If my phone interrupts me, it should either mean someone genuinely needs my attention right now or it should not be disrupting me at all. That's my notification set up.
Apps allowed to receive push notifications
Phone,
Messages,
Whatsapp,
Apple Health,
[brand] bank.
That concludes the list.
There is no reason any other app needs to be able to instantly ping me. Most apps are not notifying you because something matters; they are notifying you because they want your attention.
I do not need notifications about streaks, sales, recommendations, delivery updates etc. All that can wait until I choose to open the app. It is not urgent enough to justify interrupting me.
Yeah, this entire article is pretty transparent that it's from the sender perspective, and worried about platforms taking over "sender control".
Who is he kidding? The vast majority of apps have absolutely proven they can't be trusted to respect your attention. From my perspective, the more roadblocks the platforms put between unnecessary notifications and my phone, the better. And I don't think Apple or Google are some sort of heroes here, but I do believe their incentives better align with mine than the marketing department of some app I was forced to download because I bought a ticket once or something like that.
Notification categories are like mailing lists now. You may have unsubscribed from the daily deals email but you're still going to be auto subscribed to every new slightly modified category in perpetuity. Unless you fully disable notifications for an app (in Android at least, in my experience), new enabled by default notification categories are added all the time.
When they exist at all. Many apps that provide important notifications (like delivery tracking, drop-off time etc) put them under the same category as marketing stuff. You can't have just the transactional tracking, you have to opt-in for the marketing notifications as well.
The ridesharing apps are the most annoying about this. Yes I want to be notified when my uber driver is almost here to pick me up. No, I don't want a notification about yet another sale.
It baffles me that they do this. I have to disable push notifications from Lyft entirely, so instead they send me ride updates as text messages, which surely must cost them way more money. Why not just introduce a "ride updates only" push notification category and stop this madness?
I had to disable from Android settings all LinkedIn notifications.
I check it from time to time but I haven't missed anything, nowadays LinkedIn is mostly garbage
On iOS atleast, Live Activities are separate from Notifications. So I can still monitor food or grocery delivery even though I have turned off their notifications.
Now a few apps have started sending notifications through WhatsApp because they have my phone number. e.g. Amazon
Yeah, but I still see apps that don't implement those features. Mostly React Native/Flutter apps that don't bother implementing native features. On Android it's even more depressing.
I'm not worried about missing food notifications because they send me an email and a text (... and a fax and a hardcopy confirmation letter in the mail.)
Another sneaky behavior in Android is that categories that have yet to send a notification, which of course includes newly added auto-enabled channels, are collapsed under the 'show unused categories' button.
I can see a certain category of people screaming that WhatsApp calls are broken if that were to pass… but I do agree that no one would scream louder than app makers wanting to retain their share of human brain attention.
I want scopes like Graphene has for storage scopes. I want this on my phone and browser - let the site/app think it has everything (cookies, storage, microphone, camera, notifications, whatever it wants) but it's all empty and does nothing.
This is basically required for clueless (and even not so clueless) users.
If there's a chat app I installed 3 years ago, with no intention of giving it camera access, and I suddenly need to use that app for a video call, I don't want to be stuck debugging broken camera issues for two hours. I'd much rather have the app tell me that it doesn't have camera access.
> This is basically required for clueless (and even not so clueless) users.
I can actually confess that this hit me. Almost nothing on my phone has permission to use my camera, including my web browser (why???). I assume this was done in a fit of pique upon discovering that the setting even existed.
Roll on (god knows how many years later) and I cannot get into the gym with the link I was emailed to have my browser read a QR because my browser is just a grey screen. It was only when the member of staff suggested permissions that I realised what was going on.
The OS could tell you instead. If it is a camera app, the OS could tell you on install, that you can't start the app without given camera access, because that's what the app is.
They can, but there's an OS option that basically is "I'm going to say yes, but then effectively do no". Basically it'll pretend to the application that a permission is granted, but then just keep returning empty information or doing nothing with it. So notification perms would then be seen as enabled, but nothing is actually being send to the user.
Unfortunately Google isn't really exposing this to users, so you need something like App Ops or adb to set it up.
Of course, that way they can so they can refuse to work until you uninstall or give in to their demands. There are other operating systems that present fake data at least.
Tip: The iPhone Passwords App has basic TOTP functionality (manually create a password entry and click “Set Up Code”). I have a few dummy passwords which are effectively just labels for some login codes - it’s one less App to install.
AFAICT any TOTP app (FreeOTP+, Aegis...) works just fine with Microsoft services (or Google, etc). You don't actually need to install several TOTP apps.
Key word there being 'option'. If you choose to use push as your mechanism then enabling it is obvious. If you choose not to the app should still work. You don't need push notifications enabled on an MFA app.
Why would I want that? If it is not me, I am not going to allow the login. Making it a notification makes it more likely I could fat finger an approval.
I guess you can make the argument that you are then made aware of login attempts, but that feels more like something the host service should control.
I recently got an unsolicited OTP email from Microsoft, which led me to fear that someone had entered my password, but no: I eventually was able to confirm that the arrival of an OTP does not, in fact, require that someone enter anything beyond my email address. This is rather insane (I should not be having a blood pressure event due to Microsoft) but on the other hand I do understand the passwordless concept which is just a password-reset flow sans password-change. Perhaps a nice middle ground would be if the OTP email explicitly stated that my password was not entered.
This also happened to me about a week ago and I had the same reaction/discovery process you did. OT but I wonder if there was a recent ramp up in these attacks. It was done against an email I do not regularly use that was attached to my account as an alternate and haveibeenpwned confirmed was in a data breach back in 2020.
Some providers (looking at you, Intuit) don't seem to understand TWO factor authentication and will allow someone to bypass your password if they can intercept the SMS or email, and treat it as a normal login.
I would, but I don't need to know immediately. Plus you have the other vector of my phone sitting on a table and showing the notification to a person who can see it when they are trying to login as me.
I find it to be a poor default that sensitive data is shown on the lock screen. I change that setting as a first order of business whenever I'm setting up a new phone.
I saw a new marketing strategy recently: Someone tried to sign into something with my email. I didn't have an account, so they took the excuse to send me an email asking me to create an account.
I saw a new marketing strategy recently: Someone tried to sign into something with my email. I didn't have an account, so they took the excuse to send me an email asking me to create an account.
This has been going on since at least 2006.
Startups will "growth hack" by buying e-mail lists and feeding them into their password recovery tools.
A certain percentage of people will then follow the links and end up creating a new account on a service they had no interest in that now has their confirmed contact information, a new user, and a plausible reason to bombard them with marketing email.
I recently started getting emails from a company warning me that "I only had x days left to verify your account."
The account was supposedly registered for an organization whose name was somewhat similar to mine, so I thought somebody fat-fingered their coworker's email (the initial email was an invitation to create an account and join the org), but it might have very well been the tactic you described.
Yeah, the worst kind of software is the kind that interacts with the real world, particularly when chosen by clueless people at non-tech, real-world companies.
Conferences are great examples. You do want to submit your paper and go to a conference. To do that, they need your email address(understandable). That email ends up on dozens of email lists run by people who are doing "outreach" or something of the sort.
It's a consequence of having platforms instead of protocols.
Suppose you want delivery notifications for your packages. The seller, by contrast, wants to spam you with marketing.
If getting the notifications requires you to install their app, they're going to shovel any spam into it that they can, and then they're writing the code that runs on your device. Whereas if the software on your device is controlled by you and the notifications are received using a standard protocol, you (or someone like uBlock) can create filters to only show the notifications you actually want and discard the spam.
But for that to actually work you need the software running on the client to be under the control of the user independent of which device or service they're using, and subject to competitive pressure. Otherwise the platform uses is as a means for lock-in and then filters your notifications in the ways that benefit them rather than you, or just does a lazy job because they know you've been deprived of having a lot of other alternatives.
Unless the task is extremely well defined, protocols don't really work.
Imagine you're a shipping company and lock yourself into a parcel tracking protocol. You then decide to offer the innovative feature of parcel lockers, which need a code (or an action on your device) to open. How are you going to make the thousands of weird homebrew clients that people are using on their jailbroken Nintendo Switches or whatever to behave?
> But for that to actually work you need the software running on the client to be under the control of the user independent of which device or service ...
In other words, you need the user of the software to pay for it's development. Since that won't happen ...
I don't pay for most of my software and yet it still respects me. Of course it also isn't made by large corporations with marketing and sales departments.
> From my perspective, the more roadblocks the platforms put between unnecessary notifications and my phone, the better.
I know lots of apps behave badly when it comes to notifications but I'd still prefer if the apps controlled the level of notifications they sent. I could, of course, reduce that client-side, but I don't see why I'd want Google or Apple or any other intermediary see or control the notifications.
If an app behaves inappropriately, I could uninstall it. If a gatekeeper like Google or Apple prevent an app from sending me notifications, I'd have to change my OS, usually my hardware, too.
This forces millions of users to individually monitor and fix dozens or hundreds of apps all the time - something most don't have time for and leads to an awful experience. Centralized controls are better for the user.
TFA discusses at-length how APNs and FCM are necessary intermediaries regardless, effectively creating a technical duopoly on 'push'. We all agree it would've been preferable for things not to have gotten this way, but here we are.
> I do believe their incentives better align with mine than the marketing department of some app I was forced to download because I bought a ticket once or something like that.
Align better for now. It will get enshittified.
I try very hard to avoid installing apps specific to a particular business or organisation. So far I have only had to install a government app and some from banks. Even those are avoidable (but it would be very inconvenient to do so).
The biggest problem are apps that do both. For example, I want Uber to notify me when my driver has arrived, but I don't want it to notify me when they have a special 10% discount on my next 5 rides. It's not straightforward to block one but not the other.
If I order an Uber, I already know it is coming. I was the person who ordered it.
This is how taxis worked for decades before smartphones existed. You phoned for a taxi, then remained vaguely aware that it would arrive shortly.
The question is whether a single “it has arrived” notification is worth the surrounding noise: “driver accepted”, “driver is nearby”, “rate your driver”, “here’s 10% off your next ride”, and so on.
In most cases, it is not. The useful information is either already obvious (you can see the car outside) or you have re-opened the app to check where they are.
Operational and marketing notifications should never share the same permission. Until that is enforced at the OS level, I will treat them all as unnecessary spam.
The difference is that taxis would either give you a call that they were here or they'd just wait for. They don't care either way if you show up or not because the meter is running. The Uber is gone if you don't show up in 5 minutes. That is if you are lucky and the driver didn't mark themselves as "here" when they were 2 blocks away, which seems to be the norm here now.
Android has different types of notifications for apps and can have them filtered separately. Unfortunately, some app makers like Uber are bad about labeling. Google would need to enforce labeling so transactional and advertising notifications are separate.
iOS has the thing they call “time-sensitive notifications” which is a flag you put when submitting notification that is supposed to be Really Important Right Now. Unfortunately it’s not easy to mute everything that is not time-sensitive
The point of notifications is the convenience of not having to constantly check your phone for every single app you have (amazon delivery? just eats delivery? uber booking? claude finished its task?).
> The point of notifications is the convenience of not having to constantly check your phone for every single app you have (amazon delivery? just eats delivery? uber booking? claude finished its task?).
My phone has been on DoNoDisturb since 2010 or so. Here's the reality: I don't check for any of those things. Delivery drivers can ring the door bell. If I'm very hungry I'll keep the app open and check where they are. I literally do not care to be notified about any of the things that apps want to notify me off.
Anyone who cares to reach me knows to ring the phone twice in case of emergency to get through DnD. Anyone else? The best time to call is text me. Or schedule a time.
As for Claude, the point of clankers is that they work in the background. The robot can wait, their infinite patience is a feature.
I guess you probably have no dependents and never been oncall then if you are on no disturb. For many people, having to poll the state of multiple ongoing tasks is time consuming itself or/and focus breaking enough that some apps are deserving of having notifications.
Manually polling multiple items as you go around your day is stealing valuable mental bandwidth that could be used in better things.
Oh I’m on essentially permanent 2nd tier oncall and occasionally 1st tier. Pagerduty has an exception configured. If you message me on slack it shows up on my home screen and I will probably notice in the next 15 minutes or so.
I find polling less disruptive because my phone or watch are almost always nearby. Being in control of when to get interrupted feels better than having stuff constantly pop up while you’re doing something.
Just holding my partner’s phone spikes my cortizol. You try to type 2 sentences and get 5 notifications for random other apps and messages popping up on screen in the meantime. Literally one notification every 10 seconds on average. It’s horrible
> Manually polling multiple items as you go around your day is stealing valuable mental bandwidth that could be used in better things.
I both agree and disagree with you. I disabled notifications for everything and found myself refreshing email too often. But if I have the notifications, I can get disturbed too often which also takes mental bandwidth.
For me I found the best tactic was to selectively enable notifications (whatsapp for just one person), and delete (not silence) everything else - I don't have email on my phone now - the temptation to check it is more than the need to have it. As for things like PagerDuty, I have it send an SMS and phone me instead.
A built-in notification system that everyone from your local politician asking for your vote to your Jehovah Witness recruiter can and does abuse. Way to waste your time.
I have notifications enabled for my bank because it may alert me to transactions which were not made by me, and because it does not abuse that permission to send me marketing spam.
I have WhatsApp notifications enabled because it is the primary way people communicate where I live. If my elderly mother messages or calls me, it will most likely be through WhatsApp.
Both of those notifications contain genuinely important or time-sensitive information which may require action on my part.
That's the distinction between them. A person contacting me is fundamentally different from a brand attempting to engage me. Transaction alerts are fundamentally different from “your order is out for delivery”.
The criteria is not “did a thing happen”. It's whether the notification gives meaningful new information that is important or time-sensitive, and requires me to take action.
Snapchat has to be the all time worst offender to me about abusive level of notifications. Luckily, you can turn them off, but holy cow batman, that's a lot of notification options to deal with.
For me the worst is NextDoor. I don’t have the app installed, but they also have email notifications. There are seemingly 100 options and I turned them all off when I first made the account. Periodically they add new ones and auto-enable them for everyone. There is not universal way to shut them off, short of blocking them all together or deleting my account. The account was such a pain to setup that I’m hesitant to delete it, for the 1 time every couple years where it’s useful.
Even worse with ND e-mails are how they've absolutely perfected the cut-off character limit for what's being posted in your area. So my inbox is just perma-barraged with click-bait-y "This place on Smith Street has the best...", "Health officials are investigating an outbreak of...", etc.
Yes. I’d rather live with the temporary inconvenience of needing to open the Uber app to check the status of my ride once a month than wade through notification spam on an intermittent basis forever.
I just refuse to grant permission as my default. If I ever feel like I’m missing out, I can turn it on later. Usually I don’t and if I do I quickly regret it.
Has been like this on my phone for a while. It's crazy when you see someone who hasn't blocked everything and their phone dings multiple times a minute.
It's a failure of iOS architecture to not force applications to tag each notification with labels. App developers have to build notification management themselves for fine grained control.
Because companies are trying really hard to hide the "no" button: it's a single click to say "yes to all", but a safari through dialogues to say "no to all"
Same with websites like Youtube who don't understand a plain "no" but offer a fake choice between "yes, harvest all my data" and "ask me again later". That isn't consent, it's coercion.
Because people don't actually read what they are clicking on or even understand what they're doing. They just want to make the annoying banner go away. Same reason why people mash the next button when installing software.
Hundreds of thousands of people declaratively opt into receiving marketing with informed consent on a daily basis. Just because you don’t does not mean other people are like you.
I mean, it's kind of an insurmountable obstacle. Why bother trying to unsubscribe when you're always gonna get spam anyway? It's just gonna come back.
Also, websites are shady. If you put in a required email, they'll usually automatically check a little box for you that says "allow us to ruin your inbox?" How helpful of them.
And, I'm not even convinced that checkbox does anything.
It's definitely not insurmountable. It just takes a little bit of inbox maintenance. Pressing unsubscribe and report when I have spam in the inbox, 5 seconds here, 5 seconds there. I still get spam, but it's minuscule compared to not doing this.
I just flag every marketing email as spam. It's much more effective than unsubscribing since it tells your email provider to just redirect everything from them to spam, and it causes trouble for the sender.
Exactly. Unsubscribe is for newsletters you consciously subscribed to but no longer want. Anything unsolicited that isn't a genuine one to one communication is spam and all the "unsubscribe" option does is verify that your email is active and will be able to receive more spam.
It's not insurmountable. Spam filters have been around for decades. They're pretty good. If I didn't expect the email, I train my spam filter that it's a bad email. There are a few that get through, maybe 1-5 a week, which require flagging.
The checkboxes seem to be a placebo. Sometimes there isn't one. Sometimes it doesn't do anything. Sometimes they say "updates on your order", which apparently also means future products you might want to buy a week after you receive your order. (Marketers' English seems like a foreign language to me).
I realise that CANSPAM as a law in the US is titled appropriately but I live in the UK where declarative opt-in is a must apart from a couple “soft” opt-in scenarios.
People. Opt. In.
Stop conflating your preferences with other people’s.
Is “informed consent” that little checkbox that is checked by default? Or is it the one with the wording that says something about “discounts and offers”? Or is it the one that’s enabled because it’s a “new category” that didn’t exist when the user signed up so why not require them to opt out? Oh, I know, maybe you’re talking about the “enable notifications? Yes/Ask Me Later” dialog that is pushed on them every single time they open the app.
I’m sorry but if you honestly think the number of users who receive marketing spam have expressed “informed consent” you’re fucking high. There is a multi-trillion dollar industry devoted to tricking people into opting into spam. Stop pretending these people are expressing any consent at all.
I live in the UK where we actually have sane regulations around this stuff, I realise laws are different and folks in the US in particular don’t have much protection from spam.
Yes! iOS 27 needs to categorize notifications using AI. Apps aren't supposed to advertise to you, but some don't allow for that distinction. I want notifications on for when my sandwich is arriving, but I don't want push notifications for a promotion. Some apps are good about this, others don't allow that granularity. I hate the all or nothing.
On the flipside, I have an app that sends notifications. We don't abuse it or use it for promotions, and APNS and Google's version works perfectly fine for us.
edit: downvote all you want. Fact remains that there is no way currently to block advertisement notifications and no disincentives for those who use them.
Send everything to the iOS notification summary which you then don’t look at. Uber and others can send time sensitive notifications and those don’t go in the summary. It’s basically a junk notification folder.
periodically open the app every few minutes or so. once the driver is 5 minutes away -- go outside and wait.
it's a tradeoff. eliminating notification spam means behaving more synchronously when booking a taxi. i don't mind waiting outside for five minutes. especially if i'm not getting a random ping when i'm definitely not booking a taxi :shrugs:
Most people aren’t aware but there are laws that require granular notification consent. For example the GDPR has it. I’m currently fighting with a major bank and educating them about my rights. I want to receive security related notifications but not get spammed by “get a loan up to 50k without lifting a finger” type of bulls*. Send send this almost every week..
