As someone who was buying hardware in the late 1990s, it is hard to overstate the difference in buying experience between Sun or Digital (DEC) and someone like Dell. The former forced you into a live sales meeting, endless quote revisions, it was a nightmare. I remember calculating the server rails and power cords for a new Alpha server were going to cost more than a shipped/delivered Dell server that I could get the next day.
For us dell has become that. Rep changes all the time, what appears to be random pricing and long unclear delivery dates. Instead I can for example go to Thomas Kren and configure my serves online for a fraction of the cost. They have what they list in stock, delivered a few weeks later. Everything is clear and hasle free.
At this point I have the feeling dell just charges you what they think they can get out of you. You push back and the price drops, I don't want to negotiate.
The iron law is that vendors want to become like the old Sun. The only reason they don't is if they are trying to compete their way up the hill. Take away the old Sun and SGI and IBM and now Dell is Sun.
Difference is Sun was providing a "full stack" proprietary product including CPUs running a proprietary ISA, Unix OS, compiler, desktop environment and productivity apps. All that Dell does is assembling standard, commoditized parts and putting a free OS on it. Dell's business is incredibly easy to replicate, while Sun's business is extremely hard to replicate. Of course, Sun's competitive advantage turned into a curse when the components of that vertical integration become uncompetitive both features and price wise vs. the GNU/Linux-on-Intel stack. But regardless, Dell could never replicate the vertical integration Sun had.
Dell puts a fair amount of engineering into their servers (as do other main brands like HPE and Lenovo) so it's unfortunate they've lost their roots from the old mail order sales company.
If someone else is paying I'd prefer one of these big OEMs for established firmware lifecycle, spare parts and long gray market availability, shock (shipping) and thermodynamics engineering. If I'm paying, Asus or MiTAC (Tyan) are good enough.
Really? When I first met Sun machines they were selling 68k-based Sun machines that were better specced than Apple and Amiga and such but left them dependent on Motorola’s architecture that had indirect address and thus no future.
Sure they had an in house chip development ARM team right at the time when RISC economics looked good but CPU architecture had to work harder and harder to beat the memory wall and SPARC was not competitive by 2005 and “Hail Mary” projects like CoolThreads were difficult and expensive but did not deliver for customers.
Hypothetically customers could benefit from that proprietary stuff but they benefit even more from rapid progress is competitive, commodity markets.
When I was working at the library circa 2005 we were thinking about replacing all the Windows computers that patrons used to access the online catalog (and other services) with a web browser with Sun Rays, we probably would have bought a few hundred of them and we would have bought some SPARC servers to back them up.
We could have gotten Netscape to run under Solaris but after a week of effort I wasn't able to get Mozilla to build, so no modern web browser. So no Sun Rays. No servers. DTrace and ZFS were great innovations but without the magic of "can run the software you need to run" they are just Blub. Had Sun done the work to get Mozilla running on their machines and forgotten about DTrace and ZFS people might still be using Solaris... but instead DTrace and ZFS found an audience by supporting... the commodity OS!
(arguably at that time supporting a web browser was the #1, #2 and all the way through #99 task that the Solaris team should have been working on at that time and everything else from Java to Tcl to ZFS was a dangerous distraction... and that's how companies in a position like Sun die)
DELL was an innovator in this "incomparable pricing" model for normal consumers.
You'd price out a machine and then go look at comparable machines (difficult) from other vendors, then the next day your machine wouldn't be "the" configuration anymore. Instead there'd be half the ram and double the harddrive space for an extra $50, and if you swapped back to what it had been yesterday it'd cost you an extra $200.
IIRC they won business awards for this.
Where's Cartman when you need a vigorous "SHENANNIGANS!!!!"?
We sold web traffic analysis software to Dell Europe, around 1998 or so. They never kept their data more than a week. As the traffic and purchase behaviour they were analysing in the store changed so much every week they said (if I remember correctly) it didn’t make any sense to compare the data over time. I took from that that what they sold every week changed all the time. Of course, data storage and processing was then not as cheap as it is today, so maybe there was more to it.
> At this point I have the feeling dell just charges you what they think they can get out of you. You push back and the price drops, I don't want to negotiate.
This is precisely the role of the sales person for something like a service, but for a physical product it is pretty egregious. There is no greater sin than leaving money on the table. There's an old story about someone offering a quote that was apparently pretty low based on reaction, so they quickly add "that was for my assistant" to give way for pushing the number higher.
