This was a performance driven change. We added this as loading a cross repo issue is a much slower experience than loading an issue in the same repo due to the way the header is loaded (which is being worked on).
But we hear you on the feedback - we will roll this back while we keep pushing on the root performance causes.
[update - this change has been reverted and the previous behaviour is back]
Some hard numbers [1] as to why GitHub is struggling with stability issues, directly from GitHub's COO:
Yup, platform activity is surging. There were 1 billion commits in 2025. Now, it's 275 million per week, on pace for 14 billion this year if growth remains linear (spoiler: it won't.)
GitHub Actions has grown from 500M minutes/week in 2023 to 1B minutes/week in 2025, and now 2.1B minutes so far this week.
So we're pushing incredibly hard on more CPUs, scaling services, and strengthening GitHub’s core features.
From what I remember, it got much worse the moment they started requiring JS for displaying what would otherwise be mostly static (and thus easily cached) content.
> loading a cross repo issue is a much slower experience
Why not solve the real problem instead of putting in a janky workaround?
At risk of being cliche, it seems like you guys could benefit from the 5 Whys approach here: "Why is loading a cross repo issue slow?" and iterate until you discover the root cause, and fix that.
I suspect fixing the root cause is going to be a lot less glorious career-wise than implementing a UX change that is easier to tout at review time (well maybe not so much after this debacle).
To be honest GitHub should have like a switch for "preview stuff adopter" where you guys could give any benefits for it (maybe more copilot usage?). This way you can test with a specific public, using metrics and feedback, while testing and people could comment more about it.
I would like this personally as I hate change in general, but from their perspective it's not a great test because the sample is far from random. They should still do it though
Can you elaborate? The header meaning the top part of the page? I just checked on a recent repo I visited and it has the usual banner (which would stay the same), the repo path, some links, and some stats. Considering every page navigation would likely pull which links and stats are shown, why is this a delta to go to another repo and why are presumably 3 database entries (possible links, stars, forks) so slow?
which is also driving me nuts because it frequently fails to update the issue and PR counts when I close issues or PRs. Only a hard reload, or closing the tab and opening a new new one, fixes it.
Yeah over the past six months I've trained myself to just hit Command-R every time I switch back to a GitHub issue tab, otherwise things get stale or broken far too often.
I can't speak for GitHub but I've worked on multiple nav headers for large SaaS products and they can be ridiculously heavy weight to render given they appear on every page. They tend to be a dumping ground for features, many of which require their own permissions checks, feature flag checks, etc. it's not unusual to have to perform hierarchical permissions checks. They also tend to contain contextual info about the current nav state and dynamic information about navigable states.
A lot of this can be cached but it's easy to see why moving from one repo to another will invalidate most or all permission checks and feature flag checks.
How many checks are we talking? A well-implemented monotonic system should be able to do tens of thousands of these checks (or more) in the time budget I associate with a heavy page, and start before the entirety of the permissions/feature data is available.
Yes, pretty much this as well as some additional complexities due to the issue content being in React and the header in Rails - to the cost of approx 500-800ms p50 for a page load vs sub 100ms for a nav to an issue in the same repo (or without the header which is what we tried with this change here)
It's interesting to see that the UX issues that are annoying me when using Azure DevOps are finding their way into GitHub.
In case they are truly chasing Azure DevOps level UX, I would recommend they implement an HTML editor for issues that, depending on whether the user has dark mode or light mode enabled, saves some CSS of the respective mode and makes it unreadable if read within the other mode.
It’s always been interesting to me that multi-million and even billion dollar tech companies don’t have perfect websites in terms of UX.
Just last night I was helping my GF set up an ad for her job on LinkedIn. The UX was terrible. Like awful and basic things like save and exit were completely broken. Meanwhile LinkedIn makes what percentage of their revenue through ads? Same with google ads. It’s like these products that are in a way some of the most valuable products in the planet, are given a junior web dev and a “UX designer” who really doesn’t know anything about UX.
I don't think there is such a thing as perfect UX and I'm not asking for it. I just want them to stop making it worse.
Seriously tho, why isn't this something that a browser can do? Why can't I just split a tab and say all links from the left tab open in the right? Why not be able to scroll through history as a list of such panes like a smalltalk browser or file explorer on a mac? Maybe even a history tree, able to be forked with a click or two. Tree-style tabs are a baby step toward that, but I'm not seeing much interest out there in actually learning how to run.
KDE's hybrid file / web browser konqueror has had arbitrary tab tiling since 1999 IIRC.. still a gread tool, would just need some love and webextensions support to come back big
i occasionally need to compare two tabs. previously that meant that i had to open those two tabs in separate windows and then use window tiling to place them side by side. setting that up was a lot of work. and also it makes switching windows very hard. each side by side view would add two more windows that all need to be cycled through when i switch windows. and don't try to have more than two of those on a workspace. you'll go crazy switching between them.
with the split view it not only becomes very easy, but the split tabs also keep their position among all the other tabs, so i can keep the view permanently without cluttering up my list of windows. currently i have 5 split views in active use. that number is likely to grow...
I think it’s a nice feature. I use it to have designs on one part of the screen and implementation on the other. That way I can jump between “designs | implementation” and “PR | swagger” without managing and resizing tabs. Previously I had to jump between tabs and taking into account the newer screens provide a considerable amount of UI real estate there was screen area to utilize.