The user legitimately considers the application as hostile - hence sandboxing... Notification spam filtering is now the obvious need at the sandbox's edge, with the whole customizable arsenal we have come to expect for our inbound mail. Of course, Google will not cooperate with anything likely to reduce sacro-sanct engagement !
Except that they did. Android has notification channels. Now, I suppose we could argue that Google could be more ham-fisted about forcing apps to use them, but that's murky.
In fact, Uber on Android does use these notification channels. I just have "All Promotions & Recommendation notifications" disabled, and then "Taking a ride" channel enabled.
>Now, I suppose we could argue that Google could be more ham-fisted about forcing apps to use them, but that's murky.
I disagree. What the is the point of forcing everyone to use the Google Play Store (or whatever app store on iOS) if the store doesn't stop spammers?
People complain about Uber, but Lyft does the same shit. I got a promotional notification from Lyft and could not disable it without disabling the main notifications that tell me when drivers were arriving.
If app stores were useful instead of just rent-seeking, they would kick Lyft off until they stopped doing that.
The modern pattern in anywhere that allows it is to have dozens of ambiguously labeled toggles for nominally different notification channels, described only by a minimally brief and maximally ambiguous label. All begin as active until the user, in frustration, goes in and exhausts themselves disabling individual options without being sure which one is going to turn off the one single thing they actually want to be notified about.
Then next month, you create a new notification channel for your new promotional messages because too many people opted out of the old channels. You default that new channel to opt in, to make sure the user gets their chance to experience it and share in the delight you mean to share with them.
Presumably, you continue this until you have hundreds of such toggles and presumably some kind of dedicated Toggle Engineering Department that oversees them all. Nextdoor, Meta apps, LinkedIn, and countless others all appear to be competing for the most such toggles.
After all we wouldn't want the user to miss out on our promotion of 10% off your next refrigerator. They bought a refrigerator from us just 6 months ago, after all!
It also puts a throbbing highlight (for a few seconds) on whichever channel is associated with the notification whose gear icon you used as a shortcut into the channel list. At least for Pixel phones.
Tell me use iOS without telling me you do. Android has separate notification channel toggles, so I've turned off the marketing ones. I was shocked and aghast when I spent a year trying to use an iPhone that it didn't do this. Part of the reason I went back to my trusty Pixels.
I use Android. Lyft put marketing notifications in the default notification channel on my device. If the Play Store were useful, they'd have banned Lyft until they fixed it. (haven't gotten one in a bit so maybe they did (or maybe I set something so that the app could only message me while it's active))
Lots of iOS apps have the option, but ignore it and send you push ads anyway. Apple may require it to be present during app review, but they don't seem to enforce that it's used correctly.
Does Google actively police app's use of channels? Is there any mechanism to stop apps abusing "time critical" channels and sending unwanted marketing?
It absolutely makes sense (in a capitalist sense). Then you get more money/engagement/whatever on all of the other platforms.
It's the same reason Microsoft built functionality to let users in Europe have links open in their default browser instead of Edge but blocked that feature for the rest of the world.
And the worst part is that Apple could fix this in a heartbeat. Uber is straightforwardly in violation of App Store policies about "no advertising in push notifications", but a) they're too big to fail and b) Apple advertises via push notifications all the fucking time, so they have no leg to stand on here.
It's infuriating that the one thing the App Store monopoly could be useful for isn't even actually used in practice (if you're big enough, ofc, you and me get to eat shit if we try to evade App Store policy).
Instagram is the worst offender, I only want to receive message notifications, but I got notifications about inane random stuff I've tried to disable but it won't work. I ended up having to disable notifications altogether.
Instagram drove me so insane with that that I spent a while searching through the app to figure out how to disable it. There's a way to do it, and for a while it worked; I only got notifications about things like direct messages and posts from a few people I specifically told it to send notifications for, but I never got the "recommended" posts.
Then I got a replacement iPhone and reinstalled Instagram.app, and it defaulted to "show you notifications for everything we think you might be interested in" again, and I was too lazy to spend all that time relearning how to disable those notifications. I disabled the notifications entirely and now I open the app once a week or so to check in.
I had to do the same thing with Facebook years ago. Now I open it once a month to see who from high school got married in the last month and click the little "heart" icon and scroll until I get bored (~2 minutes). Can't trust Zuck with notifications.
Instagram run their notifications via an auction mechanism (which I suppose makes sense for an advertising company that likely built a lot of RTB systems).
I believe the App Store policy is you have to have a setting to disable ads. And Uber actually has it (though it has 8 different channels or so, apparently "Uber teen accounts" marketing was added recently).
I used the setting and am not getting Uber ads (only Uber ride notifications).
Currently my biggest problem isn't ads, it's all the apps now will find ANY excuse to send you a notification in order to keep their "Daily Active User" count high.
You turn off more and more categories and they'll still find a reason.
WellsFargo does that. Important notification and advert-BS on the same channel. If you block their notifications you don't get the near-real-time Zelle alert. Enabled you get what you want and also YOU MIGHT BE PRE-APPROVED!
On Android with notification categories it is, but iOS doesn't have that. Also, I think it's mostly a trust system. But Uber in particular does actually do it right, and you can just turn off promotional notifications.
"Marketing never met a communication system they didn't want to co-opt"
(I'm reminded of this every time a client want "WhatsApp support" in their (commercial) app, so they can "communicate with customers".)
But equally every user will have a different subset of apps they want notifications for.
For example shift workers need to know when they've been allocated a shift. Or when a shift has opened up (because someone scheduled failed to arrive etc.) One group of users consider this really important, another group of users treat it as spam.
But, per the rule above, unfortunately "useful notifications" can easily be subverted by marketing notifications. Yes I want to know my delivery driver is outside, no I don't want to know that you're running a special this week.
Unfortunately there's no way to solve this problem technically. Bad actors can (and definitely do) behave badly. But ultimately the system should work for "good citizens". In other words, the user should ultimately determine what they want to see of not. And if an app has "notifications on or off" as the only option then the user should ultimately determine that setting. Not Google. Not Apple.
Building society around the lowest-common-denominator just ends up sucking for everyone. We should actively promote good behavior, while allowing bad behavior to be punished. Not just banning everything "because it might be bad".
I totally agree. Right now the apps that can notify me are phone, text, email, what's app, and a few bank apps. You are 100% right about turning it off on everything else.
I also stopped doing store loyalty cards about 7 years ago and it's been fantastic. I actually get a lot less junk mail and spam/"legit" marketing emails. I don't have a gob of cards to sort through.
You're conflating "push notifications" with "being alerted about push notifications." I have many "important but not urgent" apps on my phone configured to just silently add their push notifications into iOS's notification center.
With an app configured to do notifications like this, no banner shows up at the time the app's notifications are delivered; and these notifications don't even show up visibly on the lock screen. You only see this type of notification if you choose to actively scroll down past the "timely" notifications that do get delivered onto your lock screen, to "catch up" on all your notifications.
Basically, these notifications are relegated to an "email inbox" that you can check or not check as you like. But unlike your email inbox, you can go "inbox zero" on your notification "inbox" whenever you like without worry, since notifications (unlike email) are inherently prohibited from being a critical path in an app workflow.
Just curious, what do you do with the increasing number of companies that use push notifications as a form of advertising venue, and how do you differentiate the security warning notification from your camera app from their special weekly annual sale notification?
The marginal cost of each notification is so low that companies simply spam users nonstop, and their A/B tests shows that the revenue is increasing. What's being lost though is that we're getting more and more agitated with these brands and their uncapped malicious behaviors. This is also true with their UI and UX as well, they keep adding banners with incredibly small close buttons, because someone will continue with the shopping after accidently missing the tiny button, and that's an added sale, who cares about 99% of users who are fuming with dissatisfaction, what are they going to throw away their $200-300 smart home device because companies abuse them?
On my Android, I aggressively mute or uninstall any apps that send push advertisements. I haven't seen one in more than 6 months now. If they think they need to advertise to me, I think I need to do without them.
Apple and Google failed to make push notifications usable for the past decade. Most important notifications drown in a sea of absolutely irrelevant nonsense. It's a very primitive mechanism where many apps compete for very little screen real estate. Beyond "something happened!" there isn't a whole lot of information in most push notifications. They are mostly not very actionable and very vague about what actually happened. And "something happened!" just isn't very useful information to me. This has de-valued the whole notion of having notifications. Whenever something interesting actually does flash by, I often miss it or can't find it back.
The push notification UX is just beyond terrible and it just got worse over time as app developers tried abusing their super power of being able to interrupt the user at will and Apple and Google tried to get on top of that. The net result is something that's very mediocre for the handful of valid uses I have left for notifications. My list is similar to yours. Things like bank approvals, 2FA stuff, etc. are useful mainly as deeplinks into apps. But other than that, it's just not worth dropping whatever I'm doing and staring at my phone.
The most used apps on my Android phone (older Google pixel model) are Firefox and gmail and just a handful of other things. As a notification channel, my email inbox is actually far more useful than mobile push notifications. They are more actionable and informative. And I can individually unsubscribe them or filter them out and easily find them back. Most apps can do both and that makes the push notifications inferior and redundant.
> The most used apps on my Android phone (older Google pixel model) are Firefox and gmail and just a handful of other things. As a notification channel, my email inbox is actually far more useful than mobile push notifications. They are more actionable and informative. And I can individually unsubscribe them or filter them out and easily find them back.
There’s also substantially more filtering happening in the inbox which is mostly useful from a user perspective.
Yahoo literally wrote a paper more than a decade ago showing how they can model predictive causal chains for emails they expect you to receive, as an example.
> Phone, Messages, Whatsapp, Apple Health, [brand] bank.
Anyone else annoyed by the fact that you can set up do-no-disturb, with exceptions for certain phone numbers, but it doesn't work for apps like WhatsApp?
I remember a while back I also had this issue on iOS (maybe around 2022 or so), though lately seems to be solved and works as one can hope so. When you're making the exceptions, do you explicitly input phone numbers or some other method? I selected contacts from my contacts list and it does the job. This is for iOS in my case. I'm not familiar with Android, so cannot give any input there..
Exactly. Senders have earned the questionable reputation that they have because they rabidly want your attention whether you want to give it or not.
I used the Southwest Airlines app recently and allowed notifications so that I could find out about things like delays and gate changes (both of which happened on my trip). Less than a week later I'm getting ads for travel "deals" pushed as notifications.
Unsurprisingly, it was difficult to find the notification setting, which was on their website, not even in the app.
Maybe it's for the best. The best practice is to have as few apps as possible. The moment an app is abusive with notifications, you know it's time to drop the app anyhow. A lot of people need that one final push to drop the app, so this could help.
And let's not forget focus modes... I have them that narrow greatly my default set of notifications, so I have a 3 tiers of notifications.
It's like the complaint I used to hear all the time: "Slack ruins work for me! OMG I can't work with constant interruptions!!" That is bewildering, because if that's how you feel, you haven't tuned your setup. Slack never interrupts me, yet I am response enough to slack messages. No one has ever complained about my response time. And I'm probably the most-messaged person on our Slack.
The other problem with Slack is that it just straight up... doesn't do what you tell it to. I have a set of notification settings that work for me. Slack goes ahead and just does something else, and you simply can't fix it to do what it's told. (Or couldn't, anyway; I've been off Slack for a while.)
The idea that software like Slack could be setup as "one size fits all" is just ludicrous to me. We have options because different people require different settings.
For Slack, I find just changing the default notification sound to a simple and subtle ding works well.
When I’m focused, I don’t hear it because it’s too subtle. But when I’m not concentrating on anything, it’s more noticeable and I don’t mind the distraction.
This might not work for everyone (“YMMV” and all), but I’ve personally found it a very effective yet simple solution.
I have no audible sounds from notifications.
They don't go to my phone, with few exceptions.
I get no popups.
Yet, I am responsive.
It was trivial to set up.
I would say the same applies to background processing as well. A random app that I don’t interact with launching every minute and wasting everything from battery to network bandwidth is simply not acceptable, and most of the time they’re loading adds or doing some other stuff that serves me no good.
I wish I could set this as the user. Apple ties background app refresh to the frequency of use, but that sucks for self-hosted photo backups. I use Immich and I don't open it too often, so Apple breaks my chosen backup system for my photos.
My notification setup is more elaborate (for one, I do keep social media notifications on, but silent) but yeah I agree in general. It frightens me seeing some other people's notification shades where they have 20+ spam notifications from all kinds of things that I wouldn't even consider installing an app for, and they're somehow fine with it.
Same: Phone, Messages, Calendar, Apple Health... nothing else can send me notifications.
On my work I also disabled all notifications except for the calendar. Even slack message our main tool for communication is not allowed to send notifications. It is almost a productivity hack :P
That said, my view is now (not novel, or unique) that I am not the customer in so many cases. Any app or platform with the slightest hint of an advertising end-game restructures my usage as the product.
The customer is instead the sender (or advertiser). So, I can't expect ideal app behavior and usage based on my intentions because I'm sold (as the product) rather than the other way around.
Maybe a cynical view, and there are exceptions, but don't think I'm far off.
I have broadly the same list as you do, but stuff like WhatsApp, Messenger, and other "non primary" communications platforms have silent notifications in the sense that they're not allowed on the lock screen or Home Screen. They simply display a notification counter.
Stuff I care about that I can't do anything about "right now" are allowed on the lock screen but quietly. That includes messages from the kids schools. Most is not even that important, like field trips "next week", but once in a while there's an "important" message I need to deal with.
I don't get what you guys are doing to be so bothered by notifications. I get them on my wirst and even then it isn't enough to take away mental bandwidth.
My bank likes to show offers, like a 10% discount in tires, but I have no car. Perhaps tree or four irrelevant messages per day.
I have MouseTimer that is an alarm that is nice to show to kids when they must wait or do something for 10 or 20 minutes. It should be able to ring and sometimes show notifications.
At this point, I'm pretty much in some form of DND at all times. I have a very small list of people that I allow the device to notify me at any time for calls/messages. Everyone else gets silenced and I'll get back to them when I choose. All other apps have notifications disabled and I'm constantly nagged about it when using those apps
I went even further and my small set of the most important applications runs in the background - rest doesn't have that privilege. I've treated my spare Samsung phone same way.
I also don't use Siri either beyond setting timers and lights in home and every application is also excluded from being "suggested". Apple for 14 years didn't bother to add support for Polish so it basically remains useless.
I'm personally just at messages. And even then I make it clear I respond when I want to. Only phone rings/notifications I get are for those in my contact list.
Take your phones back. Life is immensely better these days.
Your position is that of any normal human. Google is committed to evil however, just look at how playstore notifications are tied to sales spam. Want payment notifivations? Gotta take the ads as well, not seperate toggles, one toggle. Drink liquid shit you tech peasant. Oh? this hostility drove you to f-droid? We'll unilaterally decide every device r belong to us, so we can disable competition we dont approve of. Welcome back to the liquid shit trough, peasant.
I have it turned off for my bank. For some reason Bank of America doesn't allow me to sign in with Face ID. I always need to get a text. Only reason I keep them is because I like a brick and mortar bank nearby.
It's absolutely disgusting how most tech companies use notifications as an advertising or addiction building channel.
On the rare times I use an app like uber eats, I uninstall it directly after because the app sends multiple adverts a day through the notifications. I want a notification purely to tell me the driver is almost here. And nothing else.
I've noticed a priority inversion in recent iOS. Want to send me an SMS that matches a ban-list regex from a third party app, from a foreign phone number / obvious spam farm? No problem. The app to block you was auto-uninstalled, and the iOS notification filter will mark your message with the highest possible priority.
Want to continue a 300 message thread that I've been responding to? You're listed as my emergency contact, and called multiple times? Fuck right off. Straight to spam.
It's almost enough to get me to carry a second dumb phone or grapheneos device just so I can text and receive phone calls.
For me I definitely need Calendar and sometimes Clock (alarm). iOS is constantly freaking me out by prompting me whether or not I want to continue receiving notifications from those apps. To me those apps exist entirely for the purpose of generating notifications and it terrifies me that by repeatedly popping stupid questions like that, I'm going to accidentally answer wrong and effectively delete my most important app accidentally. It boggles my mind that somewhere someone thought Clock and Peggle were basically on equal footing here.
Pun aside, it makes sense for a notification framework including a notification delivery network to be built into these mobile OSes, because the alternative (letting applications run arbitrary background services) is usually worse.
Agreed. I have a doc appointment app, where I like the notifications to be on for reminders etc.
Lately they started sending marketing messages through that channel. Now I’m sure it’s possible to turn off the marketing messages. But I bet most people don’t know and won’t change that. It’s super annoying.
> I’m sure it’s possible to turn off the marketing messages.
Uber may have that functionality, but a surprising number of other apps don't - for example Makro, Tops, and 7/11 Thailand, three very popular Thailand retailers, use notifications for when an order is out for delivery, about to arrive, etc. But they also send constant promotion notifications every day, even with audio alerts enabled.
We must, at some point surely, reach an inflection where even everyday people are sick of this shit and start smashing their phones right?
There has always been "unpluggers" [0] amongst technologists, but the vibes are bad and getting worse. I feel like that is getting more common between "normal" people I know, but maybe outside of my country town bubble its not happening.
I was thinking we're only one or two big influencers away from a cascade, but then the ultra-influencers are never really going to commit because its their livelyhood and saying throw your phone away is self-limiting on the viral aspect.
I absolutely hate medical marketing. I recently decided to switch dermatology practices over a combination of bad data management, unwelcome marketing for cosmetic dermatology products, and unsolicited SMSs. I never consented to receive marketing or texts (or to lose my data in a data breach, or to be billed for the in person services by practices other than the one from which I received services and ignored unopened as spam because I didn’t recognize the sender, etc…)
I wish Apple would force app developers to implement different "channels" for promotional notifications vs transactional - so that you can pick and choose which ones you want.
The "you" in the title's reference to "your push notifications" is not the user, it is the marketer. That tells you everything you need to know about the value of this piece.
Something might come of introspecting why such controls are being built and desired by consumers instead of trying to frame everything as a "big tech evil!!1" narrative.
I’d recommend following your own train of thought, why is big tech so hell bent on intermediating the experience between their users and everyone else? They’ve done it for email, web search, mobile experiences, advertising, etc.