I like this version: The optician quotes the price of new glasses by saying "$100", waits to see if the customer flinches, "... for the lenses", waits again, "... each.".
> so they quickly add "that was for my assistant" to give way for pushing the number higher
I don’t understand what this means? I understand that they bumped the price. But I du not understand what is meant by it being for the assistant? Care to explain?
Pretty much, except the assistant is never going to handle things and the updated fees go up to include you and the assistant. If you only ever send out your assistant, you never make money
If you play poker, it is the equivalent of pushing 'portions' of your bet out at a time. You push the portion, observe their face and body language, and maybe push the rest.
The 'that was for my assistant' is just something they say to explain away the earlier quoted price. They are pretending that what they said wasn't the quote for this conversation, it was for another, different conversation, between the salesman and their assistant. This only works when the assistant is not in the call.
> At this point I have the feeling dell just charges you what they think they can get out of you
This is how license compliance shakedowns work. Make sure your DRM is more flexible than theicensing terms, find an overage and see what you can get the customer to agree to.
I have seen senior Oracle sales reps deliberately misreading their license doc to inflate what is owed.
I think this is a case of AI ruining the hardware market. We used to almost exclusively use Dell, but now we've been forced to look elsewhere and go through validating new vendors. The lead-times are ridiculous, and these guys now only pay attention to the really big guys. Not sure I can blame them though, you've gotta do what's best for your business.
Dell was very much moving this way before modern AI explosion.
The inconsistency on pricing alone depending on what rep was helping you also became a problem. I could often times call in direct and order via business line cheaper than dealing with my account rep. No reason to jump through so many hoops.
I prefer the "This is the price" model and just make it consistent. I can't wait until AI helps us remove multiple layers of middle men across Sales in general.
> I can't wait until AI helps us remove multiple layers of middle men across Sales in general.
Based on how things are moving, I think it's more likely that AI will _replace_ the layers of middle men (with layers of "middle AI"), then that those layers will be _removed_.
Of course, the process (of squeezing the customer) is the point. The implementation of that process, with either human account reps or sales bots, is irrelevant.
that's great as an attitude, the problem with it is that money is made up and so are prices. Business have an inventory and expenses, and hopefully they can sell enough product before their debtors come calling. There's no right price for something. Underpriced, Overpriced, the only qualifier is: is the company still in business? Sony could give Playstation 6's away for free, so long as people bought games for it, but culturally, that would look desperate, so that will have to charge money for it, when they announce it. The only thing something is worth, is what someone else is willing to pay for it.
Periodic component shortages have been a fact of life in the server business forever. Like a key supplier's factory burns down or something.
The key thing is being transparent with customers on price, availability, and delivery schedules. In the short term server vendors can use tricks and personal relationships to extract higher profit margins but eventually the senior executives at major customers figure out what's happening and start looking for other options.
> The lead-times are ridiculous, and these guys now only pay attention to the really big guys. Not sure I can blame them though, you've gotta do what's best for your business.
I don't blame them in the sense that I wish for them to be imprisoned or anything. But I wouldn't do business with them, either. I want business partners whom I can count on, not ones who will stop returning my calls as soon as a more lucrative opportunity arises.
Yeah, and then Dell more-or-less turned into that. Companies get big, customers and sales get more complicated, process gets more calcified.
SUNW would probably still be around if the whole x86 commodity pizza box revolution hadn't happened OR if they had pivoted away from SPARC earlier and went heavily into x86. Example: buying from IBM doesn't sound too different from buying from Sun, but they're still around because mainframes will be around until the Earth implodes (probably).
(I think that their channel/partner programs along with their professional services org are big reasons why Dell is still a key player in the datacenter space. You can buy Dell direct, and the sales process is still "good" as far as enterprise sales goes, but bigger players usually get their Dell kit through distributors/vendors with Dell professional services handling architecture, build-out, enablement, etc. I'm not sure how much energy SUNW invested in this, as they were before my time, but this combo is a major revenue driver for them.)
We have a couple mainframes at work, had an incident last year when someone forgot to apply a new license to one. p1, no payments happening.
I was kinda stunned; there’s a level of balls involved in selling an x millions machine and having it fail closed that I’m sort of impressed by but also utterly appalled they’d treat their big customers like that..