It's precisely because they're so big that they can afford to overhire lots of designers, which then obviously need to justify their employment by continually changing things. This isn't a problem with small and tiny companies where "UX designer" might not even be a separate job but the responsibility of someone who will care only enough to make something that works and then leave well enough alone.
i have a really good friend who did the whole UI/UX design bootcamp during the explosion of UX/UI jobs. he did okay, he's probably hopped jobs 2-3 times now and is now without a job.
i actually feel for him, it's definitely one of the career paths that's looked at as excess/waste now while companies slim up to reappropriate money for AI. but i do think there was something there, he was genuinely passionate about what he did and it's just really hard to find work doing it now.
i feel guilty saying this but i've let him talk me through some of what he does, show me how he sees and approaches design (the bulk of what he did was design the interfaces for publicly used webapps and mobile apps) and... idk. i feel like it's all acquired taste and almost a "good app developer will think of these things when they design the front end" and a lot of his insight to me broadly looked like a lot of stuff i would've considered myself as a mere sidequest and my general thought process to deliver a good app. the difference is im building the app and designing the user experience, but his entire career is silo'd to just building the user experience.
im not against breaking out the design to a dedicated resource whether thats one designer or one team who wants to try and maintain a consistent language for a company. i think this has upside to make the design experience not locked to a single developer or developer team, and opens it up to a lot more channels of input. but on the other hand, like it's not the end of the world for me as a developer to come up with a really good design & i personally have never imagined myself not considering UX/UI at every corner when I'm building something. It feels like a second nature to me, there's creative aha moments to it, i think it's generally really good for a developer to step into a users shoes and almost "debug" the experience.
where i think ui/ux has gone off the rails:
- i think it's unduly influenced web design and has been poisoned by marketing. the rise of landing pages for SaaS that say a whole lot of fucking nothing and the crossover with "marketing research". i actually literally can't stand these types of pages, i swear 75% of the time i click around and can never get a straight answer on what the product/service is. examples: https://boomi.comhttps://www.astronomer.io
- things like OP, issue links opening in popups. changing things for the sake of changing things. such a change is probably "backed by research / surveys" giving the illusion that this was a data driven-decision, making it hard to push back on.... despite on deployment = everyone universally hating it. there seems to be some heavy flaws with the data sampling/collection methods that drive these decisions. i think the field of ux/ui as its own distinguished and defined field needs to undergo a self-awareness evolution here. something that's happened quite a few times in engineering. they really need to scoot back and have one of those "sometimes the best path forward is to not change anything at all" moments collectively and learn to recognize when that is right in front of you
- sometimes (maybe more than sometimes) allowing the business to dictate design is mayhaps not a good thing. i think what im trying to say here is the existence of "hes the ux/ui guy in the department, go talk to him" gives business stakeholders misaligned incentives to just go and push a change that isn't _actually_ user oriented, but is heavily tied into some metric or some other stupid business initiative. actually the more that i think about it this is probably why a lot ui/ux careers exist (give all control of the design over to the business) and that seems like a slippery slope
It gets increasingly difficult to design a website properly when you have different teams with different goals each competing to put their little feature front-and-centre, leading to a hacky job on top of a hacky job on top of a hacky job, which in turn hurts the performance until one day someone finally decides to re-think the whole thing from scratch and pisses off >50% of its users in the process that are used to the mess.
It's way easier to nail the UX when you're still in the dozens-of-employees stage of growth and offer like five features in total.
> It’s always been interesting to me that multi-million and even billion dollar tech companies don’t have perfect websites in terms of UX.
This, but for online shops, especially clothing. Horrendously buggy, laggy, with broken navigation (especially when navigating back), filters that don't work on > 95% of online stores. Why they wouldn't fix their primary (or at the very least highest margin) income stream is beyond me, but I've had to abandon so many shopping carts just because the checkout flow is literally broken.
As someone who has built a lot of greenfield UIs while also maintaining old ones (13+ years old SaaS), I recently set up LinkedIn ads and realized the UX is abysmal considering it’s something they’re actually trying to make money from. Maybe—just maybe-I’ve seen such poor UX in a free web app that lacks a maintenance budget. The only reasonable explanation I can come up with is they have a lot of silos within the ad portion of their platform, and each team works on their little corner and no one tries to work with it end to end. Since it’s LinkedIn, this is inexcusable. You go and try to make an ad campaign and then an ad set within it containing some ads, and then come back to it a week later and try to find all these entities you created. You may land on one and take a very long time gritting your teeth and praying for a way click around until you can find another one. What‘s the net drain on worldwide GDP caused by the time-wasting UX of this component of LinkedIn?!
I agree, and I think the metrification of UX hasn't helped here.
If you read the old Win32 interface design studies, and Raymond Chen's "Old New Thing, The: Practical Development Throughout the Evolution of Windows" you realize what people click isn't always what they want.
And old UX was ensuring that it was build in a way that what the user clicked was what they wanted.
Now? Since the MBAs came in the UX is another hostile piece of software, trying to trigger you into spending money.
I visit a site/launch the app I always use with the intent of getting something done quickly, and I find that since the last time I used it someone's rearranged the deck chairs and hidden or removed the functionality I need. Something that should take a minute or two suddenly becomes rage-inducing and eats an entire day.
Or the feature is still there but they've renamed it to something totally unrelated which you would never guess. Honestly, it's like they are actively trying to lose users.
The most depressing email to receive is "Good news! We've improved our website ..."
The Win8 and Metro design disaster is what happens when you give UX free rein, instead of focusing on users they try to start design trends to impress other UX / designers (essential for their career).
I wonder how much of Apples design was basically ‘if you confuse Steve Jobs you’re fired.’ And this acted as a necessary governing force to counteract the need to impress peers.
Metro was a wonderful design for the media player app it was made for. It's great for menu-heavy interactions, not so much for representing stateful things like options and checkboxes and such. Metro isn't the problem, it's trying to shoehorn UIs into it regardless of fit that is.
I don’t agree, but that’s design, people have different opinions. I actually like the Ribbon interface, would have liked it more if they added a search box to it as well but designers hate search boxes.
100% agree, but that is in contention with the desire to invent something new. As a separate discipline where the career trajectory is determined by peers the user becomes less important.
These are all symptoms of a larger problem: very few people care about the users, and you have instead classes of workers living in a bubble, working towards either micro-optimizing metrics or trying to achieve what in their minds is the "ideal" product, pushing the latest fashions of their branch.
So UX engineers will unleash the latest fad (see Apple's glass UI, or Material Design, variations of flat gray design, etc...), PMs will insist in dumbing down UIs, engineers will push whatever micro-service architecture because it's "cool" or push for rewrites in Rust / Typescript. At the same time, it's very rare for companies to have a single person (or restricted group of people) with a global view on what the product line is trying to achieve long-term.
It really isn’t that hard if you frame it correctly.
Computers are data processing machines with input and output. People today think they are vehicles to show design skill, and that’s not what they are. Focusing on design instead of utility is how you ruin any UI/UX anywhere.
Sites like GitHub do not exist for the designer. Sites like GitHub exist for software developers. Software developers should be calling the shots on that site, not designers.