You want to use any of those things, you’ll have to pay their toll booth, figuratively or literally.
The chain of thought is quite straightforward. Functionally nobody wants an intermediary-free channel because there are adversarial entities on the other end.
I'm not a fan of Apple or Google, and it feels bad that all of our notifications pass through APNS or FCM. Megacorps shouldn't have control over our digital lives to the extent that they do, and anyone talking about this gets my full attention and support.
Except for marketers! I don't think there's a less sympathetic category of technologists, save for maybe CSAM peddlers.
You're upset that you can't get "visibility" into whether the bullshit ad you tried to ram down my throat landed on target? You're worried that I'm a dormant user and my phone will silently delete the spam you sent to try to hook me back into engaging with whatever worthless product you're hawking?
World's tiniest violin, buddy. Boo fucking hoo. Your last paragraph says the "senders" (read: spammers) who make it through the next decade intact will be, to lightly paraphrase, the ones who send messages the recipients actually wanted. You say that like it's a bad thing!
The computer is in my life because it is a tool that does the things I want. It is not an open mic for marketing sleazebags to try to sell me shit. May every single one of your attempts to invade my life and hijack my attention be flushed swiftly down the toilet.
If we had visibility, we would know what doesn't work and we would stop sending it. Almost like there are aligned interests there rather than a purely adversarial relationship.
Personally, all I've seen is marketers not getting the message and using it to perform A/B testing to come up with the basest ad they could possibly come up with to entrap more users.
Yes and that would be completely correct in many cases.
> Almost like there are aligned interests there rather than a purely adversarial relationship.
You might very well be the exception, but for something like 99% of marketing content that reaches me, our interests aren't aligned. First of all, they want to generate "needs" where there weren't any before and probably shouldn't be. A pizza ad produces the wish for unhealthy food. A fashion ad produces the wish for new clothes (even though I have enough) and probably even changes the societal dynamics of individual expression and personal style to be consumption oriented.
Second, even if I have a legitimate need for a solution, they still want me to buy their product, consume their media, give my attention to them. I, on the other hand, want to be informed by a neutral third party about the pros and cons of some product. Sure you can say "but unhappy customers are bad for us", but there are actually very few niches, where this signal is powerful enough to align incentives, because information and power asymmetry limit the customers understanding of product quality and their leverage to correct harmful market dynamics.
I can’t think of a single app I want a “Discover” tab on anymore. The moment you include one is the moment there is someone trying to game it. I definitely don’t want push notifications trying to show me something new. I’m hardly lacking in distractions
(Yes I am sure somebody can give me an example of a good use of Discover but you get my point)
is it unironically incomprehensible to you that the owner of the device should in the one who gets to decide what is and isn't spam? it's not email where you can get bombarded with shit from any random server - you can mute or uninstall an app.
You know, I would love a feature that lets me mark push notifications as spam, and optionally send them to Apple. The last part is important for a variety of reasons, one of which is that notifications can be end to end encrypted.
Spam filter push notifications.
Ideally enough spam reports for Uber Eat’s constant marketing abuse and they lose APNs access for the Bundle ID associated with the spam reports. For example.
You’re right of course, but Apple won’t do it - they’re happily running a two-tier system where Uber, eBay, Doordash can force spam notifications on you with impunity. All my settings for marketing are off - eBay still sends me notifications about coupons (and additionally there’s no way to actually contact them to complain, of course). Doordash won’t let me get delivery notifications without marketing notifications.
Apple could fully enforce their policies and fix this in a heartbeat, but they won’t.
Fine, but that’s was clearly not enough to stop the spam, nor it was enough to satisfy everyone.
There are some apps I can’t afford to mute or uninstall, such as phone, transportation, communication and work. I wish I could, but I currently can’t, I’m not privileged enough.
“Punishment by Apple” in this instance is somehow the only response anyone had to misbehaving companies.
That's not entirely fair to the author as the article also states that the consumer being in control is a must:
> Every step subtracts a degree of sender control. Some of it passes to the user, and that is a good thing: a person deciding what is allowed to interrupt them is the channel working as it should. The rest passes to the platform, and that is the part that should concern a sender, because the platform's judgment is opaque, unappealable, and increasingly made by a model rather than by a setting the user chose.
A platform has essentially two "clients" - the user and the developer. Without both, it wouldn't exist. And it is in the interest of both that the platform should have very limited arbitrary authority over them. Nobody can deny today that the platform owners today have too much power over their users and their developers, which makes it easy for them to commercially exploit them while undermining their rights. We need regulations and standards (for interoperability, which is one of the arguments being made) to counter this.
Been an app developer since 2009, worked on Android for 6 years at Google. Push notifications suck, users hate them.
Simultaneously, I cannot match the pull quote, an argument summary, to their argumentation. IIUC if the reword patent / Apple’s summarizing disappear there’s 0 reason to say it wasn’t control passing purely to the consumer.
So I’m left a bit empty as the high-minded purpose has little backing, and thus comes across as bloviating.
I’m constantly amazed how passive people are with things that steal their attention
My phone is in do not disturb mode 24/7. If your app notifies me about something pointless, it gets deleted and I start using your website instead
I have a mail rule that moves any email with the word “unsubscribe” out of the inbox into its own tagged area. Every few days, I go in and unsubscribe to everything that’s arrived.
Whenever a retail point of sale worker asks for my details or phone number or asks me to sign up to their club, I ask if there’s a discount. Because if there’s no discount - they get no details. It’s a simple exchange; offer to pay a fair price for my details and I’ll consider it. But so far my time and details are worth more than any retailer has offered to pay.
If you’re in gmail there’s literally already an interface to easily unsubscribe from stuff, it’s under “Manage subscriptions”. Yahoo similarly have a “subscriptions hub”
With the retailer asking for a phone number, I don’t see how it would ever be worth even entertaining. They will give you a discount once, then have and potentially abuse your information for years to come.
That's precisely why I don't ever accept the bribe. If I don't like the non-discounted price, then I don't buy it. Now they neither get the data nor the sale.
What's frustrating is that a lot of grocery stores do this. If you sell something absolutely necessary, such as basic foods, you should not be allowed to do the whole "mark it up to mark it down" strategy.
Also, a tip for most grocery stores (at least in the US): enter in any area code plus 867-5309. Chances are high that somebody has registered it. It's better than sharing with a family member because so many people are using it, the data becomes less useful.
Alternatively, ask the clerk to "use the store card". Usually, they will oblige.
You can make an email rule to filter those messages to trash.
I have my phone set to only ring for people in my address book. It’s probably time to do something similar for email. Not in my address book? Straight to trash.
Adding a filter is still extra work for me. Saying “no” stops the need for the rule and is less work than giving the info in the first place.
My phone is setup similarly. I did it manually back in the day, then sent some feedback to Apple, which they added in the next update about a year later. I’ve submitted a lot of feature requests, this was the only one they actually did, which is a great one. It made things much easier. Though they seemed to have changed the settings of how this works with the call screening feature. I used to have a shortcut to toggle this off and on, when I was expecting a call from an unknown number. I need to revisit my setup here, as the shortcut doesn’t actually do anything anymore.
Doing this to email is an interesting idea. If sitting in one ecosystem, maybe it would work. I’m fractured, so it’s a non-starter. Even beyond that, I think it would be an issue as there are real emails I do want to get which I’d never add to my address book, as I’d never send an email to the address. I think I’d want a whitelist for these addresses, that imported the emails from my address book as a base.
At work I had a rule like this for many year. Anything internal would pass, plus a few external domains I named. Everything else would go to a spam folder for vendor emails. Keeping up on this was hard. I ended up throwing in the towel a couple years ago after running the rule for 5-10 years. This blog post is what made me move away from this rule[0].
I'm sorry to burst your bubble but I don't think your feature request had anything to do with when apple implemented call allow listing. It's been a thing forever and they were surly aware of it since the iPhone first came out
I get your point and see it as valid, yet to nitpick most people don't feel they have a choice.
Not answering the phone or replying to people's messages is a no-no to many, which sets them in an arms race against spammers and social apps trying to get them from all fronts. And they get frustrated by us living in no-disturb land 24/7.
I don't know it could be solved, but I feel for them.
Turn off all notifications except messaging and see how your day goes. It's not going to kill you. You quickly get used to regularly checking things you actually care about, and the rest has to wait until YOU care.
I've been doing this for many years, and none of my friends or colleagues are aware of it, and they don't need to be. Notifications don't help you respond quickly, they just grab your attention from things YOU wanted to do.
I haven't checked Discord today yet. I haven't checked my email. Whenever I do want to know if my friends wrote me, or if I have some new bills, or if I need to follow up on something, I will open the respective app and deal with it.
I can put my phone next to me for hours and not get distracted.
> You quickly get used to regularly checking things you actually care about, and the rest has to wait until YOU care.
This was the biggest thing for me. Before, I was paranoid that if I turned off notifications, I'd miss something important. As though I didn't miss notifications anyway.
Getting used to regularly checking (important) things also has two wonderful side effects (at least it did for me):
* My "mental notification system" got better. Because I was less dependent on my phone doing it for me, I developed the skill more on my own.
* The apps and services that I checked less and less frequently became more obvious in how unimportant they were to me altogether. I have far fewer apps and accounts now, making me MORE punctual overall.
I'm building a productivity app right now that lives entirely on the web right now because I can't justify native development costs yet. The daily nudge (i.e., you have 47 free minutes, and this task has been sitting for 14 days) is the core retention metric. This only works as a push notification; emails get ignored. The web app sends an email. V2 goes native specifically for this reason.
The article's point about leading with the fact rather than the brand point is useful. The notification payload I'm designing already leads with the concrete numbers.
> Over fifteen years the channel has been rebuilt around one assumption: the receiver's attention is a scarce resource the platform is obliged to defend. … As a sender you are on the wrong side of that assumption, whichever way the control moved.
Fascinating how the author openly frames the situation as the sender and receiver’s interests being opposed.
Whatever has replaced the Bulk Collection of Telephony Metadata Under Section 215 of the USA PATRIOT Act informs the architecture of the Apple Push Notification, Firebase Cloud Messaging, etc. Apple owns the persistent connection to every iPhone, and only APNs can wake your app. So "self-hosting" here means running your own provider (the backend that decides what to send and hands it to APNs) instead of paying a third party like Firebase Cloud Messaging, OneSignal, or Pusher to do that for you. The last mile is never yours however. Any architecture that routes everyone's traffic through a small number of identity-aware intermediaries is, by construction, a bulk-metadata collection system waiting for a legal instrument.
[2] In December 2023 Senator Ron Wyden disclosed that the U.S. government and foreign governments had secretly compelled Google and Apple to turn over information from push notifications, including communications metadata and sometimes content. A detail that should bother any developer: app developers have no way to stop the practice if they want to send notifications on the platforms iPhones and Android rely on. Apple had been gagged from disclosing it until the program became public, after which it said it was updating its transparency reporting to detail these kinds of requests. So the architectural hypothesis isn't speculative — it's the confirmed mechanism, differing from Section 215 mainly in domain (apps vs. calls) and legal vehicle (ordinary subpoenas, FISA orders, and NSLs rather than the specific business-records theory of §215)
[1] "Its just metadata". Thanks Obama! (joking of course, no single individual is responsible for these things, it is our collective political will and its the best we can do unfortunately)
> For most of the channel's history they did very little of it visibly. The architecture was permissive of intervention; they simply chose not to intervene much. That restraint is what ended.
I guess it wasn't always visible, but they were intervening in some for or another since the beginning. At WhatsApp, push delay/suppression/coalescing was something we were always monitoring, and IIRC, it was part of the system since at least when I joined in 2011. If you don't work within the system, your users' messages don't get delivered timely.
Huh that’s interesting, do you have any further context on that? I’ve not worked on a product with anywhere near that scale before so monitoring has always been whatever I can get from commercial push platforms
I mean... record the time we first send a push message, when a client connects have it tell you if it's because it got a push or user interactive start, check the time between push and connection, add that to your choice of time series graphing tool. Graph by platform, and you can see when the platforms are delaying pushes.
Some of the delay will be ordinary things like their push service fell over or is unreliable (you also get some feedback when they don't accept push messages), or their push connection runs into silent NAT timeouts on some networks. But some of it will be things like you ran into an undocumented push quota, so Blackberry users don't get timely pushes at peak, etc. On client platforms where you have reliable background execution with network connectivity, you can potentially signal connecting clients if platform push isn't working well and have them switch to persistent connections until the push service comes back. But that was never an option for iOS; it hasn't been a reasonable option for Android since at least Android 6 when Doze was introduced... and app killers before then made it hard before then; and all the other platforms are dead. Now, push really just has to work.
AFAIK, Apple has always been willing to deprioritize pushes when you send "too many", especially when there's no user interaction; or when they added silent (voip) pushes to wake up the app, they only let you have a few silent pushes if you don't post a user visible push.
For ordinary async messaging, push latency doesn't become a big deal until it hits double digit seconds. For voice/video calls, you really want pushes to be as near to real time as possible, or the caller is gone before the callee phone rings.
Hmm, there's just a few big messaging apps and just a few os vendors, weird that you couldn't have established special treatment for pushes about user to user messages
> None of this bites evenly. The editing falls hardest on broadcast and promotional push; the notifications people actually want tend to pass through untouched or amplified.
While I have slight worries about what it means for users if Apple and Google notification services go down/censoring, I do appreciate the features that they provide to me as an end user.
So many apps use annoying and questionable marketing notifications that I'd say I have about 70% of app notifications disabled globally (because the app itself does not allow disabling notifications / has no granular control).
Am I understanding correctly that iOS notifications have to go over apns unless on the same local network as the HA server? I do appreciate that android makes this possible for ha and signal (and others) in all cases, it should be up to the user to choose centralizing the connection vs. slightly worse battery life.
I wish apple/google would implement better notification control - like the ability to turn off all marketing notifications, and a much better digest format
Notification Channels is the official way to do this on Android, but it's up to the app developer to categorize them properly. They have no incentive to allow you to turn off ads.
Actually, they do have an incentive to let you turn off ads. If they don't, many users will turn off notifications entirely. At least if they categorize them, some users will just turn off the bothersome ones.
If you're on Android, I'll always recommend Buzzkill to add very granular rules for notification filtering. I set up all kinds of filters just for the Amazon app.
On iOS I assume you're sol, that notification system is unhinged to my eyes.
You are less charitable than me. Maybe I'll adopt your approach. I first give an app the benefit of the doubt and go into the apps notification preferences and see if I can fine-tune their notifications. If not, off for all at the OS-level. If yes, I tweak it, but if I get surprised by one later, off for all at the OS-level or uninstall. It's especially annoying because I don't have notifications shown on my home screen and need to unlock with a pin so if I go through the trouble of unlocking my phone to spam and I extra annoyed with the app.
That would be nice. I wouldn’t be surprised, as on-phone models get more capable, if we don’t see them start to build an “inbox” like we see with email where you can then start seeing much more heavy processing happening.
I think that's what the Notification Organizer on Android (maybe Pixel exclusive, not sure) does. It's sorting notifications into broad categories using AI and groups them in the notification shade.
Personally, I don't see a few permanent connections as a problem. My GrapheneOS phone is degoogled, and therefore apps such as Signal fall back to a WebSocket connection. Battery life is probably somewhat impacted, but I use too few apps to notice. And in any case, this is much better than allowing Google to stick its nose into my business.
Yeah I'm disappointed this isn't pointed out in the opening paragraph. It's fair to critique Google for convincing devs that fcm is the only option, and obviously iOS is designed for Apple to do whatever they want, regardless of the owner's wishes, but Android does have other, viable, options. iOS and Android aren't equally bad here.
If my phone buzzes and I look at it and the reason was dumb, I delete the offending app and leave a 1 star review. I don’t know which of these steps are loadbearing, but my phone has gotten much quieter.
>an iPhone could not afford to let every installed application maintain its own background poll against a remote server. The proposal...a single persistent TLS connection from each device to Apple, over which any registered third party could deliver alerts.
I thought apps were sending notifications locally in the device. Like, if a messaging app receives a message, there's a network call for that. Then if the messaging app wants to tell the user they received a message, it can just hit a local API for that, right?
Is the pattern actually that the app makes another network call to the notification service to register the notification, which makes another network call to the device to deliver it?
yeah this is what author hints at with "Push as a battery problem". Apps are limited by default in what they can do in the background due to this, so most apps are in a suspended state not making network calls when you are not using them. To avoid the app having to keep running this stuff is delegated to OS which tells the app, "hey I have a push for you wake up and handle it!" You can send pushes locally but because of the background limitation it is not practical for unpredictable events like messages coming in.
Not sure i understand. Sure it doesn’t require it but Apple doesn’t trust you to handle it because you’ll probably drain the battery or spam the user or whatever, that is Apple MO, putting themselves in the middle so they can control the experience.
The single persistent connection is just to receive pushes, there is still some daemon controlled by apple in charged of dispatching to correct app.
I see the point. But honestly I am more concerned about having to constantly fight to turn off all permission allowances every time I install an app.
And the moment I have some faith and trust an app that I deem important, I get promotional junk as a "notification".
I would really like to have notifications allowed on certain apps like parking, or health etc., but all they seem to do is abuse the trust they are given, meaning I turn them off.
So where I agree with this author is certainly that more power belongs at the user.
I can't believe this still isn't a thing outside of GrapheneOS. Being able to revoke network permissions is a fundamental security and privacy tool that's willfully left out of both Android and iOS.
There's zero reason not to include it as a toggle.
On iOS it wouldn’t even be that hard. There’s already a toggle to disable use of cellular connectivity. Add a separate one for non-cellular (iPadOS can connect via Ethernet), and/or a “disallow all” toggle.
We are partly there in spirit with App Transparency keeping track of the IPs and hostnames apps connect to.
> certain apps like parking, or health etc., but all they seem to do is abuse the trust they are given, meaning I turn them off.
I've found that live activities on iOS helps with this quite a bit. Let's me keep notifications disabled on parking apps and DoorDash while still getting the tracking info I want in the live activity & dynamic island.
Otherwise, yeah, you just can't trust anyone to be respectful with notifications. Phone & a messages whitelist via focus modes are the only notifications I allow on my phone.
IMO they should be doing way more to control push notifications, there's so much more control they could give the user, and many clear violations of their policies.
One of the best apps I've bought for android is buzz kill which lets you set rules around notifications. I have cool downs on family chats and social media so it doesn't keep buzzing when things kick off, filter Amazon alerts to only "we're two stops away" and "We've delivered" messages and dismiss the rest.
I have custom buzz patterns and sounds for urgent alerts and rules that batch notifications depending what WiFi I'm on, time outs on things that don't matter after a few hours etc.