I mean surely to fuck it could have been an automated reminder email to the mothership and not just: sorry folks no banking for you today.
Also thanks to virtualization being originally invented in mainframes, and the OS design done by IBM (and Unisys as well), they get to run the old workloads on modern hardware, and also support various UNIX flavours.
You can even run most cool Python, Go, Java, Rust, PHP among others, on systems like z/OS.
In a time frame like 1999 to 2002, much more broadly than that, just the increasing gap in cost vs what you would get between buying any hardware+OS package from a legacy Unix-like vendor (Sun, DEC, SGI) and doing your own thing on a FreeBSD, OpenBSD or Linux on x86 hardware.
Dell was obviously a top choice, but by 2000 or so, Linux or FreeBSD had become good enough that you could do exactly the same thing for a fraction of the cost and run it on x86 hardware you had purchased from either Dell or one of the top 5 or 6 Dell-competitors.
As compared to, as you say, sitting through sales meetings, price quotes, revisions, a long slow and arduous process to arrive at a figure where you come to a conclusion "yeah you can spend $60,000 on these Sun servers or we can buy $20,000 of x86 servers and get creative with our own software stack to do the same thing in house".
At a certain point 1U and 2U server hardware got fast enough on boring standard x86 single or dual socket motherboards from Taiwan, in the era of like the 1.2 GHz Pentium 3 with 512KB cache, that the performance was pretty damn good and the machine cost a fraction of what the equivalent Sun or SGI rackmount machine would cost. This is in the era before a multi core x86 CPU existed. And you could be sure that its BIOS and motherboard were already OK for use with a current Linux kernel, we're talking about motherboards from Supermicro, Tyan, intel whitebox reference server boards, etc.
It's easier to see in hindsight but servers have eaten a lot of macro economic headwinds.
1999 and 2002 correspond roughly to telecom and dotcom bubbles popping. So open source operating systems were also pulled up into these positions through a lot of work and by the early 2000s it was kind of obvious this was the way forward.
2008 financial crisis really knocked the high end (RISC/UNIX/Itanium) and started the long migration to renting vs owning. In hindsight it's remarkable how much money used to just be there in the economy of all this. But I guess it still is, considering how expensive public cloud is.
> But I guess it still is, considering how expensive public cloud is.
It's not like there's less servers being sold, they're just bespoke designs from companies like the big Taiwanese ODMs that go straight into hyperscaler facilities, and we don't get to see public marketing websites and product info about them. The open compute platform stuff that's been publicly shown is like the tip of an iceberg.
I'm referencing the premium you pay to get some paltry VM in a public cloud with bizarre storage and networking that primarily make sense for the market maker versus the actual cost of the hardware. There are ways to make cloud economics work as a user, but only where staffing costs would dwarf spend (i.e. small businesses) and I've never seen them in practice.
> The open compute platform stuff that's been publicly shown is like the tip of an iceberg.
I find it a real shame it's not easier to get OCP hardware for our homelabs. Would be so much easier to operate, upgrade, and so on. And where would you even get decommissioned OCP gear?
Some of the ethernet switching top of rack 1U stuff used to show up on ebay. I think large other portions of it go into an industrial shredder/recycling process.
One of the first "expert systems" was one that helped you spec your DEC computer. It was easy to order a VAX that was missing necessary hardware (or had incompatible hardware).
If you click through to the anecdote[0] mentioned in the article, the problem was not with having to contact a sales rep. In fact the startup wanted a sales rep. Sun specifically asked them to use the website:
> More than once though, the Sun people did suggest I just buy the servers off the web site.
Dell was just the only one to callback.
I don't doubt your experience but that was at the height of Sun's business, not it's nadir as this article is about.
Sun commercial sales reps, at least in Europe, were astoundingly slimy.
I had a colleague where almost his whole job consisted of beating them in sheer sliminess, and at this he was good. I don't think we ever paid 30% of the prices they'd start at, but the process of achieving this was disgusting, and there was still the sense everyone was being screwed.
Yeah, and it wasn't just the "call for pricing" game, it was having to deal with monthly "touching base" calls which were mostly an opportunity to push corporate hospitality and other not-quite-inducements.