Ralphlauren.com should be designed by designers. Dieterrams.com should be designed by designers. Etc.
Sites for designers should be designed by people who want to show off their designs.
Sites for data entry and manipulation should be designed for those who use that information. Creatives should stay away from sites like GitHub.
Respect has to be earned, and I don't think anyone (within margin of error) with UX in their job title has earned it. Most of their work consists of shuffling design elements around for its own sake. Sometimes they strike gold (or at least silver or copper), but it never feels like that's done because they target a better design, rather they stumble upon it while making designs whose goal is to be different.
You have to go back to when it was called HIC (Human–computer interaction) to find people who weren't completely brain-dead or ad-pilled when it came to design, did actual work and research trying to make better designs, and thus were at least somewhat respected.
Most people with UX in their job title these days aren't really UX designers. They're graphic designers that now have UX in their title because that is the fashion.
> It’s like these products that are in a way some of the most valuable products in the planet, are given a junior web dev and a “UX designer” who really doesn’t know anything about UX.
What you pay attention to grows. And company's pay attention to those things that move the needle on revenue. For many successful platforms UX doesn't move the needle much anymore (if it ever did). LinkedIn has effectively won their space and a clunky UI isn't going to show up in the numbers.
LinkedIn might have amazing designers on staff, but if leadership isn't prioritizing updates and fixes it won't happen. And leadership won't prioritize it until the problem shows up in the numbers.
Companies are in it to extract as much value as possible for the least spend. Inside a bigco tech company nothing get engineering time allocated unless there is a monetary ROI attached. Which is why basic usability is neglected while features to sell you things are worked on constantly.
The other day I was visiting intercom support tool
I realized it has morphed into completely unusable tool with so many features that i don't even know what to do inside it anymore.
Same pattern I saw in many other tools and product. As time passes software becomes more and more complex, then a new one comes which simplifies something and then it also morphs into some enterprise behemoth
I would argue that senior engineers, of which I am one, are more of the problem than junior. We build fancy custom components when we should be using the existing ones.
Yes, the (senior) product and design people are part of the problem too.
What? In my experience the true seniors are the ones pushing hard for simplicity while the mediors build overly complex messes that one needs to be a rocket scientist to understand.
> that multi-million and even billion dollar tech companies don’t have perfect websites in terms of UX
I would have thought it'd be the opposite.
It implies have hundreds of teams and UI / UX often is "scaled" in weird ways where everyone does their own thing and becomes a giant mess.
Everything is "correct" when you slice it enough. So from team A's perspective this might be a gain. When you are a part of a team you only see and own this part. That's your KPI.
Unless there's real and working governance (often very very hard) then it's not happening. To get that governance you need company direction and company buy-in that stops managers trying to push new features fast to infinity.
Something about software engineering has gone wrong nobody thinks much about UX they blindly try to give functionality to the business/ customer requesting it but without considering whats already available and how to maintain status quo as much as possible. But theres also room to make things simple and intuitive.
Google released an AI music studio and their primary UI is literally an AI chat window. I absolutely hate UIs like that.
Recently I was buying furniture and it quickly became obvious that "Can I actually browse their catalogue?" is a requirement that really narrows down the search.
They make tens of billions, elsewhere to not even care about tiny UX issues like this.
At this point, it will stay broken because the amount of people complaining are not paying but are a tiny amount of people that will end up continuing to live with it.
Every bit of intuition you have using a browser, really? You click a link, the current page changes, you click back, it goes away. You cmd/ctrl click it opens in a new window, you right click and select "open in new tab/window" and it opens in a new tab / window.
Now, when you click a link in GitHub, the current page doesnt change. I want to look at the linked issue on its own page. That doesn’t occur anymore.
The page i wanted to go to pops up in a small overlay on the right hand side. The body text and content that I wanted to view is in a new, weird location, with the old page still behind it in the normal spot. It’s very unintuitive.
Thankfully either the behavior has reverted or I’m no longer in the A/B test. I can’t get the popup to happen anymore for me. (edit, nvm, behavior varies depending on repo or something? it acts completely differently on different pages, sometimes links are normal and sometimes they open in a popup. extremely annoying)
I'm completely confused by the issue, the linked page is a terribly unclear description. It doesn't clearly explain what prior behavior was, or even what the new behavior is precisely. What on earth is this garbled English supposed to mean:
> any link to an issue form an issue stared to open in a popup overlay instead of navigating to it
When I use GitHub now, I see that when I hover over a link to an issue, it provides a hover popup after a fraction of a second. I can still click the original link to navigate to the issue, or move my mouse and the popup goes away.
Is the complaint that these hover popups exist at all? Or is something else happening to certain people that they're complaining about? There isn't a link to an example page or anything. I'm just baffled here.
When you click the link it will not navigate to a new page, but instead open some kind of pop-up window with the other issue. I have been very annoyed by this behavior for the last couple of days.
I don't think GitHub has made a single UI change since ~2023 (when it went JS heavy) that I've liked. (Admittedly though, I've moved away from it for everything I have a choice about at this point, so it's possible they snuck in some good stuff when I wasn't looking.)
Also: having trouble getting this specific link to load -- just getting the unicorn error over and over.
It sounds like the root issue is that some people prefer opening new tabs while others prefer staying in the same browser window. I surfed the web when all links, even across websites always stayed in the same browser window, and I still prefer that. But I can understand that some people prefer opening new browser tabs instead.
I think web browsers should revisit how they handle links with target=_blank/_top, and show different cursers when hovering and let users customize the default behavior.
GitHub issues (well, PR comments specifically) is possibly the clearest example of developers not knowing how their users use the product. There are only 3 important user stories that matter for this workflow and none of them are done well:
- I want to review surrounding code and get context for a line level change. Can't do it without clicking multiple expanders and even that has a limit of 2 or 3. I also can't comment on surrounding unchanged code which is sometimes extremely relevant, like "copy this pattern"
- I want to see all the unaddressed issues. Ones that are not marked as resolved and not replied to, however you slice it, the issue filters simply don't work
- I don't want the PR author to be able to resolve issues without me getting indicated to verify them. The workaround is them commenting "fixed" on every issue. Make the button say "mark as resolved" and "verify resolved"
- Bonus: if you've got more than 40 comments on a PR, good luck finding some random subset of them. They're just unavailable and the UI unapologetically says "eh can't do it". Yeah small PRs but it happens.