My notifications list is now way smaller and far more relevant.
Also quickest way to sort out notifications is to take your phone off silent. Hearing everything coming in, you see more when it you can then decide if the notification should make noise, or exist at all on a per app basis.
The first thing I do on every new phone is to turn off 99% of notifications. Messaging ones and email are first to go. I cannot stand the constant beep-beep.
“ None of this bites evenly. The editing falls hardest on broadcast and promotional push; the notifications people actually want tend to pass through untouched or amplified”
So … mission accomplished then? This is pretty much how I would like it to operate.
That's my thinking too. Do marketers not also have phones, or do they somehow not see how annoying promotions and spam are? There's already so much paid advertising everywhere, and the free advertising with direct notification to the target is obviously going to be abused by these exact types of people. Guardrails must be put up by the transport layer (Google, Apple) so their own image isn't lowered to the image of the abusers that are too keen on promoting their disruptions. Few people want a convenient device that makes life inconvenient[1]
[1] I know a few people whose phones constantly beep and flash numerous times a minute, and when on, the top is completely unusable because any notification dismissed is immediately replaced, obscuring those upper buttons again. I don't understand how they tolerate it.
Isn't this a strategic going for years now, throwing random notifications to make the user use more the app?
I for once, block notifications from apps I don't want, because I don't like getting bombed with stupid ads etc.
Smart phones should absolutely allow you to lie to apps about what permissions they have, to feed apps fake data, and to basically control every single thing the app sees about the phone and the user.
The fact that they do not do this reveals that phone makers are sort of market makers between app makers and customers, creating an environment which, in a certain sense, is a neutral ground between these two types of "users."
I don't think I've got a push notification in awhile. Few months ago I switched to Lineageos and started using the web browser instead of apps. It's peaceful.
I still get notifications (SMS, email, calendar, etc) but nothing pushed
The real solution is to allow users to own their push solution, and for it to become more commonplace among apps to support alternative push providers such as Unified Push. Molly, the FOSS Android Signal client supports this configuration.
My Android phone has a long list of toggles under notification for each app. I am genuinely interested to make good use of that to optimise what notifications I recieve, but I am clueless about what each of them means . For example: badges, floating notifications, permanent notifications, unified support, alerts, etc.
Probably depends on the user. Along with push notifications for almost every app on every one of my devices, I disable the summarization.
For me the notification is the point, and the point of notifications to me is that they deserve my attention. Of the vanishingly few apps I install these days, almost nothing can say it deserves my attention. Even my bank doesn’t get those privileges.
The interaction is: marketers keep trying to abuse these systems and platform owners and users keep having to find ways to fight them off. Some of these efforts have unfortunate downstream consequences (like bad summaries).
There are apps that need 100% notification delivery reliability (in a sense that OS or delivery server itself will not be allowed to decide to just drop PN for policy reasons). I guess Google can only "solve" this (by simply not passing them through any "classification") for their own apps only?
For some android phone brands delivery reliability is like 40-50%. Some brands are better (reaching 80%, still bad) some very bad (usually the chinese brands, for some reason).
And the user has no say in this. They can't say: deliver every notification without classification. Or "allow this app to wakeup whenever it wants". Everything is babysat by the great overseer, even if you write the app yourself for yourself.
Have fun writing a telephony app that receives only 50% of calls. :D
Massively overlong article that really could have done with an editor. Although obviously editors cost money, and I'm reading it for free, so I can scarcely complain. Nevertheless, some concision would have been appreciated.
I'm very unclear to me what the thesis of the article actually is. Yes, push notifications run through the vendor's servers. Yes, Apple fucked up hard by modifying the text within them - and I contend that such modification is impossible to perform automatically without unreliability becoming the norm.
The author also appears to believe that "broadcast copy" - otherwise known as Spam by those who like to write slightly more honestly - is a legitimate use of push notifications. It is manifestly not, and any app that tries will at the very least be immediately silenced. I wish I could find the tweet that put this sentiment more entertainingly than I ever could.
If App developers continue to abuse the push notification system in this way, Apple and Google will be forced to take steps to solve what becomes an end-user's problem. Yet another tragedy of the commons.
It’s a through line from an article I posted last week about the similar situation in email, which has a lot more depth as inbox providers have substantially more published papers and patents showing their intermediation.
The next post will be highlighting the difference between the actual state of the art techniques being deployed by large tech co’s (LinkedIn and Pinterest, for example) vs what’s available via commercial marketing providers and how most marketers don’t even make the most of the tools they pay for.
> The author also appears to believe that "broadcast copy" - otherwise known as Spam by those who like to write slightly more honestly - is a legitimate use of push notifications. It is manifestly not, and any app that tries will at the very least be immediately silenced.
Cool man, but it might surprise you to find out that people knowingly opt into receiving this stuff and actually consent to it.
> Cool man, but it might surprise you to find out that people knowingly opt into receiving this stuff and actually consent to it.
Consent is more than pressing 'Allow' on a notification pop-up. It's often not even informed consent, as those pop-ups are usually a part of some onboarding flow where users are just trying to get to the value the app promises and pressing 'ok' to everything.
Even if people do indeed want notifications at the time of the ask, one doesn't really know if the message provider will wind up spamming, that's a matter of trust. And once opted-in, even if the users no longer want notifications, a lot just don't know how to turn them off. People are often incredibly accepting of sub-par experiences like this because of the friction and capability demanded of them to opt-out. My parents get tons of spam notifications that would pass your test of 'knowingly opt into receiving' but that when asked they say they do not want.
Finally there's myriad dark patterns that tons of apps use, like changing and resetting notification preferences among others.
I'd hazard a guess that observed opt-in rates far exceed users actual desires, so I wouldn't put much stock in them. I do agree that there are some people that like them tho!
Fwiw I've worked on both the delivery side (OneSignal) and developer side (Margins) so I've lived these choices and trade-offs. My believe is in terms of power dynamics, senders generally don't deserve their power to interrupt and should not possess that power. At best, they offer opportunities, which ideally are verified somehow prior to being presented to users. I'm happy that devices and ecosystems are moving in the direction of triaging and filtering sender content, as power needs to lie in the user's holistic, most pre-frontal cortex driven expression of their desired experience, and not just one moment's opt-in button they pressed.
Thank you for writing the article, good discussion points.
Yeah, that's true about the allow, and for sure marketing and product teams are deploying misleading consent priming which doesn't fully explain to the user what they're actually allowing in the first place, or setting baselines that are too permissive vs what the user is expecting.
> I'm happy that devices and ecosystems are moving in the direction of triaging and filtering sender content, as power needs to lie in the user's holistic
I don't disagree necessarily, but I see it as them putting themselves in a position to act as a toll collector, which has already happened with email and web search and is only getting worse with the introduction of LLM's into both of those things.
It's a bummer this article ended up doing much better than my email one, as I think that might better position the problem in a lot of user's minds and highlight just how much surveillance is sitting on top of those free inboxes.
For whatever reason, I get very few push notifications on my phone. Compared to my days at Blackberry, it's probably 10% as frequent that I get interrupted by my phone.
So good for me.
But there's some really scary stuff in here happening to other people that I'm not even aware of.
The default must be pull, unless opt in for push. At the moment I would like notifications once a day or once a week for most apps. But instead I ha e turned it off completely, because of the push abuse. If I can configure to pull all the notifications on a predetermined cycle, it makes my life even better
I've had all notifications turned off expect for my immediate family (parents, wife) for years now. I'd get rid of my phone before going back to getting buzzed and dinged constantly.
iOS really needs LLM-based notification filtering. This would take care of promotional notification spam overnight. It would even enable fine-grained user filtering like "notify when - someone is messaging me about plans for today."
Channels are a great first level, and iOS absolutely needs to implement an Android-tier version of this.
But channels continue to be abused, even on Android. When all deterministic controls fail...
Secondly, channels are set by the developer (or platform). In an ideal world, I want to define whatever channels I care about, and turn them on/off at will.
I need a feature to block my bank's incessant nagging about cash-back deals while keeping the ones about transactions.
Right now on iOS there is no way to do this. And yes, I've explicitly turned off the cash-back deals notifications in my bank app's settings and that is completely ignored.
I don't know if there's an iOS equivalent, but Buzzkill on Android is really great for this. I set up filters to hide all the stupid Amazon ad notifications.
> 2 to 5 notifications per week is the optimal range for most apps and exceeding it materially increases uninstalls;
Wow. Y’all must be much more tolerant of your time being wasted than i am. One notification from an app I didn’t need/request/expect is cause for deletion. 2-5 per week would be enough to go and rate the app 1/5 on the AppStore and actively recommend everyone I know to delete the app.
> visibility into all of this is poor by design, and getting worse.
Good! I pay Apple big money to protect me (user) from you (abusive app developer, abusive by definition since you talk about my attention as if it were your property)
Predictably people are moving back to SMS for notifications. Not as nice for linking to your app but once the user opts in you don't have to deal with the Apple/Google complexity.
I live a push-notification free life, aka no push ever.
I also saw elderly people receiving such notifications and
not knowing why. Then I realised that this system is designed
to abuse the elderly, so I am now totally against it.
Google/Outlook/etc intervening with email was a good thing and saved email with spam filtering and content ranking. Mobile Carriers have not effectively intervened with phone screening and voice calls are practically dead.
Intervening with push notifs could be a good thing. Notifs are approaching the point of uselessness. I turn all off by default now.
Oh man... I just wish someone invented some form of organization where workers could negotiate with employers in a more equal footing by doing this in a collective way.
If my phone interrupts me, it should either mean someone genuinely needs my attention right now or it should not be disrupting me at all. That's my notification set up.
Apps allowed to receive push notifications
Phone, Messages, Whatsapp, Apple Health, [brand] bank.
That concludes the list.
There is no reason any other app needs to be able to instantly ping me. Most apps are not notifying you because something matters; they are notifying you because they want your attention.
I do not need notifications about streaks, sales, recommendations, delivery updates etc. All that can wait until I choose to open the app. It is not urgent enough to justify interrupting me.
Yeah, this entire article is pretty transparent that it's from the sender perspective, and worried about platforms taking over "sender control".
Who is he kidding? The vast majority of apps have absolutely proven they can't be trusted to respect your attention. From my perspective, the more roadblocks the platforms put between unnecessary notifications and my phone, the better. And I don't think Apple or Google are some sort of heroes here, but I do believe their incentives better align with mine than the marketing department of some app I was forced to download because I bought a ticket once or something like that.
Notification categories are like mailing lists now. You may have unsubscribed from the daily deals email but you're still going to be auto subscribed to every new slightly modified category in perpetuity. Unless you fully disable notifications for an app (in Android at least, in my experience), new enabled by default notification categories are added all the time.
When they exist at all. Many apps that provide important notifications (like delivery tracking, drop-off time etc) put them under the same category as marketing stuff. You can't have just the transactional tracking, you have to opt-in for the marketing notifications as well.
The ridesharing apps are the most annoying about this. Yes I want to be notified when my uber driver is almost here to pick me up. No, I don't want a notification about yet another sale.
It baffles me that they do this. I have to disable push notifications from Lyft entirely, so instead they send me ride updates as text messages, which surely must cost them way more money. Why not just introduce a "ride updates only" push notification category and stop this madness?
Have you looked at the Uber app recently? 90% of it is promotionshit.
I am just looking for a fucking taxi.
Are there any VCs looking to give away a few billion dollars to disrupt the ossified, wasteful, poor customer experience taxi app market?
I had to disable from Android settings all LinkedIn notifications. I check it from time to time but I haven't missed anything, nowadays LinkedIn is mostly garbage
On iOS atleast, Live Activities are separate from Notifications. So I can still monitor food or grocery delivery even though I have turned off their notifications.
Now a few apps have started sending notifications through WhatsApp because they have my phone number. e.g. Amazon
Yeah, but I still see apps that don't implement those features. Mostly React Native/Flutter apps that don't bother implementing native features. On Android it's even more depressing.
I'm not worried about missing food notifications because they send me an email and a text (... and a fax and a hardcopy confirmation letter in the mail.)
On iOS atleast, Live Activities are separate from Notifications.
Should be, but not always. There are plenty of apps that still mix marketing and functional notification.
Hell, even Apple does this, especially on new devices.
[Settings]: Log in to your iCloud account to sync data.
Three minutes later…
[Settings]: You qualify for three free months of Apple Music!
Tbf both of those are promotional. If I'm not logged into iCloud, it's for a reason.
There’s the other direction too. You only get a couple toggles, and something you actually need is behind both, so you can’t not get all notifications
Another sneaky behavior in Android is that categories that have yet to send a notification, which of course includes newly added auto-enabled channels, are collapsed under the 'show unused categories' button.
iOS asks you if you want to allow notifications when each new app is started. You can just say no there and you're done.
It would be better if they were totally opt-in of course (1), but that's not bloody likely to happen.
(1) As in off by default with no questions.
I can see a certain category of people screaming that WhatsApp calls are broken if that were to pass… but I do agree that no one would scream louder than app makers wanting to retain their share of human brain attention.
As far as i know my Whatsapp is muted and the mute is muted again. But it still rings for voice calls.
It's integrated somehow with the phone app on iOS, whatsapp calls show alongside GSM calls.
I recently had to setup Microsoft Authenticator. It refused to register a code unless I enabled notifications.
You are a two factor app. I should never be in a situation where there is an unexpected login I need to verify.
I want scopes like Graphene has for storage scopes. I want this on my phone and browser - let the site/app think it has everything (cookies, storage, microphone, camera, notifications, whatever it wants) but it's all empty and does nothing.
Apps can know whether you granted permission?? That sounds like a security flaw.
This is basically required for clueless (and even not so clueless) users.
If there's a chat app I installed 3 years ago, with no intention of giving it camera access, and I suddenly need to use that app for a video call, I don't want to be stuck debugging broken camera issues for two hours. I'd much rather have the app tell me that it doesn't have camera access.
> This is basically required for clueless (and even not so clueless) users.
I can actually confess that this hit me. Almost nothing on my phone has permission to use my camera, including my web browser (why???). I assume this was done in a fit of pique upon discovering that the setting even existed.
Roll on (god knows how many years later) and I cannot get into the gym with the link I was emailed to have my browser read a QR because my browser is just a grey screen. It was only when the member of staff suggested permissions that I realised what was going on.
I'm the problem, it's me
The OS could tell you instead. If it is a camera app, the OS could tell you on install, that you can't start the app without given camera access, because that's what the app is.
They can, but there's an OS option that basically is "I'm going to say yes, but then effectively do no". Basically it'll pretend to the application that a permission is granted, but then just keep returning empty information or doing nothing with it. So notification perms would then be seen as enabled, but nothing is actually being send to the user.
Unfortunately Google isn't really exposing this to users, so you need something like App Ops or adb to set it up.
Yep. Just today I had a tram/bus ticket purchase app refuse to work unless I grant it Phone access.
Of course, that way they can so they can refuse to work until you uninstall or give in to their demands. There are other operating systems that present fake data at least.
Tip: The iPhone Passwords App has basic TOTP functionality (manually create a password entry and click “Set Up Code”). I have a few dummy passwords which are effectively just labels for some login codes - it’s one less App to install.
Unfortunately Microsoft Authenticator does more than TOTP and usually its not up to the user to decide which two factor implementation is accepted.
Some Microsoft setups ONLY allow Authenticator - can’t use 1pass etc. I have recently fallen into this pit
AFAICT any TOTP app (FreeOTP+, Aegis...) works just fine with Microsoft services (or Google, etc). You don't actually need to install several TOTP apps.
Microsoft Authenticator is not standard TOTP, but their own private flavor.
Okta has push as an option. Maybe msft has that too.
Key word there being 'option'. If you choose to use push as your mechanism then enabling it is obvious. If you choose not to the app should still work. You don't need push notifications enabled on an MFA app.
> I should never be in a situation where there is an unexpected login I need to verify.
Isn't that kind of the point? If someone else is trying to login somewhere with your credentials, your two factor will ping up?
Why would I want that? If it is not me, I am not going to allow the login. Making it a notification makes it more likely I could fat finger an approval.
I guess you can make the argument that you are then made aware of login attempts, but that feels more like something the host service should control.
> Why would I want that?
Because to get that far they entered your password? Which you might like to change?
You did mention: "You are a two factor app."
If they've got past your first factor, you might want to know.
I recently got an unsolicited OTP email from Microsoft, which led me to fear that someone had entered my password, but no: I eventually was able to confirm that the arrival of an OTP does not, in fact, require that someone enter anything beyond my email address. This is rather insane (I should not be having a blood pressure event due to Microsoft) but on the other hand I do understand the passwordless concept which is just a password-reset flow sans password-change. Perhaps a nice middle ground would be if the OTP email explicitly stated that my password was not entered.
This also happened to me about a week ago and I had the same reaction/discovery process you did. OT but I wonder if there was a recent ramp up in these attacks. It was done against an email I do not regularly use that was attached to my account as an alternate and haveibeenpwned confirmed was in a data breach back in 2020.
Some providers (looking at you, Intuit) don't seem to understand TWO factor authentication and will allow someone to bypass your password if they can intercept the SMS or email, and treat it as a normal login.
Our Okta is setup so that it usually does the two-factor before asking for password.
I would, but I don't need to know immediately. Plus you have the other vector of my phone sitting on a table and showing the notification to a person who can see it when they are trying to login as me.
I find it to be a poor default that sensitive data is shown on the lock screen. I change that setting as a first order of business whenever I'm setting up a new phone.
I saw a new marketing strategy recently: Someone tried to sign into something with my email. I didn't have an account, so they took the excuse to send me an email asking me to create an account.
I saw a new marketing strategy recently: Someone tried to sign into something with my email. I didn't have an account, so they took the excuse to send me an email asking me to create an account.
This has been going on since at least 2006.
Startups will "growth hack" by buying e-mail lists and feeding them into their password recovery tools.
A certain percentage of people will then follow the links and end up creating a new account on a service they had no interest in that now has their confirmed contact information, a new user, and a plausible reason to bombard them with marketing email.
I recently started getting emails from a company warning me that "I only had x days left to verify your account."
The account was supposedly registered for an organization whose name was somewhat similar to mine, so I thought somebody fat-fingered their coworker's email (the initial email was an invitation to create an account and join the org), but it might have very well been the tactic you described.
huh, is that why my google authenticator app pops up randomly? i always figured it was a bug in the app or in android.
This is all a consequence from running software that doesn't respect you and notification are just one of many symptoms.
I'd rather choose better software than let Google/Apple decide what software running on my device is allowed to do.