Whereas Dell just gave us a flat 30% discount on list pricing, and were clear in advance about what level of annual spend would trigger an increase to 35%. We'd order online and our sales rep would phone a few hours later to confirm. Other than that, they'd initiate contact maybe once a year when Dell refreshed their server or storage product lines and would offer to either send some bumpf or arrange for some demo kit.
Dealing with them was so much less hassle all round, you never got covered in slime, and you never felt like they were on the verge of offering you bribes.
Yeah that fits. I was in London (though not the City) and it was definitely that sort of scene.
Thinking about it, we moved to Dell as part of the run up to IPO, and I suspect had we not done so Sun would have been much harder to deal with afterwards.
What I recall from that time is Sun were very happy they'd sold a huge pile of servers to be the core of Interlink, and a whole load of people went off to work on that.
I’ve had that experience with dell consistently over the last 6+ years. While with other vendors I can do everything over email from start to finish, Dell insists on meetings, follow-ups on topics that I said I am not interested in talking about, services that I do not need, meetings with teams I am not interested in meeting, reps that straight up skip meetings, so I have to console a junior rep that got stuck in a meeting that is above their pay grade. Different reps same slimy bs.
Yeah it was so much more accessible to play around with configs and pricing and then all you needed was a credit card.
Even now I think enterprise companies should be more open about what's possible and what the price and delivery date would be without having to talk/chat/email anybody. Even for something like a Juniper $100k+ device it would be kind of cool to configure it.
And that is the "thing". Every failed mini company had the same exact issue, seems customers had to beg and/or go through hell to purchase equipment. Never mind that the hardware was more expensive than what you cold buy from companies like Dell.
IBM is not much better than what I hear, but somehow they keep surviving. Probably because of their mainframe business.
> IBM is not much better than what I hear, but somehow they keep surviving. Probably because of their mainframe business.
Imagine every bank you know going through tech modernization and maybe 70% of those are now using Red Hat OpenShift.
They own Red Hat (OpenShift kubernetes, Red Hat Enterprise Linux, Ansible Automation Platform) AND HashiCorp (Terraform, Nomad, Consul, Packer, Vault, etc.) now, on top of their previous mainframe, services gigs, global financing arm, and calling everything they sell "watson".
I needed to add a bunch of capacity running Sun/Solaris back in the late-90s. Same thing: pointless meetings with sales reps without clarity in pricing. Finally, I found the gear we were looking for on eBay, bought it, drove over and picked it up in my Ford Explorer, and had it up-and-running in one day. To my surprise, people in my company were PO'ed. But I had air cover because the CEO found out and loved it.
I've written several times here about all the mistakes that Sun made during the 00s that doomed it. Among the many mistakes it made are:
- Cancelling -if briefly- Solaris on x86 in 2002. This killed Solaris in the minds of many who didn't want to be locked into Sun for SPARC.
- Failing to make a deal with Google in 2002. Apparently Sun insisted on knowing how many servers Google had, something that Google considered a high-value secret, so Sun failed to make a deal with Google, so Google ended up using Linux and contributing to Linux. This was a tremendous mind-share disaster -- it's hard to overestimate the damage done by this.
- Closing Sun PS (professional services). Bad bad move, possibly the worst of them.
- Not giving up on J2ME earlier -- it's not the sort of thing that could last forever, and Steve Jobs killed it with the iPhone. This was a case of vendor lock-in clouding Sun's decision making.
- Failure to recognize that Sun needed to become a systems company, not a CPU company.
- Failure to respond to Active Directory. This was yet another case of vendor lock-in clouding Sun's decision making: the Sun DS product team was milking their existing customers more than they wanted to go after more business with a sustainable strategy.
- UltraSPARC was more than a decade too late to make up for SPARC falling way behind x86_64. Sun needed to give up on SPARC, but again, vendor lock-in sounds sweet but turns out to be poison.
- Failure to make a deal with Apple for it to use ZFS in OS X.
- The MySQL purchase. WTF, this was horrible and stupid. The only interesting effect of this was to make Sun a target of acquisition for Oracle. But of course, it turns out that Oracle -a company built on building mind-share- had become too blinded by vendor lock-in just like Sun, so...
There were numerous other mistakes along the way. These are the most salient, for me anyway.
What's shocking is how long it took Sun to fail under those circumstances!
Also shocking is how much amazing stuff came out of Solaris engineering and the systems division!