Popup or inline i don't really care, the baseline workflow is completely uninformed.
I noticed this feature earlier in the week and found it helpful and intuitive. I suspect they tested this well and most users liked it. GitHub usually has top rate UX.
Performance is poor and there are a million other reasons to beat up on GH. This is not one of them.
Alas, GitHub has been plagued by bugs and UX regressions year after year.
I reported a bug last year about being unable to quote code blocks. It's quite a basic yet fundamental feature, right? They acknowledged the bug and moved on. To this day, quoting a block of code is still broken [1].
They simply don't care. I suppose their attention is focused on other subjects...
Anyway, I kind of accepted the "enshitification" of things I used to like. Fortunately, in this case, we can still hack our way around using custom userscripts [2].
I was responsible for this going out. The goal was to provide a more consistent user experience in that what happens when you click an issue would be the same in more places where we use the issue viewer (sub-issues on an issue, our dedicated issues dashboard (https://github.com/issues), GitHub Projects, and others). Like you mentioned, you also wouldn't lose your place when clicking an issue reference when reading a discussion. There were some performance improvements that came with the change too. It was well intentioned, but we hear you, and thanks for the feedback. We missed the mark on this one and it's been rolled back.
Interesting to see that Microsoft is now also ruining the old
UI. That was the only advantage GitHub would still have over Gitlab,
as Gitlab's UI was always horrible. And now Microsoft nerfs GitHub
here. This is epic.
There's browser extensions to bring back more user control on youtube, facebook, trello* and many others; looks like someone should make one for github soon.
*the markdown enabler needs updating last I checked
Last week there was a new Plex update to their already bad new redesign where they changed the main font. And I hated it, but it also reminded me, this font may not be objectively worse than the last one so much as the regular change is what has made me come to hate using the app. We don't give enough credit to maintaining the status quo. If software was getting better and discomfort with new designs was a tax we had to pay, then fine. That was still the world of ten years ago, perhaps. Now we're deeply into the era of software getting worse. The design changes from employees who have full time permanent jobs and need to make themselves busy aren't balancing actual progress.
If there's two things HN hates it's UX designers, PMs and off-by-one errors. As a UX designer I have to laugh: one of the most important parts of building good UX is humility and a willingness to be wrong. The confidence with with many on HN assert that they know what the right UX is for any app is exactly the same error that bad UX designers make.
As always in product the user's frustration is real and important but their ideas for a fix are almost never the best choice for the product, the company, or most users.
Tay.ai (Microsoft's infamous chatbot) and copilot are too busy vibe coding GitHub into the ground to look at the issue. There is no CEO of GitHub anymore to respond, which means no-one cares anymore.
Unfortunately GitHub, err Microsoft, stopped listening a long time ago. From the feed to text contrast to many more issues, their community feedback repo has become a place where complaints go to die.
Super annoying when I saw this. Initially I assumed I'd strayed into some quadrant of the UI space I hadn't been in before. But no they just broke it for no reason. Well, presumably the reason was someone expected to get a bonus.
This is why I kind of think that UI/UX should be handled by normal developers who do other things as well. People whose sole job is UI/UX must do things like this in order to stay employed, normal developers don’t. So teach normal developers how to think about UI and UX so that changes stop happening solely because a specialist needs to change something that does not need changing.
Sorry, UI/UX people, but if you were proceeding towards some finely crafted experience, you’d have honed in on it by now. You would have a set of rules that could be followed to present information in both a pleasing way and a useful way simultaneously and everyone would know how things work because everyone followed the same rules. None of that has happened. You are just changing things to change them.
This title is misleading: yes, a bunch of users didn't liked it. But of course there's UI research and likely A/B testing showed github that this might be preferable to the majority of users.
Personally, I don't like it much. It sounds like leakage from AzDO design. Maybe a option to turn it off would be the best way out.
Disclaimer: I work for msft, although I've no connection to github, ado or any other such tool.
This was a performance driven change. We added this as loading a cross repo issue is a much slower experience than loading an issue in the same repo due to the way the header is loaded (which is being worked on).
But we hear you on the feedback - we will roll this back while we keep pushing on the root performance causes.
[update - this change has been reverted and the previous behaviour is back]
How did the performance of GitHub become so slow in the first place? It didn't used to be this bad years ago.
Some hard numbers [1] as to why GitHub is struggling with stability issues, directly from GitHub's COO:
Yup, platform activity is surging. There were 1 billion commits in 2025. Now, it's 275 million per week, on pace for 14 billion this year if growth remains linear (spoiler: it won't.)
GitHub Actions has grown from 500M minutes/week in 2023 to 1B minutes/week in 2025, and now 2.1B minutes so far this week.
So we're pushing incredibly hard on more CPUs, scaling services, and strengthening GitHub’s core features.
1: https://x.com/kdaigle/status/2040164759836778878
All of which can be handled with horizontal scaling of identical components.
None of which explains poor latency when opening UI elements, which is more likely be explained by overuse of SPA or spaghetti code in microservices.
Update: yup, that’s exactly it, just as I guessed: https://news.ycombinator.com/item?id=47912867
AI. GitHub usage has exploded recently due to the ease at which code can be generated.
Not just due to code generation, but to AI code scraping and inspection.
From what I remember, it got much worse the moment they started requiring JS for displaying what would otherwise be mostly static (and thus easily cached) content.
> loading a cross repo issue is a much slower experience
Why not solve the real problem instead of putting in a janky workaround?
At risk of being cliche, it seems like you guys could benefit from the 5 Whys approach here: "Why is loading a cross repo issue slow?" and iterate until you discover the root cause, and fix that.
I suspect fixing the root cause is going to be a lot less glorious career-wise than implementing a UX change that is easier to tout at review time (well maybe not so much after this debacle).
> update - this change has been reverted and the previous behaviour is back
was an on-call engineer paged for this on the weekend just to roll a revert instead of waiting until Monday?
we both know the answer to that
To be honest GitHub should have like a switch for "preview stuff adopter" where you guys could give any benefits for it (maybe more copilot usage?). This way you can test with a specific public, using metrics and feedback, while testing and people could comment more about it.