Usually you don't have any choice of the software if you need to use a particular product or service.
Yeah, the worst kind of software is the kind that interacts with the real world, particularly when chosen by clueless people at non-tech, real-world companies.
Conferences are great examples. You do want to submit your paper and go to a conference. To do that, they need your email address(understandable). That email ends up on dozens of email lists run by people who are doing "outreach" or something of the sort.
You can usually choose to "need" to use a particular product or service though. It's not always worth it, but there's a choice.
You mean it's a consequence of very large amounts of people refusing to pay for software, at essentially any other cost ...
Of course you could describe almost all of the internet that way.
It's a consequence of having platforms instead of protocols.
Suppose you want delivery notifications for your packages. The seller, by contrast, wants to spam you with marketing.
If getting the notifications requires you to install their app, they're going to shovel any spam into it that they can, and then they're writing the code that runs on your device. Whereas if the software on your device is controlled by you and the notifications are received using a standard protocol, you (or someone like uBlock) can create filters to only show the notifications you actually want and discard the spam.
But for that to actually work you need the software running on the client to be under the control of the user independent of which device or service they're using, and subject to competitive pressure. Otherwise the platform uses is as a means for lock-in and then filters your notifications in the ways that benefit them rather than you, or just does a lazy job because they know you've been deprived of having a lot of other alternatives.
Unless the task is extremely well defined, protocols don't really work.
Imagine you're a shipping company and lock yourself into a parcel tracking protocol. You then decide to offer the innovative feature of parcel lockers, which need a code (or an action on your device) to open. How are you going to make the thousands of weird homebrew clients that people are using on their jailbroken Nintendo Switches or whatever to behave?
> But for that to actually work you need the software running on the client to be under the control of the user independent of which device or service ...
In other words, you need the user of the software to pay for it's development. Since that won't happen ...
That isn't the only way.
I don't pay for most of my software and yet it still respects me. Of course it also isn't made by large corporations with marketing and sales departments.
The vast majority of spammers aren't large corporations but really small ones. Scammy ones. At least judging by my spam folder.
> From my perspective, the more roadblocks the platforms put between unnecessary notifications and my phone, the better.
I know lots of apps behave badly when it comes to notifications but I'd still prefer if the apps controlled the level of notifications they sent. I could, of course, reduce that client-side, but I don't see why I'd want Google or Apple or any other intermediary see or control the notifications.
If an app behaves inappropriately, I could uninstall it. If a gatekeeper like Google or Apple prevent an app from sending me notifications, I'd have to change my OS, usually my hardware, too.
This forces millions of users to individually monitor and fix dozens or hundreds of apps all the time - something most don't have time for and leads to an awful experience. Centralized controls are better for the user.
TFA discusses at-length how APNs and FCM are necessary intermediaries regardless, effectively creating a technical duopoly on 'push'. We all agree it would've been preferable for things not to have gotten this way, but here we are.
> I do believe their incentives better align with mine than the marketing department of some app I was forced to download because I bought a ticket once or something like that.
Align better for now. It will get enshittified.
I try very hard to avoid installing apps specific to a particular business or organisation. So far I have only had to install a government app and some from banks. Even those are avoidable (but it would be very inconvenient to do so).
The biggest problem are apps that do both. For example, I want Uber to notify me when my driver has arrived, but I don't want it to notify me when they have a special 10% discount on my next 5 rides. It's not straightforward to block one but not the other.
If I order an Uber, I already know it is coming. I was the person who ordered it.
This is how taxis worked for decades before smartphones existed. You phoned for a taxi, then remained vaguely aware that it would arrive shortly.
The question is whether a single “it has arrived” notification is worth the surrounding noise: “driver accepted”, “driver is nearby”, “rate your driver”, “here’s 10% off your next ride”, and so on.
In most cases, it is not. The useful information is either already obvious (you can see the car outside) or you have re-opened the app to check where they are.
Operational and marketing notifications should never share the same permission. Until that is enforced at the OS level, I will treat them all as unnecessary spam.
The difference is that taxis would either give you a call that they were here or they'd just wait for. They don't care either way if you show up or not because the meter is running. The Uber is gone if you don't show up in 5 minutes. That is if you are lucky and the driver didn't mark themselves as "here" when they were 2 blocks away, which seems to be the norm here now.
Android has different types of notifications for apps and can have them filtered separately. Unfortunately, some app makers like Uber are bad about labeling. Google would need to enforce labeling so transactional and advertising notifications are separate.
“Bad about labeling” is pretending that might be an accident. Uber has repeatedly demonstrated they will do all the dark patterns.
iOS has the thing they call “time-sensitive notifications” which is a flag you put when submitting notification that is supposed to be Really Important Right Now. Unfortunately it’s not easy to mute everything that is not time-sensitive
The point of notifications is the convenience of not having to constantly check your phone for every single app you have (amazon delivery? just eats delivery? uber booking? claude finished its task?).
> The point of notifications is the convenience of not having to constantly check your phone for every single app you have (amazon delivery? just eats delivery? uber booking? claude finished its task?).
My phone has been on DoNoDisturb since 2010 or so. Here's the reality: I don't check for any of those things. Delivery drivers can ring the door bell. If I'm very hungry I'll keep the app open and check where they are. I literally do not care to be notified about any of the things that apps want to notify me off.
Anyone who cares to reach me knows to ring the phone twice in case of emergency to get through DnD. Anyone else? The best time to call is text me. Or schedule a time.
As for Claude, the point of clankers is that they work in the background. The robot can wait, their infinite patience is a feature.
Same here! If this phone made an unexpected sound I'd smash it with a brick.
I guess you probably have no dependents and never been oncall then if you are on no disturb. For many people, having to poll the state of multiple ongoing tasks is time consuming itself or/and focus breaking enough that some apps are deserving of having notifications.
Manually polling multiple items as you go around your day is stealing valuable mental bandwidth that could be used in better things.
Oh I’m on essentially permanent 2nd tier oncall and occasionally 1st tier. Pagerduty has an exception configured. If you message me on slack it shows up on my home screen and I will probably notice in the next 15 minutes or so.
I find polling less disruptive because my phone or watch are almost always nearby. Being in control of when to get interrupted feels better than having stuff constantly pop up while you’re doing something.
Just holding my partner’s phone spikes my cortizol. You try to type 2 sentences and get 5 notifications for random other apps and messages popping up on screen in the meantime. Literally one notification every 10 seconds on average. It’s horrible
> Manually polling multiple items as you go around your day is stealing valuable mental bandwidth that could be used in better things.
I both agree and disagree with you. I disabled notifications for everything and found myself refreshing email too often. But if I have the notifications, I can get disturbed too often which also takes mental bandwidth.
For me I found the best tactic was to selectively enable notifications (whatsapp for just one person), and delete (not silence) everything else - I don't have email on my phone now - the temptation to check it is more than the need to have it. As for things like PagerDuty, I have it send an SMS and phone me instead.
> the best tactic was to selectively enable notifications (whatsapp for just one person), and delete (not silence) everything else
Hundred percent agree. I wouldn't support having notifications for everything enabled.
Your house already has a built-in notification system, which can be activated by the delivery guy, once he is physically at your location.
A built-in notification system that everyone from your local politician asking for your vote to your Jehovah Witness recruiter can and does abuse. Way to waste your time.
You can silence it just as easy and in difference to the digital alternatives, the amount of time wasted is symmetric.
And the inconvenience to constantly having to check your phone
>If I order an Uber, I already know it is coming. I was the person who ordered it.
Which makes me wonder why you have notifications for your bank and Whatsapp enabled.
If I have an account, I know what transactions are coming out of it. I was the person who owned the account.
If I have someone's number, I know if I want to see messages from them. I was the person who gave them my number.
Seems really sill that you have notifications enabled for those apps. If you care about missing something, you'd just check them anyway.
I have notifications enabled for my bank because it may alert me to transactions which were not made by me, and because it does not abuse that permission to send me marketing spam.
I have WhatsApp notifications enabled because it is the primary way people communicate where I live. If my elderly mother messages or calls me, it will most likely be through WhatsApp.
Both of those notifications contain genuinely important or time-sensitive information which may require action on my part.
That's the distinction between them. A person contacting me is fundamentally different from a brand attempting to engage me. Transaction alerts are fundamentally different from “your order is out for delivery”.
The criteria is not “did a thing happen”. It's whether the notification gives meaningful new information that is important or time-sensitive, and requires me to take action.
Most app notifications fail that test completely.
>If I have an account, I know what transactions are coming out of it. I was the person who owned the account.
What if you receive notification about transaction that you didnt make? :)
For me, it's quite straightforward. If an app makes an unsolicited spammy push, it's notifications-off. No exceptions.
Snapchat has to be the all time worst offender to me about abusive level of notifications. Luckily, you can turn them off, but holy cow batman, that's a lot of notification options to deal with.
For me the worst is NextDoor. I don’t have the app installed, but they also have email notifications. There are seemingly 100 options and I turned them all off when I first made the account. Periodically they add new ones and auto-enable them for everyone. There is not universal way to shut them off, short of blocking them all together or deleting my account. The account was such a pain to setup that I’m hesitant to delete it, for the 1 time every couple years where it’s useful.
Yep, I straight up deleted my Nextdoor. Nextdoor itself is spammy and ad ridden. And most of the content in the community is also just spam anyway.
My neighborhood has a WhatsApp. I engage with it if I feel like it. Works great
I flagged them as spam the other day.
Even worse with ND e-mails are how they've absolutely perfected the cut-off character limit for what's being posted in your area. So my inbox is just perma-barraged with click-bait-y "This place on Smith Street has the best...", "Health officials are investigating an outbreak of...", etc.
That is why I just filter ND emails to spam. Those patterns just reaffirm I don’t want the kind of content they offer.
Remember when Android used to let notification senders hijack turning your screen on, Snapchat used that one a lot.
That's "growth hacking" for you
Yes. I’d rather live with the temporary inconvenience of needing to open the Uber app to check the status of my ride once a month than wade through notification spam on an intermittent basis forever.
Agreed, there’s a level of trust and as soon as the app breaks that trust with spam, notifications get turned off for that app.
I just refuse to grant permission as my default. If I ever feel like I’m missing out, I can turn it on later. Usually I don’t and if I do I quickly regret it.
Has been like this on my phone for a while. It's crazy when you see someone who hasn't blocked everything and their phone dings multiple times a minute.
It's a failure of iOS architecture to not force applications to tag each notification with labels. App developers have to build notification management themselves for fine grained control.
Android has notification channels, but I'm not sure how widely it's used: https://developer.android.com/develop/ui/compose/notificatio...
This is the kind of thing where we actually would like to have store policy enforce correct use of APIs.
No one willingly says "yes" to advertisements, but people will say "yes" to important-updates(-and-advertisements).
Then why is it whenever I watch someone use their computer they always accept cookies?
Because companies are trying really hard to hide the "no" button: it's a single click to say "yes to all", but a safari through dialogues to say "no to all"
Same with websites like Youtube who don't understand a plain "no" but offer a fake choice between "yes, harvest all my data" and "ask me again later". That isn't consent, it's coercion.
Because people don't actually read what they are clicking on or even understand what they're doing. They just want to make the annoying banner go away. Same reason why people mash the next button when installing software.
1. accepting cookies is not the same as opting-in to advertisement
2. because most of the time, any other option is bloody inconvenient
They are choosing the lowest friction option.
Hundreds of thousands of people declaratively opt into receiving marketing with informed consent on a daily basis. Just because you don’t does not mean other people are like you.
Yes… seeing my wife’s email inbox is mind blowing.
Maybe she didn’t opt in, but she will never unsubscribe from anything.
Emails from every site she’s ever shopped at.
> Emails from every site she’s ever shopped at.
This too is frustrating. Spam is not allowed unless you have a "prior relationship".
But fuck me, that single toddler's bike I bought many years ago for my then toddler no longer qualifies as us having a relationship.
I mean, it's kind of an insurmountable obstacle. Why bother trying to unsubscribe when you're always gonna get spam anyway? It's just gonna come back.
Also, websites are shady. If you put in a required email, they'll usually automatically check a little box for you that says "allow us to ruin your inbox?" How helpful of them.
And, I'm not even convinced that checkbox does anything.
It's definitely not insurmountable. It just takes a little bit of inbox maintenance. Pressing unsubscribe and report when I have spam in the inbox, 5 seconds here, 5 seconds there. I still get spam, but it's minuscule compared to not doing this.
I just flag every marketing email as spam. It's much more effective than unsubscribing since it tells your email provider to just redirect everything from them to spam, and it causes trouble for the sender.
Exactly. Unsubscribe is for newsletters you consciously subscribed to but no longer want. Anything unsolicited that isn't a genuine one to one communication is spam and all the "unsubscribe" option does is verify that your email is active and will be able to receive more spam.
It's not insurmountable. Spam filters have been around for decades. They're pretty good. If I didn't expect the email, I train my spam filter that it's a bad email. There are a few that get through, maybe 1-5 a week, which require flagging.
The checkboxes seem to be a placebo. Sometimes there isn't one. Sometimes it doesn't do anything. Sometimes they say "updates on your order", which apparently also means future products you might want to buy a week after you receive your order. (Marketers' English seems like a foreign language to me).
I realise that CANSPAM as a law in the US is titled appropriately but I live in the UK where declarative opt-in is a must apart from a couple “soft” opt-in scenarios.
People. Opt. In.
Stop conflating your preferences with other people’s.
Spammers ignore both opt-in and opt-out, they do not care about UK or US law. That is where hosting provider enforcement matters.
Agreed but I don’t find the framing of labelling marketers as spammers particularly accurate or useful.
"Report as spam and block sender"
Repeat until the emails stop.
Is “informed consent” that little checkbox that is checked by default? Or is it the one with the wording that says something about “discounts and offers”? Or is it the one that’s enabled because it’s a “new category” that didn’t exist when the user signed up so why not require them to opt out? Oh, I know, maybe you’re talking about the “enable notifications? Yes/Ask Me Later” dialog that is pushed on them every single time they open the app.
I’m sorry but if you honestly think the number of users who receive marketing spam have expressed “informed consent” you’re fucking high. There is a multi-trillion dollar industry devoted to tricking people into opting into spam. Stop pretending these people are expressing any consent at all.
I live in the UK where we actually have sane regulations around this stuff, I realise laws are different and folks in the US in particular don’t have much protection from spam.
Apple should add "promotional notifications" section to iOS, then ban everyone who don't put their marketing bs into that category
Yes! iOS 27 needs to categorize notifications using AI. Apps aren't supposed to advertise to you, but some don't allow for that distinction. I want notifications on for when my sandwich is arriving, but I don't want push notifications for a promotion. Some apps are good about this, others don't allow that granularity. I hate the all or nothing.
On the flipside, I have an app that sends notifications. We don't abuse it or use it for promotions, and APNS and Google's version works perfectly fine for us.
Apple isn't your friend though.
edit: downvote all you want. Fact remains that there is no way currently to block advertisement notifications and no disincentives for those who use them.
Send everything to the iOS notification summary which you then don’t look at. Uber and others can send time sensitive notifications and those don’t go in the summary. It’s basically a junk notification folder.
Works well.
Some banks also do this, and offer no way to segregate marketing from utilitarian push notifications. This is borderline abuse of trust IMO.
It's not borderline, it's absolutely crossed the line.
periodically open the app every few minutes or so. once the driver is 5 minutes away -- go outside and wait.
it's a tradeoff. eliminating notification spam means behaving more synchronously when booking a taxi. i don't mind waiting outside for five minutes. especially if i'm not getting a random ping when i'm definitely not booking a taxi :shrugs:
I prefer temporarily toggling notifications on because I really don’t trust my internal metronome.
You could set a short timer on your phone. On an iphone, its two clicks away.
Enabling notifications is not difficult. And the driver could arrive before or after the initial estimate.
Most people aren’t aware but there are laws that require granular notification consent. For example the GDPR has it. I’m currently fighting with a major bank and educating them about my rights. I want to receive security related notifications but not get spammed by “get a loan up to 50k without lifting a finger” type of bulls*. Send send this almost every week..
The user legitimately considers the application as hostile - hence sandboxing... Notification spam filtering is now the obvious need at the sandbox's edge, with the whole customizable arsenal we have come to expect for our inbound mail. Of course, Google will not cooperate with anything likely to reduce sacro-sanct engagement !
I definitely run all my emails through an LLM filter and wish I could do the same for push notifications!
Except that they did. Android has notification channels. Now, I suppose we could argue that Google could be more ham-fisted about forcing apps to use them, but that's murky.
In fact, Uber on Android does use these notification channels. I just have "All Promotions & Recommendation notifications" disabled, and then "Taking a ride" channel enabled.
>Now, I suppose we could argue that Google could be more ham-fisted about forcing apps to use them, but that's murky.
I disagree. What the is the point of forcing everyone to use the Google Play Store (or whatever app store on iOS) if the store doesn't stop spammers?
People complain about Uber, but Lyft does the same shit. I got a promotional notification from Lyft and could not disable it without disabling the main notifications that tell me when drivers were arriving.
If app stores were useful instead of just rent-seeking, they would kick Lyft off until they stopped doing that.
I don’t know about uber specifically but most of the apps I use have a separate toggle for things like marketing. Maybe it’s an EU thing?
The modern pattern in anywhere that allows it is to have dozens of ambiguously labeled toggles for nominally different notification channels, described only by a minimally brief and maximally ambiguous label. All begin as active until the user, in frustration, goes in and exhausts themselves disabling individual options without being sure which one is going to turn off the one single thing they actually want to be notified about.
Then next month, you create a new notification channel for your new promotional messages because too many people opted out of the old channels. You default that new channel to opt in, to make sure the user gets their chance to experience it and share in the delight you mean to share with them.
Presumably, you continue this until you have hundreds of such toggles and presumably some kind of dedicated Toggle Engineering Department that oversees them all. Nextdoor, Meta apps, LinkedIn, and countless others all appear to be competing for the most such toggles.
After all we wouldn't want the user to miss out on our promotion of 10% off your next refrigerator. They bought a refrigerator from us just 6 months ago, after all!
Though Android does help a little for existing toggles by giving you an 'About 129 notifications per day' blurb below the entry.
It also puts a throbbing highlight (for a few seconds) on whichever channel is associated with the notification whose gear icon you used as a shortcut into the channel list. At least for Pixel phones.
That's how the design is supposed to work. But marketing realizes that no one voluntarily receives ads, so they mix em.
Tell me use iOS without telling me you do. Android has separate notification channel toggles, so I've turned off the marketing ones. I was shocked and aghast when I spent a year trying to use an iPhone that it didn't do this. Part of the reason I went back to my trusty Pixels.