> - UltraSPARC was more than a decade too late to make up for SPARC falling way behind x86_64. Sun needed to give up on SPARC, but again, vendor lock-in sounds sweet but turns out to be poison.
UltraSPARC came out in 1995, and Athlon 64 in 2003. I'm confused on how it could be late?
> - The MySQL purchase. WTF, this was horrible and stupid. The only interesting effect of this was to make Sun a target of acquisition for Oracle. [..]
I disagree with this assessment strongly. I believe Oracle wanted Java and only Java. All other Sun properties were collateral damage of the acquisition. It might be weird that Oracle now owns MySQL, but Oracle never cared about it. (Of course, both your and my assessment are mere opinions.)
I sold my Sun stock at the internet bubble top for $70 a share. A few months later it was $7. I think of that when I see PE ratios in the hundreds for Tesla, SpaceX and other AI stocks.
Good timing. But on the other hand if you had sold AAPL, MSFT, NVDA, ORCL, AVGO, ADI, KLAC, LRCX, etc at the Internet bubble top then you would have missed out on huge gains since then.
Your point is correct, but let's not gloss over that most of those had 80% drawdowns or more. The nasdaq took 15 years to recover and it's only been 25 years total, meaning 60% of the time since then you'd have been living with a paper loss. Timing is difficult but powerful. (Just not as powerful as 25 years of waiting.)
> […] etc at the Internet bubble top then you would have missed out on huge gains since then.
Only if you never repurchased.
Or you could DCA out: have a standard policy that no single stock can be more that x% of your portfolio, and as any one gets above that limit you crystallize profits. It's no different than having a 60/40 portfolio and rebalancing your equities if they get >60%.
I lean more towards passive and (total) market index funds, and just riding the roller coaster, as anything else is fairly either impossible or too time consuming:
Since we're sharing, I have fond memories of my using my the Sun thin clients at university. They were slow to start up and use for the most part, but once you got to the terminal... it was great.
Being a hacker-type, I did everything with it.
Email? Forget waiting for some bloated email client, I'm using pine.
Want some html on my personal page? Use vi.
I think they didn't take away my access for years after college, and I would sometimes login and poke around just for fun.
I do blame them, along with the rise of Java and the awful gui toolkit motif for convincing me to drop out of computer science and switch to Computational Math for my bachelors degree. I was so frustrated that our grades were dependent on learning something that I saw as completely worthless in light of the dotcom boom happening at the time. I wanted them to teach us web technologies, or at least some good system code, but all I got was broken x11 and awful performance.
I think motif was fine for its time, but definitely not much better than OpenLook.
However, 30 years ago, developing a standard SunOS app with devGuide was the most enjoyable and productive programming I've ever experienced. Simply layout your GUI, link actions to methods, generate code, fill in the abstracted methods, compile and run. I didn't have to learn X or any other bloated low level graphics library.
I used a machine that had 8MB of RAM and a huge b&w monitor. One day I found out it would crash if your process forked more than 127 times. These things were so cool at the time but are nothing compared to a Raspberry Pi.
It's absolutely crazy to realize that even a Pi 400 running from MicroSD would smoke classic workstations of yesteryear, and yet by today's standards it's slow.
> I do blame them, along with the rise of Java ...for convincing me to drop out of computer science and switch to Computational Math for my bachelors degree.
I avoided studying CS for this reason: I saw my girlfriend's Java textbooks and thought if I had to appreciate this, I wasn't cut out to be a programmer, and stuck to network engineering. Years later, after 'Gang of Four' style OOP started dying, I have some code in a new things people use here everyday.
I was spoilt by SGI and their wonderful interface, where every detail was awesomely cool. The Sun machines were just old-fashioned at this superficial but important level of how it all looked, but only if you knew better, and that meant experience of SGI. In the world of IRIX, we had CPU-eater desktop wallpaper, meanwhile, on Sun, you had some badly drawn window decorations.
I imagine this to be the awful GUI that you speak of.
I say look and feel is superficial, however, when you have to learn a variant of UNIX with slightly different command line switches to the UNIX you know, with an interface worse than even Windows (3.11), for it to run dog slow, a lot of copium is needed. A poor interface gets in the way of what you are trying to do, a cool interface just lets you get on with it.