I would like this personally as I hate change in general, but from their perspective it's not a great test because the sample is far from random. They should still do it though
Can you elaborate? The header meaning the top part of the page? I just checked on a recent repo I visited and it has the usual banner (which would stay the same), the repo path, some links, and some stats. Considering every page navigation would likely pull which links and stats are shown, why is this a delta to go to another repo and why are presumably 3 database entries (possible links, stars, forks) so slow?
Navigation within a repository does not reload the page, only the section below the header.
which is also driving me nuts because it frequently fails to update the issue and PR counts when I close issues or PRs. Only a hard reload, or closing the tab and opening a new new one, fixes it.
Yeah over the past six months I've trained myself to just hit Command-R every time I switch back to a GitHub issue tab, otherwise things get stale or broken far too often.
I can't speak for GitHub but I've worked on multiple nav headers for large SaaS products and they can be ridiculously heavy weight to render given they appear on every page. They tend to be a dumping ground for features, many of which require their own permissions checks, feature flag checks, etc. it's not unusual to have to perform hierarchical permissions checks. They also tend to contain contextual info about the current nav state and dynamic information about navigable states.
A lot of this can be cached but it's easy to see why moving from one repo to another will invalidate most or all permission checks and feature flag checks.
How many checks are we talking? A well-implemented monotonic system should be able to do tens of thousands of these checks (or more) in the time budget I associate with a heavy page, and start before the entirety of the permissions/feature data is available.
Yes, pretty much this as well as some additional complexities due to the issue content being in React and the header in Rails - to the cost of approx 500-800ms p50 for a page load vs sub 100ms for a nav to an issue in the same repo (or without the header which is what we tried with this change here)
Has the team considered going back all-in on Rails and SSR instead of this hybrid approach?
[delayed]
I'm curious, what causes the rails header to be so slow? They have a pretty good fragment caching story, don't they?
That does seem very long so it's good you're working on it
It's interesting to see that the UX issues that are annoying me when using Azure DevOps are finding their way into GitHub.
In case they are truly chasing Azure DevOps level UX, I would recommend they implement an HTML editor for issues that, depending on whether the user has dark mode or light mode enabled, saves some CSS of the respective mode and makes it unreadable if read within the other mode.
They should also order the comments in order of recency top to bottom so you have to read the page in reverse.
It’s always been interesting to me that multi-million and even billion dollar tech companies don’t have perfect websites in terms of UX.
Just last night I was helping my GF set up an ad for her job on LinkedIn. The UX was terrible. Like awful and basic things like save and exit were completely broken. Meanwhile LinkedIn makes what percentage of their revenue through ads? Same with google ads. It’s like these products that are in a way some of the most valuable products in the planet, are given a junior web dev and a “UX designer” who really doesn’t know anything about UX.
I don't think there is such a thing as perfect UX and I'm not asking for it. I just want them to stop making it worse.
Seriously tho, why isn't this something that a browser can do? Why can't I just split a tab and say all links from the left tab open in the right? Why not be able to scroll through history as a list of such panes like a smalltalk browser or file explorer on a mac? Maybe even a history tree, able to be forked with a click or two. Tree-style tabs are a baby step toward that, but I'm not seeing much interest out there in actually learning how to run.
Split tabs are now in:
https://support.mozilla.org/en-US/kb/split-view-firefox
Chrome also has split tabs since Feb '26
right click a link, open in split view
KDE's hybrid file / web browser konqueror has had arbitrary tab tiling since 1999 IIRC.. still a gread tool, would just need some love and webextensions support to come back big
I hate that feature and I hate that they keep bloating browser which was lightweight.
Just for the record.
When was Chrome lightweight? 15 years ago?
I also don't understand this feature. Like yo, we heard you like tabs, so we put tabs in your tabs so you can tabulate while you tabulate. Huh?
i occasionally need to compare two tabs. previously that meant that i had to open those two tabs in separate windows and then use window tiling to place them side by side. setting that up was a lot of work. and also it makes switching windows very hard. each side by side view would add two more windows that all need to be cycled through when i switch windows. and don't try to have more than two of those on a workspace. you'll go crazy switching between them.
with the split view it not only becomes very easy, but the split tabs also keep their position among all the other tabs, so i can keep the view permanently without cluttering up my list of windows. currently i have 5 split views in active use. that number is likely to grow...
I think it’s a nice feature. I use it to have designs on one part of the screen and implementation on the other. That way I can jump between “designs | implementation” and “PR | swagger” without managing and resizing tabs. Previously I had to jump between tabs and taking into account the newer screens provide a considerable amount of UI real estate there was screen area to utilize.
You might like Zen Browser
Just in case you aren’t aware, Edge can split a tab and open links from the left side on the right.
What’s edge ?
It’s a de-googled spyware app in case you’re looking to diversify your personal information loss portfolio across multiple firms.
https://en.wikipedia.org/wiki/Microsoft_Edge
It's precisely because they're so big that they can afford to overhire lots of designers, which then obviously need to justify their employment by continually changing things. This isn't a problem with small and tiny companies where "UX designer" might not even be a separate job but the responsibility of someone who will care only enough to make something that works and then leave well enough alone.
i have a really good friend who did the whole UI/UX design bootcamp during the explosion of UX/UI jobs. he did okay, he's probably hopped jobs 2-3 times now and is now without a job.
i actually feel for him, it's definitely one of the career paths that's looked at as excess/waste now while companies slim up to reappropriate money for AI. but i do think there was something there, he was genuinely passionate about what he did and it's just really hard to find work doing it now.
i feel guilty saying this but i've let him talk me through some of what he does, show me how he sees and approaches design (the bulk of what he did was design the interfaces for publicly used webapps and mobile apps) and... idk. i feel like it's all acquired taste and almost a "good app developer will think of these things when they design the front end" and a lot of his insight to me broadly looked like a lot of stuff i would've considered myself as a mere sidequest and my general thought process to deliver a good app. the difference is im building the app and designing the user experience, but his entire career is silo'd to just building the user experience.
im not against breaking out the design to a dedicated resource whether thats one designer or one team who wants to try and maintain a consistent language for a company. i think this has upside to make the design experience not locked to a single developer or developer team, and opens it up to a lot more channels of input. but on the other hand, like it's not the end of the world for me as a developer to come up with a really good design & i personally have never imagined myself not considering UX/UI at every corner when I'm building something. It feels like a second nature to me, there's creative aha moments to it, i think it's generally really good for a developer to step into a users shoes and almost "debug" the experience.