I use Android. Lyft put marketing notifications in the default notification channel on my device. If the Play Store were useful, they'd have banned Lyft until they fixed it. (haven't gotten one in a bit so maybe they did (or maybe I set something so that the app could only message me while it's active))
While iOS doesn’t do this at the OS level I’ve never seen an app that didn’t have these options. I assumed it’s required by Apple.
Lots of iOS apps have the option, but ignore it and send you push ads anyway. Apple may require it to be present during app review, but they don't seem to enforce that it's used correctly.
Does Google actively police app's use of channels? Is there any mechanism to stop apps abusing "time critical" channels and sending unwanted marketing?
They technically allow you to, but make it really annoying to. Uber for example:
Account > Settings > Accessibility > Communication Settings
Most apps are cross-platform. If you're already required to do it on Android, going out of your way to avoid it on iOS doesn't make a lot of sense.
It absolutely makes sense (in a capitalist sense). Then you get more money/engagement/whatever on all of the other platforms.
It's the same reason Microsoft built functionality to let users in Europe have links open in their default browser instead of Edge but blocked that feature for the rest of the world.
Yep exactly this. The app developers are the problem, but Apple and Google are not helping here.
This. So much this.
And the worst part is that Apple could fix this in a heartbeat. Uber is straightforwardly in violation of App Store policies about "no advertising in push notifications", but a) they're too big to fail and b) Apple advertises via push notifications all the fucking time, so they have no leg to stand on here.
It's infuriating that the one thing the App Store monopoly could be useful for isn't even actually used in practice (if you're big enough, ofc, you and me get to eat shit if we try to evade App Store policy).
Instagram is the worst offender, I only want to receive message notifications, but I got notifications about inane random stuff I've tried to disable but it won't work. I ended up having to disable notifications altogether.
Instagram drove me so insane with that that I spent a while searching through the app to figure out how to disable it. There's a way to do it, and for a while it worked; I only got notifications about things like direct messages and posts from a few people I specifically told it to send notifications for, but I never got the "recommended" posts.
Then I got a replacement iPhone and reinstalled Instagram.app, and it defaulted to "show you notifications for everything we think you might be interested in" again, and I was too lazy to spend all that time relearning how to disable those notifications. I disabled the notifications entirely and now I open the app once a week or so to check in.
I had to do the same thing with Facebook years ago. Now I open it once a month to see who from high school got married in the last month and click the little "heart" icon and scroll until I get bored (~2 minutes). Can't trust Zuck with notifications.
> Now I open it once a month
So do I. Login on the web incognito.
Instagram run their notifications via an auction mechanism (which I suppose makes sense for an advertising company that likely built a lot of RTB systems).
https://arxiv.org/abs/2302.04835
I believe the App Store policy is you have to have a setting to disable ads. And Uber actually has it (though it has 8 different channels or so, apparently "Uber teen accounts" marketing was added recently).
I used the setting and am not getting Uber ads (only Uber ride notifications).
Currently my biggest problem isn't ads, it's all the apps now will find ANY excuse to send you a notification in order to keep their "Daily Active User" count high.
You turn off more and more categories and they'll still find a reason.
Presumably enforcing this would trigger an immediate legal response where Uber claims Apple is using their monopolistic control over the App Store.
I would love if Apple enforced that rule, but they certainly don't
WellsFargo does that. Important notification and advert-BS on the same channel. If you block their notifications you don't get the near-real-time Zelle alert. Enabled you get what you want and also YOU MIGHT BE PRE-APPROVED!
On Android with notification categories it is, but iOS doesn't have that. Also, I think it's mostly a trust system. But Uber in particular does actually do it right, and you can just turn off promotional notifications.
"Marketing never met a communication system they didn't want to co-opt"
(I'm reminded of this every time a client want "WhatsApp support" in their (commercial) app, so they can "communicate with customers".)
But equally every user will have a different subset of apps they want notifications for.
For example shift workers need to know when they've been allocated a shift. Or when a shift has opened up (because someone scheduled failed to arrive etc.) One group of users consider this really important, another group of users treat it as spam.
But, per the rule above, unfortunately "useful notifications" can easily be subverted by marketing notifications. Yes I want to know my delivery driver is outside, no I don't want to know that you're running a special this week.
Unfortunately there's no way to solve this problem technically. Bad actors can (and definitely do) behave badly. But ultimately the system should work for "good citizens". In other words, the user should ultimately determine what they want to see of not. And if an app has "notifications on or off" as the only option then the user should ultimately determine that setting. Not Google. Not Apple.
Building society around the lowest-common-denominator just ends up sucking for everyone. We should actively promote good behavior, while allowing bad behavior to be punished. Not just banning everything "because it might be bad".
I totally agree. Right now the apps that can notify me are phone, text, email, what's app, and a few bank apps. You are 100% right about turning it off on everything else.
I also stopped doing store loyalty cards about 7 years ago and it's been fantastic. I actually get a lot less junk mail and spam/"legit" marketing emails. I don't have a gob of cards to sort through.
Corporations should not speak unless spoken to.
You're conflating "push notifications" with "being alerted about push notifications." I have many "important but not urgent" apps on my phone configured to just silently add their push notifications into iOS's notification center.
With an app configured to do notifications like this, no banner shows up at the time the app's notifications are delivered; and these notifications don't even show up visibly on the lock screen. You only see this type of notification if you choose to actively scroll down past the "timely" notifications that do get delivered onto your lock screen, to "catch up" on all your notifications.
Basically, these notifications are relegated to an "email inbox" that you can check or not check as you like. But unlike your email inbox, you can go "inbox zero" on your notification "inbox" whenever you like without worry, since notifications (unlike email) are inherently prohibited from being a critical path in an app workflow.
Just curious, what do you do with the increasing number of companies that use push notifications as a form of advertising venue, and how do you differentiate the security warning notification from your camera app from their special weekly annual sale notification?
The marginal cost of each notification is so low that companies simply spam users nonstop, and their A/B tests shows that the revenue is increasing. What's being lost though is that we're getting more and more agitated with these brands and their uncapped malicious behaviors. This is also true with their UI and UX as well, they keep adding banners with incredibly small close buttons, because someone will continue with the shopping after accidently missing the tiny button, and that's an added sale, who cares about 99% of users who are fuming with dissatisfaction, what are they going to throw away their $200-300 smart home device because companies abuse them?
On my Android, I aggressively mute or uninstall any apps that send push advertisements. I haven't seen one in more than 6 months now. If they think they need to advertise to me, I think I need to do without them.
Apple and Google failed to make push notifications usable for the past decade. Most important notifications drown in a sea of absolutely irrelevant nonsense. It's a very primitive mechanism where many apps compete for very little screen real estate. Beyond "something happened!" there isn't a whole lot of information in most push notifications. They are mostly not very actionable and very vague about what actually happened. And "something happened!" just isn't very useful information to me. This has de-valued the whole notion of having notifications. Whenever something interesting actually does flash by, I often miss it or can't find it back.
The push notification UX is just beyond terrible and it just got worse over time as app developers tried abusing their super power of being able to interrupt the user at will and Apple and Google tried to get on top of that. The net result is something that's very mediocre for the handful of valid uses I have left for notifications. My list is similar to yours. Things like bank approvals, 2FA stuff, etc. are useful mainly as deeplinks into apps. But other than that, it's just not worth dropping whatever I'm doing and staring at my phone.
The most used apps on my Android phone (older Google pixel model) are Firefox and gmail and just a handful of other things. As a notification channel, my email inbox is actually far more useful than mobile push notifications. They are more actionable and informative. And I can individually unsubscribe them or filter them out and easily find them back. Most apps can do both and that makes the push notifications inferior and redundant.
> The most used apps on my Android phone (older Google pixel model) are Firefox and gmail and just a handful of other things. As a notification channel, my email inbox is actually far more useful than mobile push notifications. They are more actionable and informative. And I can individually unsubscribe them or filter them out and easily find them back.
There’s also substantially more filtering happening in the inbox which is mostly useful from a user perspective.
Yahoo literally wrote a paper more than a decade ago showing how they can model predictive causal chains for emails they expect you to receive, as an example.
https://dl.acm.org/doi/10.1145/2740908.2741694
Gee lets take a 5 inch screen phone and have every notification stack up in 1 inch worth of space. I really hate ios18. Too bad ios26 is even worse.
> Apps allowed to receive push notifications
> Phone, Messages, Whatsapp, Apple Health, [brand] bank.
Anyone else annoyed by the fact that you can set up do-no-disturb, with exceptions for certain phone numbers, but it doesn't work for apps like WhatsApp?
I remember a while back I also had this issue on iOS (maybe around 2022 or so), though lately seems to be solved and works as one can hope so. When you're making the exceptions, do you explicitly input phone numbers or some other method? I selected contacts from my contacts list and it does the job. This is for iOS in my case. I'm not familiar with Android, so cannot give any input there..
On Android I can "star" certain contacts.
Exactly. Senders have earned the questionable reputation that they have because they rabidly want your attention whether you want to give it or not.
I used the Southwest Airlines app recently and allowed notifications so that I could find out about things like delays and gate changes (both of which happened on my trip). Less than a week later I'm getting ads for travel "deals" pushed as notifications.
Unsurprisingly, it was difficult to find the notification setting, which was on their website, not even in the app.
Maybe it's for the best. The best practice is to have as few apps as possible. The moment an app is abusive with notifications, you know it's time to drop the app anyhow. A lot of people need that one final push to drop the app, so this could help.
Agreed.
And let's not forget focus modes... I have them that narrow greatly my default set of notifications, so I have a 3 tiers of notifications.
It's like the complaint I used to hear all the time: "Slack ruins work for me! OMG I can't work with constant interruptions!!" That is bewildering, because if that's how you feel, you haven't tuned your setup. Slack never interrupts me, yet I am response enough to slack messages. No one has ever complained about my response time. And I'm probably the most-messaged person on our Slack.
> if that's how you feel, you haven't tuned your setup
The withering cry of the software engineer "just tune your setup!" This is simply not a thing that people will do.
The defaults are so, so important. They are crucial. The vast majority of people rely on the defaults to be sane. The defaults should be sane.
The other problem with Slack is that it just straight up... doesn't do what you tell it to. I have a set of notification settings that work for me. Slack goes ahead and just does something else, and you simply can't fix it to do what it's told. (Or couldn't, anyway; I've been off Slack for a while.)
Absolutely agreed.
How much time must everyone be asked to waste to “tune” a working set of applications to something reasonably sane for human beings.
Sure, what is sane for one human might not be for the next, but it’s not as if trends cannot be discerned.
How ridiculous would it be to be told “if you don’t want people constantly barging into your office, lock the door”?
It wasn't much time at all. Honestly, the push back on this always baffles me.
And when I had an office, I closed (not locked) the door to signify I was in a focus mode. I don't get your point.
The idea that software like Slack could be setup as "one size fits all" is just ludicrous to me. We have options because different people require different settings.
For Slack, I find just changing the default notification sound to a simple and subtle ding works well.
When I’m focused, I don’t hear it because it’s too subtle. But when I’m not concentrating on anything, it’s more noticeable and I don’t mind the distraction.
This might not work for everyone (“YMMV” and all), but I’ve personally found it a very effective yet simple solution.
If you are very present on slack, ofcourse you dont feel that you are interruped.
I don't know what that means.
I have no audible sounds from notifications. They don't go to my phone, with few exceptions. I get no popups. Yet, I am responsive. It was trivial to set up.
I would say the same applies to background processing as well. A random app that I don’t interact with launching every minute and wasting everything from battery to network bandwidth is simply not acceptable, and most of the time they’re loading adds or doing some other stuff that serves me no good.
I wish I could set this as the user. Apple ties background app refresh to the frequency of use, but that sucks for self-hosted photo backups. I use Immich and I don't open it too often, so Apple breaks my chosen backup system for my photos.
My notification setup is more elaborate (for one, I do keep social media notifications on, but silent) but yeah I agree in general. It frightens me seeing some other people's notification shades where they have 20+ spam notifications from all kinds of things that I wouldn't even consider installing an app for, and they're somehow fine with it.
Same: Phone, Messages, Calendar, Apple Health... nothing else can send me notifications.
On my work I also disabled all notifications except for the calendar. Even slack message our main tool for communication is not allowed to send notifications. It is almost a productivity hack :P
I agree with your points.
That said, my view is now (not novel, or unique) that I am not the customer in so many cases. Any app or platform with the slightest hint of an advertising end-game restructures my usage as the product.
The customer is instead the sender (or advertiser). So, I can't expect ideal app behavior and usage based on my intentions because I'm sold (as the product) rather than the other way around.
Maybe a cynical view, and there are exceptions, but don't think I'm far off.
The worst are apps that bundle genuinely urgent notifications with maketing brain-manipulation promotional crap.
Uber is a notorious example. I do genuinely want Uber notifications for when I use Uber. I do not care about whatever promotion it pushes at me.
I classify them even further.
I have broadly the same list as you do, but stuff like WhatsApp, Messenger, and other "non primary" communications platforms have silent notifications in the sense that they're not allowed on the lock screen or Home Screen. They simply display a notification counter.
Stuff I care about that I can't do anything about "right now" are allowed on the lock screen but quietly. That includes messages from the kids schools. Most is not even that important, like field trips "next week", but once in a while there's an "important" message I need to deal with.
Might as well use a dumb phone
I don't get what you guys are doing to be so bothered by notifications. I get them on my wirst and even then it isn't enough to take away mental bandwidth.
My bank likes to show offers, like a 10% discount in tires, but I have no car. Perhaps tree or four irrelevant messages per day.
I have MouseTimer that is an alarm that is nice to show to kids when they must wait or do something for 10 or 20 minutes. It should be able to ring and sometimes show notifications.
Health? Why, are you worried you'll miss the notification that you have a heart attack?
As for Whatsapp, maybe you're not in enough group chats that you still allow notifications...
You can mute group chats in whatsapp.
I also allow emails, but I'm very agressive at filtering promotion/junk emails with skip inbox rules: https://support.google.com/mail/answer/6579?hl=en
> Phone, Messages
At this point, I'm pretty much in some form of DND at all times. I have a very small list of people that I allow the device to notify me at any time for calls/messages. Everyone else gets silenced and I'll get back to them when I choose. All other apps have notifications disabled and I'm constantly nagged about it when using those apps
I went even further and my small set of the most important applications runs in the background - rest doesn't have that privilege. I've treated my spare Samsung phone same way.
I also don't use Siri either beyond setting timers and lights in home and every application is also excluded from being "suggested". Apple for 14 years didn't bother to add support for Polish so it basically remains useless.
Exactly. Same for me, except I don't have an iPhone and therefore no "Apple Health". I will take care of my own health, or not, on my own.
So I would say: only humans can send me notifications. That includes me in the case of 2FA. But no machine ever, for any reason.
I'm personally just at messages. And even then I make it clear I respond when I want to. Only phone rings/notifications I get are for those in my contact list.
Take your phones back. Life is immensely better these days.
Your position is that of any normal human. Google is committed to evil however, just look at how playstore notifications are tied to sales spam. Want payment notifivations? Gotta take the ads as well, not seperate toggles, one toggle. Drink liquid shit you tech peasant. Oh? this hostility drove you to f-droid? We'll unilaterally decide every device r belong to us, so we can disable competition we dont approve of. Welcome back to the liquid shit trough, peasant.
I disable all group chat notifications too, only direct messages trigger notifications for me.
Fully agree, the apps are to blame for misusing notifications for marketing and ads, they are the ones doing this to themselves.
I have it turned off for my bank. For some reason Bank of America doesn't allow me to sign in with Face ID. I always need to get a text. Only reason I keep them is because I like a brick and mortar bank nearby.
There ought to be a flag to group all such notifications together and present them when the user wants to get to those notifications.
Attention(/time) is our most valuable resource. Protect it ruthlessly.
I have my phone always in Do Not Disturb. That's it.
if a app messages me it is uninstalled. we have too many installed apps anyways.
It's absolutely disgusting how most tech companies use notifications as an advertising or addiction building channel.
On the rare times I use an app like uber eats, I uninstall it directly after because the app sends multiple adverts a day through the notifications. I want a notification purely to tell me the driver is almost here. And nothing else.
Phone, Calendar, Health - that's it for me
I've noticed a priority inversion in recent iOS. Want to send me an SMS that matches a ban-list regex from a third party app, from a foreign phone number / obvious spam farm? No problem. The app to block you was auto-uninstalled, and the iOS notification filter will mark your message with the highest possible priority.
Want to continue a 300 message thread that I've been responding to? You're listed as my emergency contact, and called multiple times? Fuck right off. Straight to spam.
It's almost enough to get me to carry a second dumb phone or grapheneos device just so I can text and receive phone calls.
For me I definitely need Calendar and sometimes Clock (alarm). iOS is constantly freaking me out by prompting me whether or not I want to continue receiving notifications from those apps. To me those apps exist entirely for the purpose of generating notifications and it terrifies me that by repeatedly popping stupid questions like that, I'm going to accidentally answer wrong and effectively delete my most important app accidentally. It boggles my mind that somewhere someone thought Clock and Peggle were basically on equal footing here.
Spot on.
To your list, I would add a calendar and reminders app.
Apple and Google are pushing push notifications?
No way!
Pun aside, it makes sense for a notification framework including a notification delivery network to be built into these mobile OSes, because the alternative (letting applications run arbitrary background services) is usually worse.
I feel like this article reads like the author is upset that Apple + Google prevent / control certain types of notifications (read: spam)
> Cross-sell, upsell, education and discovery can work on push
Push notifications should only be for transactional notifications. I don't want another inbox for junk.
Agreed. I have a doc appointment app, where I like the notifications to be on for reminders etc.
Lately they started sending marketing messages through that channel. Now I’m sure it’s possible to turn off the marketing messages. But I bet most people don’t know and won’t change that. It’s super annoying.
Same for things like Uber.
I do want to know when a car is arriving.
I don't want messages asking if I'm hungry.
Hi whstl,
Are you hungry? Open your Uber Eats app now for 10% off.
/this message sent through PalantirFinder -- from marketing and coupons to ordnance, we deliver everything!
> I’m sure it’s possible to turn off the marketing messages.
Uber may have that functionality, but a surprising number of other apps don't - for example Makro, Tops, and 7/11 Thailand, three very popular Thailand retailers, use notifications for when an order is out for delivery, about to arrive, etc. But they also send constant promotion notifications every day, even with audio alerts enabled.
We must, at some point surely, reach an inflection where even everyday people are sick of this shit and start smashing their phones right?