Back in the early 90s, I ran several procurements for university computer labs. We were under severe budget constraints, and, as a computer science department, we had constraints not normally found in such labs (for example, we needed to use certain compilers, so they had to support the workstation architectures). We wrote RFPs in accordance with university policies, and, while a number of potential bidders couldn't meet them, the companies in question all understood the constraints, and unsuccessful bidders were constructive, on the grounds of “consider us next time, ok?” All of them were like this, except one. The only time I have ever had someone threaten my job was the Sun Microsystems sales rep, after they lost a competition.
> Sun had become bored with the mechanics of running a business.
I would contend that Sun was never interested in running a business. They always cared more about building amazing technology and suffered the sales to make money.
You can't deny that Sun hardware was top tier and always ahead of its time (until Oracle bought them).
Edit: I had on rose colored glasses. I was thinking of late 90s SUN. They did in fact start getting crushed in the early 2000s.
I think this sorta forgets how slow the cpus were - I was running mostly ASIC design workloads on big Sun hardware in 2001 and we eventually got some x86 boxes that were only 32 bit but could run some of our jobs - and the single-threaded perf of these commodity machines was amazing compared to the Sparc boxes. When x86-64 came around, this was the death of all the Sun hardware in that group.
Also sometime in 2000 I had interviewed at Compaq - the old Alpha group who I think later worked on Itanium at Intel. They were constantly frustrated because they couldn't even get a build from Synopsys of the design tools they needed to run on Alpha and were forced to use Sun hardware they described as "gasping and wheezing" to design their crazy fast cpu.
Sun lost on a bunch of fronts, but I think the amazing thing is they let happen to them what they did to the other big UNIX vendors before - they were eaten from the bottom by faster cheaper hardware.
Yes, I had the same experience while using various Sun computers between 1995 and 2005. During all that decade the Sun computers were reliable, but excessively slow.
Around the beginning of 2006, I had a laptop with a 64-bit AMD Turion CPU and I could run on it Cadence and Mentor tools for ASIC design and the simulations were much faster than on a big Sun multi-processor server. Later that year our team finally got some servers with AMD Opteron, which accelerated dramatically the work in comparison with the traditional Sun servers.
I recall a fork/exec heavy shell script at a previous job that ran about an order of magnitude (~25s vs 3s) slower on Solaris/SparcV9 vs Linux/Intel P3 circa 2003. The standard retort back in those days was to rewrite it in Perl.
I can deny it. It was OK, but not amazing. I had to make pretty frequent repair calls on their gear--much more frequently than the less-technologically-amazing-but-solid-as-a-rock Dell and Compaq gear. IBM, Digital, and HP (yes, that HP, before they lost the Way) also had very good UNIX gear, but Sun had the mindshare during the '90s.
In terms of performance, even Linux 0.x on a 33Mhz 486-class processor could outperform a uniprocessor Sun/SPARC box at a fraction of the price.
I also question the sanity of anyone who thinks Sun's choice of 13W3 for a video connector was a good one. :-)
I loved Sun hardware but I'll deny that they were ahead of their time after 2003 or so. I had a boss who loved his pizza box but Intel/Linux crushed them in the 2000s on a price/performance basis.
Sun's various divisions all cared very much about their respective vendor lock-in, so they didn't innovate fast enough. The divisions that didn't have that innovated. On the way downward Sun cannibalized itself (think Sun PS getting closed).
One of the things that Larry Ellison found utterly baffling when he acquired Sun was the way that sales commission was calculated. It was all on the value of the contract, with no requirement for the contract to be profitable.
At least in the later stages of life Sun sales folks were literally selling the company into bankruptcy. They'd find out what competitors were offering, undercut it to the degree they'd make a loss, just to get the contract and their sweet commission.
To not realise you've got this going on really speaks to how little they were interested in the actual running of the business.
> You can't deny that Sun hardware was top tier and always ahead of its time (until Oracle bought them).
From Brian Cantrill's original HN comment in 2011(https://news.ycombinator.com/item?id=2287033): "I never hesitated to tell anyone that was listening that our x86 boxes were starting to smoke the hell out of UltraSPARC". SPARC stopped being ahead of its time way before being bought by Oracle.
>> You can't deny that Sun hardware was top tier and always ahead of its time (until Oracle bought them).
> SPARC stopped being ahead of its time way before being bought by Oracle.