where i think ui/ux has gone off the rails:
- i think it's unduly influenced web design and has been poisoned by marketing. the rise of landing pages for SaaS that say a whole lot of fucking nothing and the crossover with "marketing research". i actually literally can't stand these types of pages, i swear 75% of the time i click around and can never get a straight answer on what the product/service is. examples: https://boomi.com https://www.astronomer.io
- things like OP, issue links opening in popups. changing things for the sake of changing things. such a change is probably "backed by research / surveys" giving the illusion that this was a data driven-decision, making it hard to push back on.... despite on deployment = everyone universally hating it. there seems to be some heavy flaws with the data sampling/collection methods that drive these decisions. i think the field of ux/ui as its own distinguished and defined field needs to undergo a self-awareness evolution here. something that's happened quite a few times in engineering. they really need to scoot back and have one of those "sometimes the best path forward is to not change anything at all" moments collectively and learn to recognize when that is right in front of you
- sometimes (maybe more than sometimes) allowing the business to dictate design is mayhaps not a good thing. i think what im trying to say here is the existence of "hes the ux/ui guy in the department, go talk to him" gives business stakeholders misaligned incentives to just go and push a change that isn't _actually_ user oriented, but is heavily tied into some metric or some other stupid business initiative. actually the more that i think about it this is probably why a lot ui/ux careers exist (give all control of the design over to the business) and that seems like a slippery slope
It gets increasingly difficult to design a website properly when you have different teams with different goals each competing to put their little feature front-and-centre, leading to a hacky job on top of a hacky job on top of a hacky job, which in turn hurts the performance until one day someone finally decides to re-think the whole thing from scratch and pisses off >50% of its users in the process that are used to the mess.
It's way easier to nail the UX when you're still in the dozens-of-employees stage of growth and offer like five features in total.
> It’s always been interesting to me that multi-million and even billion dollar tech companies don’t have perfect websites in terms of UX.
This, but for online shops, especially clothing. Horrendously buggy, laggy, with broken navigation (especially when navigating back), filters that don't work on > 95% of online stores. Why they wouldn't fix their primary (or at the very least highest margin) income stream is beyond me, but I've had to abandon so many shopping carts just because the checkout flow is literally broken.
The larger the company, the more it will be designed according to internal incentives, and less by people actually using their own product.
As someone who has built a lot of greenfield UIs while also maintaining old ones (13+ years old SaaS), I recently set up LinkedIn ads and realized the UX is abysmal considering it’s something they’re actually trying to make money from. Maybe—just maybe-I’ve seen such poor UX in a free web app that lacks a maintenance budget. The only reasonable explanation I can come up with is they have a lot of silos within the ad portion of their platform, and each team works on their little corner and no one tries to work with it end to end. Since it’s LinkedIn, this is inexcusable. You go and try to make an ad campaign and then an ad set within it containing some ads, and then come back to it a week later and try to find all these entities you created. You may land on one and take a very long time gritting your teeth and praying for a way click around until you can find another one. What‘s the net drain on worldwide GDP caused by the time-wasting UX of this component of LinkedIn?!
UX is really, really hard - and for some reason still not fully respected as a discipline.
I agree, and I think the metrification of UX hasn't helped here.
If you read the old Win32 interface design studies, and Raymond Chen's "Old New Thing, The: Practical Development Throughout the Evolution of Windows" you realize what people click isn't always what they want.
And old UX was ensuring that it was build in a way that what the user clicked was what they wanted.
Now? Since the MBAs came in the UX is another hostile piece of software, trying to trigger you into spending money.
Fast track to loss of respect:
I visit a site/launch the app I always use with the intent of getting something done quickly, and I find that since the last time I used it someone's rearranged the deck chairs and hidden or removed the functionality I need. Something that should take a minute or two suddenly becomes rage-inducing and eats an entire day.
Or the feature is still there but they've renamed it to something totally unrelated which you would never guess. Honestly, it's like they are actively trying to lose users.
The most depressing email to receive is "Good news! We've improved our website ..."
The Win8 and Metro design disaster is what happens when you give UX free rein, instead of focusing on users they try to start design trends to impress other UX / designers (essential for their career).
I wonder how much of Apples design was basically ‘if you confuse Steve Jobs you’re fired.’ And this acted as a necessary governing force to counteract the need to impress peers.
Metro was a wonderful design for the media player app it was made for. It's great for menu-heavy interactions, not so much for representing stateful things like options and checkboxes and such. Metro isn't the problem, it's trying to shoehorn UIs into it regardless of fit that is.
I don’t agree, but that’s design, people have different opinions. I actually like the Ribbon interface, would have liked it more if they added a search box to it as well but designers hate search boxes.
Part of UX is leveraging what users are already familiar with.
100% agree, but that is in contention with the desire to invent something new. As a separate discipline where the career trajectory is determined by peers the user becomes less important.
These are all symptoms of a larger problem: very few people care about the users, and you have instead classes of workers living in a bubble, working towards either micro-optimizing metrics or trying to achieve what in their minds is the "ideal" product, pushing the latest fashions of their branch.
So UX engineers will unleash the latest fad (see Apple's glass UI, or Material Design, variations of flat gray design, etc...), PMs will insist in dumbing down UIs, engineers will push whatever micro-service architecture because it's "cool" or push for rewrites in Rust / Typescript. At the same time, it's very rare for companies to have a single person (or restricted group of people) with a global view on what the product line is trying to achieve long-term.
It really isn’t that hard if you frame it correctly.
Computers are data processing machines with input and output. People today think they are vehicles to show design skill, and that’s not what they are. Focusing on design instead of utility is how you ruin any UI/UX anywhere.
Sites like GitHub do not exist for the designer. Sites like GitHub exist for software developers. Software developers should be calling the shots on that site, not designers.
Ralphlauren.com should be designed by designers. Dieterrams.com should be designed by designers. Etc.
Sites for designers should be designed by people who want to show off their designs.
Sites for data entry and manipulation should be designed for those who use that information. Creatives should stay away from sites like GitHub.