There has always been "unpluggers" [0] amongst technologists, but the vibes are bad and getting worse. I feel like that is getting more common between "normal" people I know, but maybe outside of my country town bubble its not happening.
I was thinking we're only one or two big influencers away from a cascade, but then the ultra-influencers are never really going to commit because its their livelyhood and saying throw your phone away is self-limiting on the viral aspect.
I guess we're just stuck under the boot.
^0 https://biggaybunny.tumblr.com/post/166787080920/tech-enthus...
> We must, at some point surely, reach an inflection where even everyday people are sick of this shit and start smashing their phones right?
Never underestimate the ignorance of the average person…
I am talking about those people who consume content while ads blink around the content in a all four directions and they don‘t even actively notice.
I absolutely hate medical marketing. I recently decided to switch dermatology practices over a combination of bad data management, unwelcome marketing for cosmetic dermatology products, and unsolicited SMSs. I never consented to receive marketing or texts (or to lose my data in a data breach, or to be billed for the in person services by practices other than the one from which I received services and ignored unopened as spam because I didn’t recognize the sender, etc…)
But I digress.
And soon, those appointment reminders might quietly be dismissed by your phone without you being any the wiser.
I wish Apple would force app developers to implement different "channels" for promotional notifications vs transactional - so that you can pick and choose which ones you want.
In-app notifications settings should do this if they are trustworthy.
The "you" in the title's reference to "your push notifications" is not the user, it is the marketer. That tells you everything you need to know about the value of this piece.
Not upset, but increasingly concerned that all channels are being intermediated by big tech.
Something might come of introspecting why such controls are being built and desired by consumers instead of trying to frame everything as a "big tech evil!!1" narrative.
I’d recommend following your own train of thought, why is big tech so hell bent on intermediating the experience between their users and everyone else? They’ve done it for email, web search, mobile experiences, advertising, etc.
You want to use any of those things, you’ll have to pay their toll booth, figuratively or literally.
The chain of thought is quite straightforward. Functionally nobody wants an intermediary-free channel because there are adversarial entities on the other end.
Do you think that logic holds true for web search as well? Because this is happening there in a much stronger way too.
I'm not a fan of Apple or Google, and it feels bad that all of our notifications pass through APNS or FCM. Megacorps shouldn't have control over our digital lives to the extent that they do, and anyone talking about this gets my full attention and support.
Except for marketers! I don't think there's a less sympathetic category of technologists, save for maybe CSAM peddlers.
You're upset that you can't get "visibility" into whether the bullshit ad you tried to ram down my throat landed on target? You're worried that I'm a dormant user and my phone will silently delete the spam you sent to try to hook me back into engaging with whatever worthless product you're hawking?
World's tiniest violin, buddy. Boo fucking hoo. Your last paragraph says the "senders" (read: spammers) who make it through the next decade intact will be, to lightly paraphrase, the ones who send messages the recipients actually wanted. You say that like it's a bad thing!
The computer is in my life because it is a tool that does the things I want. It is not an open mic for marketing sleazebags to try to sell me shit. May every single one of your attempts to invade my life and hijack my attention be flushed swiftly down the toilet.
If we had visibility, we would know what doesn't work and we would stop sending it. Almost like there are aligned interests there rather than a purely adversarial relationship.
Personally, all I've seen is marketers not getting the message and using it to perform A/B testing to come up with the basest ad they could possibly come up with to entrap more users.
One could take an equally uncharitable approach to what engineers and product teams do.
Yes and that would be completely correct in many cases.
> Almost like there are aligned interests there rather than a purely adversarial relationship.
You might very well be the exception, but for something like 99% of marketing content that reaches me, our interests aren't aligned. First of all, they want to generate "needs" where there weren't any before and probably shouldn't be. A pizza ad produces the wish for unhealthy food. A fashion ad produces the wish for new clothes (even though I have enough) and probably even changes the societal dynamics of individual expression and personal style to be consumption oriented.
Second, even if I have a legitimate need for a solution, they still want me to buy their product, consume their media, give my attention to them. I, on the other hand, want to be informed by a neutral third party about the pros and cons of some product. Sure you can say "but unhappy customers are bad for us", but there are actually very few niches, where this signal is powerful enough to align incentives, because information and power asymmetry limit the customers understanding of product quality and their leverage to correct harmful market dynamics.
Depends. Blackberry 10 hub was strongly designed as a shared inbox instead of a loose system of notification like ios or android.
And it was awesome.
I've definitely had notifications I consider spam direct from Google before. Apple/Google are not trustworthy.
Yeah these channels used to be respected in that way.
And then app developers discovered that hooks like "look what you missed" work on users and so now we all have to get them in the same category.
>discovery
I can’t think of a single app I want a “Discover” tab on anymore. The moment you include one is the moment there is someone trying to game it. I definitely don’t want push notifications trying to show me something new. I’m hardly lacking in distractions
(Yes I am sure somebody can give me an example of a good use of Discover but you get my point)
> (read: spam)
is it unironically incomprehensible to you that the owner of the device should in the one who gets to decide what is and isn't spam? it's not email where you can get bombarded with shit from any random server - you can mute or uninstall an app.
It should but apps don't let us decide.
An intermediate seems to be trying to fix it.
Is it ideal? No. But it's the spammers who are to blame.
You know, I would love a feature that lets me mark push notifications as spam, and optionally send them to Apple. The last part is important for a variety of reasons, one of which is that notifications can be end to end encrypted.
Spam filter push notifications.
Ideally enough spam reports for Uber Eat’s constant marketing abuse and they lose APNs access for the Bundle ID associated with the spam reports. For example.
You’re right of course, but Apple won’t do it - they’re happily running a two-tier system where Uber, eBay, Doordash can force spam notifications on you with impunity. All my settings for marketing are off - eBay still sends me notifications about coupons (and additionally there’s no way to actually contact them to complain, of course). Doordash won’t let me get delivery notifications without marketing notifications.
Apple could fully enforce their policies and fix this in a heartbeat, but they won’t.
like I said, you decide by muting or removing the offending app.
Fine, but that’s was clearly not enough to stop the spam, nor it was enough to satisfy everyone.
There are some apps I can’t afford to mute or uninstall, such as phone, transportation, communication and work. I wish I could, but I currently can’t, I’m not privileged enough.
“Punishment by Apple” in this instance is somehow the only response anyone had to misbehaving companies.
That's not entirely fair to the author as the article also states that the consumer being in control is a must:
> Every step subtracts a degree of sender control. Some of it passes to the user, and that is a good thing: a person deciding what is allowed to interrupt them is the channel working as it should. The rest passes to the platform, and that is the part that should concern a sender, because the platform's judgment is opaque, unappealable, and increasingly made by a model rather than by a setting the user chose.
A platform has essentially two "clients" - the user and the developer. Without both, it wouldn't exist. And it is in the interest of both that the platform should have very limited arbitrary authority over them. Nobody can deny today that the platform owners today have too much power over their users and their developers, which makes it easy for them to commercially exploit them while undermining their rights. We need regulations and standards (for interoperability, which is one of the arguments being made) to counter this.
Been an app developer since 2009, worked on Android for 6 years at Google. Push notifications suck, users hate them.
Simultaneously, I cannot match the pull quote, an argument summary, to their argumentation. IIUC if the reword patent / Apple’s summarizing disappear there’s 0 reason to say it wasn’t control passing purely to the consumer.
So I’m left a bit empty as the high-minded purpose has little backing, and thus comes across as bloviating.
I’m constantly amazed how passive people are with things that steal their attention
My phone is in do not disturb mode 24/7. If your app notifies me about something pointless, it gets deleted and I start using your website instead
I have a mail rule that moves any email with the word “unsubscribe” out of the inbox into its own tagged area. Every few days, I go in and unsubscribe to everything that’s arrived.
Whenever a retail point of sale worker asks for my details or phone number or asks me to sign up to their club, I ask if there’s a discount. Because if there’s no discount - they get no details. It’s a simple exchange; offer to pay a fair price for my details and I’ll consider it. But so far my time and details are worth more than any retailer has offered to pay.
That unsubscribe rule is genius. (Obvious in hindsight, as the best things are.) Thanks for that.
if you're in Gmail this filter works really well for me:
("list-unsubscribe" OR "unsubscribe" OR "list-id")
If you’re in gmail there’s literally already an interface to easily unsubscribe from stuff, it’s under “Manage subscriptions”. Yahoo similarly have a “subscriptions hub”
With the retailer asking for a phone number, I don’t see how it would ever be worth even entertaining. They will give you a discount once, then have and potentially abuse your information for years to come.
That's precisely why I don't ever accept the bribe. If I don't like the non-discounted price, then I don't buy it. Now they neither get the data nor the sale.
What's frustrating is that a lot of grocery stores do this. If you sell something absolutely necessary, such as basic foods, you should not be allowed to do the whole "mark it up to mark it down" strategy.
Also, a tip for most grocery stores (at least in the US): enter in any area code plus 867-5309. Chances are high that somebody has registered it. It's better than sharing with a family member because so many people are using it, the data becomes less useful.
Alternatively, ask the clerk to "use the store card". Usually, they will oblige.
You can make an email rule to filter those messages to trash.
I have my phone set to only ring for people in my address book. It’s probably time to do something similar for email. Not in my address book? Straight to trash.
Adding a filter is still extra work for me. Saying “no” stops the need for the rule and is less work than giving the info in the first place.
My phone is setup similarly. I did it manually back in the day, then sent some feedback to Apple, which they added in the next update about a year later. I’ve submitted a lot of feature requests, this was the only one they actually did, which is a great one. It made things much easier. Though they seemed to have changed the settings of how this works with the call screening feature. I used to have a shortcut to toggle this off and on, when I was expecting a call from an unknown number. I need to revisit my setup here, as the shortcut doesn’t actually do anything anymore.
Doing this to email is an interesting idea. If sitting in one ecosystem, maybe it would work. I’m fractured, so it’s a non-starter. Even beyond that, I think it would be an issue as there are real emails I do want to get which I’d never add to my address book, as I’d never send an email to the address. I think I’d want a whitelist for these addresses, that imported the emails from my address book as a base.
At work I had a rule like this for many year. Anything internal would pass, plus a few external domains I named. Everything else would go to a spam folder for vendor emails. Keeping up on this was hard. I ended up throwing in the towel a couple years ago after running the rule for 5-10 years. This blog post is what made me move away from this rule[0].
[0] https://herman.bearblog.dev/digital-hygiene-emails/
I'm sorry to burst your bubble but I don't think your feature request had anything to do with when apple implemented call allow listing. It's been a thing forever and they were surly aware of it since the iPhone first came out
Yes - and a fair price would incorporate that.
Hence I doubt retailers will ever consider offering a fair price.
I made a more advanced version of the unsubscribe thing to kill all unnecessary notifications from email. I made it open-sourced:
https://unfuck.email
> passive
I get your point and see it as valid, yet to nitpick most people don't feel they have a choice.
Not answering the phone or replying to people's messages is a no-no to many, which sets them in an arms race against spammers and social apps trying to get them from all fronts. And they get frustrated by us living in no-disturb land 24/7.
I don't know it could be solved, but I feel for them.
For sure - some folks have to unwind expectations they’ve let themselves be under around responsiveness to messages.
Feels like an education issue rather than a tech issue.. thoughts?
I think it’s cultural. Someone sends you a message and knows you likely also have your phone on you. If you don’t respond they might feel ignored.
It can be hard to set phone boundaries with work. There are certain people who get very upset if I don’t answer their phone call.
Turn off all notifications except messaging and see how your day goes. It's not going to kill you. You quickly get used to regularly checking things you actually care about, and the rest has to wait until YOU care.
I've been doing this for many years, and none of my friends or colleagues are aware of it, and they don't need to be. Notifications don't help you respond quickly, they just grab your attention from things YOU wanted to do.
I haven't checked Discord today yet. I haven't checked my email. Whenever I do want to know if my friends wrote me, or if I have some new bills, or if I need to follow up on something, I will open the respective app and deal with it.
I can put my phone next to me for hours and not get distracted.
> You quickly get used to regularly checking things you actually care about, and the rest has to wait until YOU care.
This was the biggest thing for me. Before, I was paranoid that if I turned off notifications, I'd miss something important. As though I didn't miss notifications anyway.
Getting used to regularly checking (important) things also has two wonderful side effects (at least it did for me):
* My "mental notification system" got better. Because I was less dependent on my phone doing it for me, I developed the skill more on my own. * The apps and services that I checked less and less frequently became more obvious in how unimportant they were to me altogether. I have far fewer apps and accounts now, making me MORE punctual overall.
This part is factually incorrect:
"...a notification lives only in the notification centre, which clears, drops and summarises what passes through it and retains nothing reliably."
Your notification center reliably retains information. Something like an inbox does exist, just not in userland: https://www.forbes.com/sites/larsdaniel/2026/04/10/fbi-pulle...
I'm building a productivity app right now that lives entirely on the web right now because I can't justify native development costs yet. The daily nudge (i.e., you have 47 free minutes, and this task has been sitting for 14 days) is the core retention metric. This only works as a push notification; emails get ignored. The web app sends an email. V2 goes native specifically for this reason.
The article's point about leading with the fact rather than the brand point is useful. The notification payload I'm designing already leads with the concrete numbers.
> Over fifteen years the channel has been rebuilt around one assumption: the receiver's attention is a scarce resource the platform is obliged to defend. … As a sender you are on the wrong side of that assumption, whichever way the control moved.
Fascinating how the author openly frames the situation as the sender and receiver’s interests being opposed.
They are not necessarily opposed, just in tension.
A zealous guard of your attention will occasionally block something you would like to have seen.
That being said, yes most notifications are garbage and should be blocked.
A fairly uncharitable read, I’d argue it states that the platform is acting on the platform’s interest, not the user’s.
Speaking from the standpoint of a user: I consider my attention a scarce resource that needs defending.
To the extent a platform has the same assumption, its interests are aligned with mine.
To the extent a sender does not have this assumption, I want the platform to defend my attention on my behalf.
Whatever has replaced the Bulk Collection of Telephony Metadata Under Section 215 of the USA PATRIOT Act informs the architecture of the Apple Push Notification, Firebase Cloud Messaging, etc. Apple owns the persistent connection to every iPhone, and only APNs can wake your app. So "self-hosting" here means running your own provider (the backend that decides what to send and hands it to APNs) instead of paying a third party like Firebase Cloud Messaging, OneSignal, or Pusher to do that for you. The last mile is never yours however. Any architecture that routes everyone's traffic through a small number of identity-aware intermediaries is, by construction, a bulk-metadata collection system waiting for a legal instrument.
[2] In December 2023 Senator Ron Wyden disclosed that the U.S. government and foreign governments had secretly compelled Google and Apple to turn over information from push notifications, including communications metadata and sometimes content. A detail that should bother any developer: app developers have no way to stop the practice if they want to send notifications on the platforms iPhones and Android rely on. Apple had been gagged from disclosing it until the program became public, after which it said it was updating its transparency reporting to detail these kinds of requests. So the architectural hypothesis isn't speculative — it's the confirmed mechanism, differing from Section 215 mainly in domain (apps vs. calls) and legal vehicle (ordinary subpoenas, FISA orders, and NSLs rather than the specific business-records theory of §215)
[1] "Its just metadata". Thanks Obama! (joking of course, no single individual is responsible for these things, it is our collective political will and its the best we can do unfortunately)
[1] https://www.youtube.com/watch?v=9iUdm0QMDM0 [2] https://epic.org/sen-wyden-reveals-government-surveillance-o...
> For most of the channel's history they did very little of it visibly. The architecture was permissive of intervention; they simply chose not to intervene much. That restraint is what ended.
I guess it wasn't always visible, but they were intervening in some for or another since the beginning. At WhatsApp, push delay/suppression/coalescing was something we were always monitoring, and IIRC, it was part of the system since at least when I joined in 2011. If you don't work within the system, your users' messages don't get delivered timely.
Huh that’s interesting, do you have any further context on that? I’ve not worked on a product with anywhere near that scale before so monitoring has always been whatever I can get from commercial push platforms
I mean... record the time we first send a push message, when a client connects have it tell you if it's because it got a push or user interactive start, check the time between push and connection, add that to your choice of time series graphing tool. Graph by platform, and you can see when the platforms are delaying pushes.
Some of the delay will be ordinary things like their push service fell over or is unreliable (you also get some feedback when they don't accept push messages), or their push connection runs into silent NAT timeouts on some networks. But some of it will be things like you ran into an undocumented push quota, so Blackberry users don't get timely pushes at peak, etc. On client platforms where you have reliable background execution with network connectivity, you can potentially signal connecting clients if platform push isn't working well and have them switch to persistent connections until the push service comes back. But that was never an option for iOS; it hasn't been a reasonable option for Android since at least Android 6 when Doze was introduced... and app killers before then made it hard before then; and all the other platforms are dead. Now, push really just has to work.
AFAIK, Apple has always been willing to deprioritize pushes when you send "too many", especially when there's no user interaction; or when they added silent (voip) pushes to wake up the app, they only let you have a few silent pushes if you don't post a user visible push.
For ordinary async messaging, push latency doesn't become a big deal until it hits double digit seconds. For voice/video calls, you really want pushes to be as near to real time as possible, or the caller is gone before the callee phone rings.
Hmm, there's just a few big messaging apps and just a few os vendors, weird that you couldn't have established special treatment for pushes about user to user messages
> None of this bites evenly. The editing falls hardest on broadcast and promotional push; the notifications people actually want tend to pass through untouched or amplified.
Sounds fine with me?
While I have slight worries about what it means for users if Apple and Google notification services go down/censoring, I do appreciate the features that they provide to me as an end user.
So many apps use annoying and questionable marketing notifications that I'd say I have about 70% of app notifications disabled globally (because the app itself does not allow disabling notifications / has no granular control).
However, it seems that SOME self hosted services can directly notify you without APNS / FCM. As an example, see https://companion.home-assistant.io/docs/notifications/notif...
Am I understanding correctly that iOS notifications have to go over apns unless on the same local network as the HA server? I do appreciate that android makes this possible for ha and signal (and others) in all cases, it should be up to the user to choose centralizing the connection vs. slightly worse battery life.
As far as I understand it, yes, you have to be connected locally.
The question is, will it work over Tailscale/WireGuard? I'd assume so, but I've honestly never tested local pushes as I simply haven't needed them :)
I wish apple/google would implement better notification control - like the ability to turn off all marketing notifications, and a much better digest format
Notification Channels is the official way to do this on Android, but it's up to the app developer to categorize them properly. They have no incentive to allow you to turn off ads.