Respect has to be earned, and I don't think anyone (within margin of error) with UX in their job title has earned it. Most of their work consists of shuffling design elements around for its own sake. Sometimes they strike gold (or at least silver or copper), but it never feels like that's done because they target a better design, rather they stumble upon it while making designs whose goal is to be different.
You have to go back to when it was called HIC (Human–computer interaction) to find people who weren't completely brain-dead or ad-pilled when it came to design, did actual work and research trying to make better designs, and thus were at least somewhat respected.
Most people with UX in their job title these days aren't really UX designers. They're graphic designers that now have UX in their title because that is the fashion.
They're closer to artists; and of course art isn't practical, it's meant to be artistic.
> It’s like these products that are in a way some of the most valuable products in the planet, are given a junior web dev and a “UX designer” who really doesn’t know anything about UX.
What you pay attention to grows. And company's pay attention to those things that move the needle on revenue. For many successful platforms UX doesn't move the needle much anymore (if it ever did). LinkedIn has effectively won their space and a clunky UI isn't going to show up in the numbers.
LinkedIn might have amazing designers on staff, but if leadership isn't prioritizing updates and fixes it won't happen. And leadership won't prioritize it until the problem shows up in the numbers.
Companies are in it to extract as much value as possible for the least spend. Inside a bigco tech company nothing get engineering time allocated unless there is a monetary ROI attached. Which is why basic usability is neglected while features to sell you things are worked on constantly.
It's more like no one cares about UX. People keep using the product and they keep printing. Why invest in a UX researcher or designer?
The other day I was visiting intercom support tool
I realized it has morphed into completely unusable tool with so many features that i don't even know what to do inside it anymore.
Same pattern I saw in many other tools and product. As time passes software becomes more and more complex, then a new one comes which simplifies something and then it also morphs into some enterprise behemoth
I would argue that senior engineers, of which I am one, are more of the problem than junior. We build fancy custom components when we should be using the existing ones.
Yes, the (senior) product and design people are part of the problem too.
We need to build simpler software that works.
What? In my experience the true seniors are the ones pushing hard for simplicity while the mediors build overly complex messes that one needs to be a rocket scientist to understand.
> that multi-million and even billion dollar tech companies don’t have perfect websites in terms of UX
I would have thought it'd be the opposite.
It implies have hundreds of teams and UI / UX often is "scaled" in weird ways where everyone does their own thing and becomes a giant mess.
Everything is "correct" when you slice it enough. So from team A's perspective this might be a gain. When you are a part of a team you only see and own this part. That's your KPI.
Unless there's real and working governance (often very very hard) then it's not happening. To get that governance you need company direction and company buy-in that stops managers trying to push new features fast to infinity.
Something about software engineering has gone wrong nobody thinks much about UX they blindly try to give functionality to the business/ customer requesting it but without considering whats already available and how to maintain status quo as much as possible. But theres also room to make things simple and intuitive.
Google released an AI music studio and their primary UI is literally an AI chat window. I absolutely hate UIs like that.
The metric for perfect is
-Does it drive more people to the app -Does it maximize time spent on the site -etc
Your idea of perfect is very different than the one LinkedIn is using
Recently I was buying furniture and it quickly became obvious that "Can I actually browse their catalogue?" is a requirement that really narrows down the search.
They make tens of billions, elsewhere to not even care about tiny UX issues like this.
At this point, it will stay broken because the amount of people complaining are not paying but are a tiny amount of people that will end up continuing to live with it.
So it won't be fixed.
Great, the UX feature I probably hate the most in Jira, now on Github.
This was exactly my thought. It breaks every bit of intuition I have using a browser, and makes pages run even slower.
Every bit of intuition you have using a browser, really? You click a link, the current page changes, you click back, it goes away. You cmd/ctrl click it opens in a new window, you right click and select "open in new tab/window" and it opens in a new tab / window.
Now, when you click a link in GitHub, the current page doesnt change. I want to look at the linked issue on its own page. That doesn’t occur anymore.
The page i wanted to go to pops up in a small overlay on the right hand side. The body text and content that I wanted to view is in a new, weird location, with the old page still behind it in the normal spot. It’s very unintuitive.
Thankfully either the behavior has reverted or I’m no longer in the A/B test. I can’t get the popup to happen anymore for me. (edit, nvm, behavior varies depending on repo or something? it acts completely differently on different pages, sometimes links are normal and sometimes they open in a popup. extremely annoying)
And GitLab, too!
I'm completely confused by the issue, the linked page is a terribly unclear description. It doesn't clearly explain what prior behavior was, or even what the new behavior is precisely. What on earth is this garbled English supposed to mean:
> any link to an issue form an issue stared to open in a popup overlay instead of navigating to it
When I use GitHub now, I see that when I hover over a link to an issue, it provides a hover popup after a fraction of a second. I can still click the original link to navigate to the issue, or move my mouse and the popup goes away.
Is the complaint that these hover popups exist at all? Or is something else happening to certain people that they're complaining about? There isn't a link to an example page or anything. I'm just baffled here.
When you click the link it will not navigate to a new page, but instead open some kind of pop-up window with the other issue. I have been very annoyed by this behavior for the last couple of days.
I don't think GitHub has made a single UI change since ~2023 (when it went JS heavy) that I've liked. (Admittedly though, I've moved away from it for everything I have a choice about at this point, so it's possible they snuck in some good stuff when I wasn't looking.)
Also: having trouble getting this specific link to load -- just getting the unicorn error over and over.
It will probably suffer the same fate as the most-upvoted discussion of all time in the GitHub Community repo: https://github.com/orgs/community/discussions/66188
no reaction
It sounds like the root issue is that some people prefer opening new tabs while others prefer staying in the same browser window. I surfed the web when all links, even across websites always stayed in the same browser window, and I still prefer that. But I can understand that some people prefer opening new browser tabs instead.
I think web browsers should revisit how they handle links with target=_blank/_top, and show different cursers when hovering and let users customize the default behavior.
This is not about new tabs. It refers to an in-page panel that displays the content of the linked issue instead of navigating to it.
Idk, it almost seem a workaround for slow/broken go-back? If go-back is fast and state preserving, it's basically a fullscreen modal.
All(?) browser open links in a new tab when middle-clicked?
GitHub issues (well, PR comments specifically) is possibly the clearest example of developers not knowing how their users use the product. There are only 3 important user stories that matter for this workflow and none of them are done well:
- I want to review surrounding code and get context for a line level change. Can't do it without clicking multiple expanders and even that has a limit of 2 or 3. I also can't comment on surrounding unchanged code which is sometimes extremely relevant, like "copy this pattern"
- I want to see all the unaddressed issues. Ones that are not marked as resolved and not replied to, however you slice it, the issue filters simply don't work
- I don't want the PR author to be able to resolve issues without me getting indicated to verify them. The workaround is them commenting "fixed" on every issue. Make the button say "mark as resolved" and "verify resolved"
- Bonus: if you've got more than 40 comments on a PR, good luck finding some random subset of them. They're just unavailable and the UI unapologetically says "eh can't do it". Yeah small PRs but it happens.
Popup or inline i don't really care, the baseline workflow is completely uninformed.
Links should be links. Stop making them into something else.
I noticed this feature earlier in the week and found it helpful and intuitive. I suspect they tested this well and most users liked it. GitHub usually has top rate UX.
Performance is poor and there are a million other reasons to beat up on GH. This is not one of them.
Just improve what you have GitHub. Stop the AI bloatware. You will lose that race anyway, obviously.
I still don't understand what's the point of any full screen popups are
And they pushed this as every major browser introduced a "Split View" feature...
I get this issue preview on Projects, although I don't like it there either, but as a hook on any issue link is just terrible UX, zero benefits IMHO.
Alas, GitHub has been plagued by bugs and UX regressions year after year.
I reported a bug last year about being unable to quote code blocks. It's quite a basic yet fundamental feature, right? They acknowledged the bug and moved on. To this day, quoting a block of code is still broken [1].
They simply don't care. I suppose their attention is focused on other subjects...
Anyway, I kind of accepted the "enshitification" of things I used to like. Fortunately, in this case, we can still hack our way around using custom userscripts [2].
[1] https://imgur.com/a/github-bug-cant-quote-blocks-of-code-Z9O...
[2] https://github.com/orgs/community/discussions/192665#discuss...
shift+cmd click is the macos shortcut for "open this link a new tab and go to that tab"
Works on these psuedo links all over github
I wish they’d focus on making their platform reliable and more stable.
Why would they care to empower competitors?
Being able to see a detail view without navigating away from the list view is a better user experience and the more common practice now
I was responsible for this going out. The goal was to provide a more consistent user experience in that what happens when you click an issue would be the same in more places where we use the issue viewer (sub-issues on an issue, our dedicated issues dashboard (https://github.com/issues), GitHub Projects, and others). Like you mentioned, you also wouldn't lose your place when clicking an issue reference when reading a discussion. There were some performance improvements that came with the change too. It was well intentioned, but we hear you, and thanks for the feedback. We missed the mark on this one and it's been rolled back.
Interesting to see that Microsoft is now also ruining the old UI. That was the only advantage GitHub would still have over Gitlab, as Gitlab's UI was always horrible. And now Microsoft nerfs GitHub here. This is epic.
There's browser extensions to bring back more user control on youtube, facebook, trello* and many others; looks like someone should make one for github soon.
*the markdown enabler needs updating last I checked
There's already something like this for GitHub: https://github.com/refined-github/refined-github
Thanks for sharing - loads of great ideas there for the GitHub team.
They keep adding “fancy” UI and hijacking standard browser behaviour that is infuriating on a daily basis.
Please consider a lofi version for people that want to select text without navigating to a different page.
Last week there was a new Plex update to their already bad new redesign where they changed the main font. And I hated it, but it also reminded me, this font may not be objectively worse than the last one so much as the regular change is what has made me come to hate using the app. We don't give enough credit to maintaining the status quo. If software was getting better and discomfort with new designs was a tax we had to pay, then fine. That was still the world of ten years ago, perhaps. Now we're deeply into the era of software getting worse. The design changes from employees who have full time permanent jobs and need to make themselves busy aren't balancing actual progress.
It took me a while to realize it was not a bug. Utterly insane that this went through QA.
> Utterly insane that this went through QA.
Big assumption you're making there.
If there's two things HN hates it's UX designers, PMs and off-by-one errors. As a UX designer I have to laugh: one of the most important parts of building good UX is humility and a willingness to be wrong. The confidence with with many on HN assert that they know what the right UX is for any app is exactly the same error that bad UX designers make.
As always in product the user's frustration is real and important but their ideas for a fix are almost never the best choice for the product, the company, or most users.
If anyone knows someone at GitHub and can tap them on the shoulder, please ask them to revert this terrible change.
Tay.ai (Microsoft's infamous chatbot) and copilot are too busy vibe coding GitHub into the ground to look at the issue. There is no CEO of GitHub anymore to respond, which means no-one cares anymore.
Unfortunately GitHub, err Microsoft, stopped listening a long time ago. From the feed to text contrast to many more issues, their community feedback repo has become a place where complaints go to die.
I don't hate the change iself. But I do hate that it is inconsistant.
Super annoying when I saw this. Initially I assumed I'd strayed into some quadrant of the UI space I hadn't been in before. But no they just broke it for no reason. Well, presumably the reason was someone expected to get a bonus.
This is why I kind of think that UI/UX should be handled by normal developers who do other things as well. People whose sole job is UI/UX must do things like this in order to stay employed, normal developers don’t. So teach normal developers how to think about UI and UX so that changes stop happening solely because a specialist needs to change something that does not need changing.
Sorry, UI/UX people, but if you were proceeding towards some finely crafted experience, you’d have honed in on it by now. You would have a set of rules that could be followed to present information in both a pleasing way and a useful way simultaneously and everyone would know how things work because everyone followed the same rules. None of that has happened. You are just changing things to change them.
This title is misleading: yes, a bunch of users didn't liked it. But of course there's UI research and likely A/B testing showed github that this might be preferable to the majority of users.
Personally, I don't like it much. It sounds like leakage from AzDO design. Maybe a option to turn it off would be the best way out.
Disclaimer: I work for msft, although I've no connection to github, ado or any other such tool.
> A/B testing showed github that this might be preferable
A/B testing can’t measure preference, only interaction.
I guarantee you it's not preferable to the majority of github users.
Can you actually show us this research and a/b testing?
Except there was a different comment from someone that actually knew how it came around, and it was an ugly performance workaround.