Actually, they do have an incentive to let you turn off ads. If they don't, many users will turn off notifications entirely. At least if they categorize them, some users will just turn off the bothersome ones.
I'd say most people don't though, and those people might be more influential to advertisements anyways so it's a net win for them.
If you're on Android, I'll always recommend Buzzkill to add very granular rules for notification filtering. I set up all kinds of filters just for the Amazon app.
On iOS I assume you're sol, that notification system is unhinged to my eyes.
I just turn off all notifications for any app that sends me marketing notifications.
You are less charitable than me. Maybe I'll adopt your approach. I first give an app the benefit of the doubt and go into the apps notification preferences and see if I can fine-tune their notifications. If not, off for all at the OS-level. If yes, I tweak it, but if I get surprised by one later, off for all at the OS-level or uninstall. It's especially annoying because I don't have notifications shown on my home screen and need to unlock with a pin so if I go through the trouble of unlocking my phone to spam and I extra annoyed with the app.
That would be nice. I wouldn’t be surprised, as on-phone models get more capable, if we don’t see them start to build an “inbox” like we see with email where you can then start seeing much more heavy processing happening.
I think that's what the Notification Organizer on Android (maybe Pixel exclusive, not sure) does. It's sorting notifications into broad categories using AI and groups them in the notification shade.
Makes sense, Google definitely have a lot more experience in that space with gmail than Apple do.
Too bad about the walled garden or you'd have this tweak already installed years ago.
Personally, I don't see a few permanent connections as a problem. My GrapheneOS phone is degoogled, and therefore apps such as Signal fall back to a WebSocket connection. Battery life is probably somewhat impacted, but I use too few apps to notice. And in any case, this is much better than allowing Google to stick its nose into my business.
Yeah I'm disappointed this isn't pointed out in the opening paragraph. It's fair to critique Google for convincing devs that fcm is the only option, and obviously iOS is designed for Apple to do whatever they want, regardless of the owner's wishes, but Android does have other, viable, options. iOS and Android aren't equally bad here.
If my phone buzzes and I look at it and the reason was dumb, I delete the offending app and leave a 1 star review. I don’t know which of these steps are loadbearing, but my phone has gotten much quieter.
The entire architecture here is surprising to me.
>an iPhone could not afford to let every installed application maintain its own background poll against a remote server. The proposal...a single persistent TLS connection from each device to Apple, over which any registered third party could deliver alerts.
I thought apps were sending notifications locally in the device. Like, if a messaging app receives a message, there's a network call for that. Then if the messaging app wants to tell the user they received a message, it can just hit a local API for that, right?
Is the pattern actually that the app makes another network call to the notification service to register the notification, which makes another network call to the device to deliver it?
yeah this is what author hints at with "Push as a battery problem". Apps are limited by default in what they can do in the background due to this, so most apps are in a suspended state not making network calls when you are not using them. To avoid the app having to keep running this stuff is delegated to OS which tells the app, "hey I have a push for you wake up and handle it!" You can send pushes locally but because of the background limitation it is not practical for unpredictable events like messages coming in.
What you're describing seems reasonable, but it doesn't align with what I quoted, unless I'm missing something, which I very well could be.
Having apps sleep and a daemon wake them to handle notifications doesn't require all of the notifications coming from Apple.
Not sure i understand. Sure it doesn’t require it but Apple doesn’t trust you to handle it because you’ll probably drain the battery or spam the user or whatever, that is Apple MO, putting themselves in the middle so they can control the experience.
The single persistent connection is just to receive pushes, there is still some daemon controlled by apple in charged of dispatching to correct app.
Marketing and advertising people ruin everything they touch.
>"But salesmen exist because customers don't actually know what they want"
"I want item ABC rev D with options X, Y, & Z – I do not wish to pay for undercoating"
>[salesperson looks at autist dumbfoundedly] "I'm not sure we can do that; don't even know what Y & Z are..."
"It's a valid configuration, please just do this for me."
----
Why does my state still require new car sales via a dealership model?!
Don’t forget the engineers and product folks who build these systems.
I see the point. But honestly I am more concerned about having to constantly fight to turn off all permission allowances every time I install an app.
And the moment I have some faith and trust an app that I deem important, I get promotional junk as a "notification".
I would really like to have notifications allowed on certain apps like parking, or health etc., but all they seem to do is abuse the trust they are given, meaning I turn them off.
So where I agree with this author is certainly that more power belongs at the user.
Apart from this, what is most needed in both platforms is an application firewall - not every app needs to be allowed to connect to the internet.
I can't believe this still isn't a thing outside of GrapheneOS. Being able to revoke network permissions is a fundamental security and privacy tool that's willfully left out of both Android and iOS.
There's zero reason not to include it as a toggle.
On iOS it wouldn’t even be that hard. There’s already a toggle to disable use of cellular connectivity. Add a separate one for non-cellular (iPadOS can connect via Ethernet), and/or a “disallow all” toggle.
We are partly there in spirit with App Transparency keeping track of the IPs and hostnames apps connect to.
Apparently chinese versions of ios (specifically for China) already have this feature because the Chinese government mandates it!
"having to constantly fight to turn off all permission allowances every time I install an app"
Are you really installing that many apps that this is so hard?
> certain apps like parking, or health etc., but all they seem to do is abuse the trust they are given, meaning I turn them off.
I've found that live activities on iOS helps with this quite a bit. Let's me keep notifications disabled on parking apps and DoorDash while still getting the tracking info I want in the live activity & dynamic island.
Otherwise, yeah, you just can't trust anyone to be respectful with notifications. Phone & a messages whitelist via focus modes are the only notifications I allow on my phone.
IMO they should be doing way more to control push notifications, there's so much more control they could give the user, and many clear violations of their policies.
One of the best apps I've bought for android is buzz kill which lets you set rules around notifications. I have cool downs on family chats and social media so it doesn't keep buzzing when things kick off, filter Amazon alerts to only "we're two stops away" and "We've delivered" messages and dismiss the rest.
I have custom buzz patterns and sounds for urgent alerts and rules that batch notifications depending what WiFi I'm on, time outs on things that don't matter after a few hours etc.
My notifications list is now way smaller and far more relevant.
Also quickest way to sort out notifications is to take your phone off silent. Hearing everything coming in, you see more when it you can then decide if the notification should make noise, or exist at all on a per app basis.
The first thing I do on every new phone is to turn off 99% of notifications. Messaging ones and email are first to go. I cannot stand the constant beep-beep.
“ None of this bites evenly. The editing falls hardest on broadcast and promotional push; the notifications people actually want tend to pass through untouched or amplified”
So … mission accomplished then? This is pretty much how I would like it to operate.
That's my thinking too. Do marketers not also have phones, or do they somehow not see how annoying promotions and spam are? There's already so much paid advertising everywhere, and the free advertising with direct notification to the target is obviously going to be abused by these exact types of people. Guardrails must be put up by the transport layer (Google, Apple) so their own image isn't lowered to the image of the abusers that are too keen on promoting their disruptions. Few people want a convenient device that makes life inconvenient[1]
[1] I know a few people whose phones constantly beep and flash numerous times a minute, and when on, the top is completely unusable because any notification dismissed is immediately replaced, obscuring those upper buttons again. I don't understand how they tolerate it.
Isn't this a strategic going for years now, throwing random notifications to make the user use more the app? I for once, block notifications from apps I don't want, because I don't like getting bombed with stupid ads etc.
Am I supposed to feel sorry for developers? How did this make it to page one?
Seeing how they think about these things can be interesting.
I dont uninstall apps that annoy me with notifications, but I do disable them. Most of my notifications these days are news or texts. So be it
Smart phones should absolutely allow you to lie to apps about what permissions they have, to feed apps fake data, and to basically control every single thing the app sees about the phone and the user.
The fact that they do not do this reveals that phone makers are sort of market makers between app makers and customers, creating an environment which, in a certain sense, is a neutral ground between these two types of "users."
> that answers to the user rather than to you. You cannot out-shout it, and there is no appeal.
"now here's a list of how to get around that!"
I don't think I've got a push notification in awhile. Few months ago I switched to Lineageos and started using the web browser instead of apps. It's peaceful.
I still get notifications (SMS, email, calendar, etc) but nothing pushed
> Google followed in 2010 with Cloud to Device Messaging, then Google Cloud Messaging in 2012, then Firebase Cloud Messaging in 2016
Classic
I’m curious if anything meaningful changed along with these name changes or it’s mostly issues of branding/implementation.
The real solution is to allow users to own their push solution, and for it to become more commonplace among apps to support alternative push providers such as Unified Push. Molly, the FOSS Android Signal client supports this configuration.
My Android phone has a long list of toggles under notification for each app. I am genuinely interested to make good use of that to optimise what notifications I recieve, but I am clueless about what each of them means . For example: badges, floating notifications, permanent notifications, unified support, alerts, etc.
Push notifications are for the user, not the marketer.
From the author's blog: "I do Revenue Operation, helping Marketing, Sales and Customer Success teams with data, process and technology."
You think there might be some sort of interaction between both facets there?
How is bad summarisation good for a user, for example?
Probably depends on the user. Along with push notifications for almost every app on every one of my devices, I disable the summarization.
For me the notification is the point, and the point of notifications to me is that they deserve my attention. Of the vanishingly few apps I install these days, almost nothing can say it deserves my attention. Even my bank doesn’t get those privileges.
I largely do the same, and keep my phone on dnd mode.
> You think there might be some sort of interaction between both facets there?
With the exception of one trying to extract currency from the other, in exchange for something of dubious value—no.
The interaction is: marketers keep trying to abuse these systems and platform owners and users keep having to find ways to fight them off. Some of these efforts have unfortunate downstream consequences (like bad summaries).
There are apps that need 100% notification delivery reliability (in a sense that OS or delivery server itself will not be allowed to decide to just drop PN for policy reasons). I guess Google can only "solve" this (by simply not passing them through any "classification") for their own apps only?
For some android phone brands delivery reliability is like 40-50%. Some brands are better (reaching 80%, still bad) some very bad (usually the chinese brands, for some reason).
And the user has no say in this. They can't say: deliver every notification without classification. Or "allow this app to wakeup whenever it wants". Everything is babysat by the great overseer, even if you write the app yourself for yourself.
Have fun writing a telephony app that receives only 50% of calls. :D
Vast majority of software should not be able to send a push notification. Send an email if you need to alert me.
Massively overlong article that really could have done with an editor. Although obviously editors cost money, and I'm reading it for free, so I can scarcely complain. Nevertheless, some concision would have been appreciated.
I'm very unclear to me what the thesis of the article actually is. Yes, push notifications run through the vendor's servers. Yes, Apple fucked up hard by modifying the text within them - and I contend that such modification is impossible to perform automatically without unreliability becoming the norm.
The author also appears to believe that "broadcast copy" - otherwise known as Spam by those who like to write slightly more honestly - is a legitimate use of push notifications. It is manifestly not, and any app that tries will at the very least be immediately silenced. I wish I could find the tweet that put this sentiment more entertainingly than I ever could.
If App developers continue to abuse the push notification system in this way, Apple and Google will be forced to take steps to solve what becomes an end-user's problem. Yet another tragedy of the commons.
It’s a through line from an article I posted last week about the similar situation in email, which has a lot more depth as inbox providers have substantially more published papers and patents showing their intermediation.
https://www.jacquescorbytuech.com/writing/what-google-yahoo-...
The next post will be highlighting the difference between the actual state of the art techniques being deployed by large tech co’s (LinkedIn and Pinterest, for example) vs what’s available via commercial marketing providers and how most marketers don’t even make the most of the tools they pay for.
> The author also appears to believe that "broadcast copy" - otherwise known as Spam by those who like to write slightly more honestly - is a legitimate use of push notifications. It is manifestly not, and any app that tries will at the very least be immediately silenced.
Cool man, but it might surprise you to find out that people knowingly opt into receiving this stuff and actually consent to it.
> Cool man, but it might surprise you to find out that people knowingly opt into receiving this stuff and actually consent to it.
Consent is more than pressing 'Allow' on a notification pop-up. It's often not even informed consent, as those pop-ups are usually a part of some onboarding flow where users are just trying to get to the value the app promises and pressing 'ok' to everything.
Even if people do indeed want notifications at the time of the ask, one doesn't really know if the message provider will wind up spamming, that's a matter of trust. And once opted-in, even if the users no longer want notifications, a lot just don't know how to turn them off. People are often incredibly accepting of sub-par experiences like this because of the friction and capability demanded of them to opt-out. My parents get tons of spam notifications that would pass your test of 'knowingly opt into receiving' but that when asked they say they do not want.
Finally there's myriad dark patterns that tons of apps use, like changing and resetting notification preferences among others.
I'd hazard a guess that observed opt-in rates far exceed users actual desires, so I wouldn't put much stock in them. I do agree that there are some people that like them tho!
Fwiw I've worked on both the delivery side (OneSignal) and developer side (Margins) so I've lived these choices and trade-offs. My believe is in terms of power dynamics, senders generally don't deserve their power to interrupt and should not possess that power. At best, they offer opportunities, which ideally are verified somehow prior to being presented to users. I'm happy that devices and ecosystems are moving in the direction of triaging and filtering sender content, as power needs to lie in the user's holistic, most pre-frontal cortex driven expression of their desired experience, and not just one moment's opt-in button they pressed.
Thank you for writing the article, good discussion points.
Yeah, that's true about the allow, and for sure marketing and product teams are deploying misleading consent priming which doesn't fully explain to the user what they're actually allowing in the first place, or setting baselines that are too permissive vs what the user is expecting.
> I'm happy that devices and ecosystems are moving in the direction of triaging and filtering sender content, as power needs to lie in the user's holistic
I don't disagree necessarily, but I see it as them putting themselves in a position to act as a toll collector, which has already happened with email and web search and is only getting worse with the introduction of LLM's into both of those things.
It's a bummer this article ended up doing much better than my email one, as I think that might better position the problem in a lot of user's minds and highlight just how much surveillance is sitting on top of those free inboxes.
For whatever reason, I get very few push notifications on my phone. Compared to my days at Blackberry, it's probably 10% as frequent that I get interrupted by my phone.
So good for me.
But there's some really scary stuff in here happening to other people that I'm not even aware of.
This article separately needs a summary at top.
Not everything needs to be summarised
The default must be pull, unless opt in for push. At the moment I would like notifications once a day or once a week for most apps. But instead I ha e turned it off completely, because of the push abuse. If I can configure to pull all the notifications on a predetermined cycle, it makes my life even better
> What the marketer can see
> Your visibility into all of this is poor by design, and getting worse.
Great! This article is all good news, it seems.
What I'm doing: leaving them all off
Android is better because they allow you to change individual notifications right from the notifications themselves + granularly do it there also.
On iOS I have to find the right setting page and then all notifications are either on or off. Doesn’t make sense.
You can long press on the notification (or swipe left?) and pick "Turn Off..." among other options
https://support.apple.com/en-us/108781#manage-alerts
turn off is the only option basically, try an android phone bro
This only works if the app properly tags notification categories, no?
It also shows what category the notification was tagged with. An app that improperly tags notification categories gets one of two results from me:
1. Uninstall the app
2. If the app is non-optional for some reason, block all notifications.
I feel like the CAN-SPAM Act should apply to push notifications as well. I don’t know of any case that has tested this.
I've had all notifications turned off expect for my immediate family (parents, wife) for years now. I'd get rid of my phone before going back to getting buzzed and dinged constantly.
iOS really needs LLM-based notification filtering. This would take care of promotional notification spam overnight. It would even enable fine-grained user filtering like "notify when - someone is messaging me about plans for today."
Because Apple makes iOS, they could have a much more rigorous solution, like Google’s.
A nondeterministic, energy hungry classifier is inferior to writing a policy to define channels.
You really need both.
Channels are a great first level, and iOS absolutely needs to implement an Android-tier version of this.
But channels continue to be abused, even on Android. When all deterministic controls fail...
Secondly, channels are set by the developer (or platform). In an ideal world, I want to define whatever channels I care about, and turn them on/off at will.
I need a feature to block my bank's incessant nagging about cash-back deals while keeping the ones about transactions.
Right now on iOS there is no way to do this. And yes, I've explicitly turned off the cash-back deals notifications in my bank app's settings and that is completely ignored.
I don't know if there's an iOS equivalent, but Buzzkill on Android is really great for this. I set up filters to hide all the stupid Amazon ad notifications.
let me pour one out for all my homies in the marketing department...
> 2 to 5 notifications per week is the optimal range for most apps and exceeding it materially increases uninstalls;
Wow. Y’all must be much more tolerant of your time being wasted than i am. One notification from an app I didn’t need/request/expect is cause for deletion. 2-5 per week would be enough to go and rate the app 1/5 on the AppStore and actively recommend everyone I know to delete the app.
> visibility into all of this is poor by design, and getting worse.
Good! I pay Apple big money to protect me (user) from you (abusive app developer, abusive by definition since you talk about my attention as if it were your property)
That's why my next phone will neither be Android or iPhone.
I'm surprised that the article is this long with zero mention of Senator Wyden's concerns vis-a-vis Google and Apple's Push Notification system: https://www.wyden.senate.gov/imo/media/doc/wyden_smartphone_...
I’m in the UK so I don’t catch all us news, good spot though
Also can the browsers finally acknowledge that allowing websites to request the ability to send desktop notifications was a terrible mistake?
This sure sounds like a marketer spending far too many words crying that they've lost surveillance on their customers. Boo hoo, don't care.
How does this not mention alternatives? Here: UnifiedPush.org
And the author is also wrong that all notifications on my phone go via Google. Signal and Mastodon notifications are set up via Sunup.
They seem to have given up. Don't do that please...
Predictably people are moving back to SMS for notifications. Not as nice for linking to your app but once the user opts in you don't have to deal with the Apple/Google complexity.
SMS will go exactly the same direction as push as far as being intermediated.
I live a push-notification free life, aka no push ever.
I also saw elderly people receiving such notifications and not knowing why. Then I realised that this system is designed to abuse the elderly, so I am now totally against it.
Google/Outlook/etc intervening with email was a good thing and saved email with spam filtering and content ranking. Mobile Carriers have not effectively intervened with phone screening and voice calls are practically dead.
Intervening with push notifs could be a good thing. Notifs are approaching the point of uselessness. I turn all off by default now.
What’s fun is all that “intervening” infrastructure they’ve introduced also doubles up as a big surveillance network.
Oh man... I just wish someone invented some form of organization where workers could negotiate with employers in a more equal footing by doing this in a collective way.
[flagged]
We've banned this account.
We detached this comment from https://news.ycombinator.com/item?id=48301060 and marked it off topic.