You seem to be focusing strictly on performance, but there were other factors: e.g., remote management is still a bit of a mess with x86. With SPARC everything can be done on the CLI (serials consoles or SSH), with x86 quite often you're still clicking on "open KVM window" for so many things. And of course many/most vendors make you purchase an 'advanced license' to use the remote KVM functionality of the LOM system.
Yeah, that was amazing. And so, I learned how to provide LILO + agetty on COM1 on every server I was admining :) So, unless there was HW fault, I could just connect to server via serial to check things out.
It's path dependence. If you built a Sun-syle x86 server with a serial console and inserted a random Linux installer CD, the installer would probably only output VGA and thus it wouldn't work.
I think the most important thing Sun got wrong was having no credible onboarding path for new clients. Why would someone invest in SPARC and Solaris when a generic x86 with Linux would do 99% of what the SPARC box would do, for far less? Why would someone who is developing their software on an x86 box with Linux care about deploying it to SPARC+Solaris? Or SPARC+Linux? Or x86+Solaris? There is a huge advantage in deploying to the same platform you develop for, which means, not having a credible desktop will hurt you in the long run.
The same applies to IBM's POWER and Z - how many green field projects are being built for them? Apart from weirdos (such as myself) who is even interested in that? And even then, I will not suggest building anything new on AIX or z/OS, or Linux on either, unless there is a really compelling (such as "we got the hardware for free", or "we really need instruction-cycle latencies to run inference and we can't match that with any other NPU") reason.
Don't get me wrong - I love using uncommon hardware with uncommon OSs (which expose bugs that are affected by endianness, or Linuxisms, or blindly assuming a string is ASCII, or that a file stays put while you use it).
> Why would someone who is developing their software on an x86 box with Linux care about ... Solaris?
There was a time when ZFS would have been a good answer for this. Of course, now it's available on FreeBSD and Linux.
At one point also there was zones, but again, containerization is pretty standard elsewhere now.
I guess also if you go further back, like the Linux 2.6 era, which is close to when Solaris was open sourced ... the 2.6 release involved a concerted effort to improve SMP support. So maybe you'd be interested more in commercial Unix in the 2.4 era, for better SMP. But again, Linux surpassed this ~20 years ago.
I’m interested in oxide, but I feel like you could level the same criticism. The startup cost of using azure or aws is almost nothing. The cost to use oxide is 6 figures?
I feel like they need to run a public cloud just to let people tinker with lightweight stuff and scale up to owning the thing.
That said, I’m also a moron about certain things, and this might be a moronic take.
> but I feel like you could level the same criticism.
Exactly. Oxide builds machines for the few companies that won't go with Azure or AWS, or that got so big Azure and AWS are no longer competitive (and it takes A LOT to be in that space, because pricing is ridiculously flexible). This is a niche market.
The question I always ask is "why would I deploy my software on your platform?". The answer is, most of the time, I wouldn't. I don't want to play platform bingo with company money.
I mean, flashback to 2010 ... zones, dtrace, smf, iscsi and zfs were very well integrated together and lightyears beyond what was available in the Linux space at the time.
Zones were lightweight, more akin to the Linux vserver project or lxc. You could make a zone on its own zfs filesystem, clone the zone (which creates a zfs clone), launch the clone, run a throw-away task, tear the zone down which also discards the zfs snapshot+clone. It was quick and cheap but also half a decade before Docker was released or popularized.
zfs integrated into other parts of the system just as well. eg: iSCSI volumes backed by a zpool and export properties configured in zfs. All of these technologies were built with the idea of being "self-healing." Meanwhile, Linux has a disparate iSCSI daemon, lvm+ext4/reiserfs, initd, and probably 20 different tools to learn if you want to grab metrics across the stack like dtrace on Solaris was capable of.
The out of band management was amazing too. Yes, some x86 hardware had ilom but like everything else in the space, it's bolted on and not truly integrated.
Sun made really nice stuff that outpaced Linux in a lot of ways. The ecosystem came from a time of competing BSDs and looked for paying customers for the engineering marvel they had created. They were used to customers figuring out how to use Unix, and the Linux crowd at the time still fit that rubric. The OS was not optimized for low-latency terminal interaction and thus "felt slow" (aka: Slowaris) but a well tuned DB on Sparc really cooked.
Finally, I went to a few Sun events and had Sun reps come and look at what we were doing. They were definitely happy to help with any kind of outreach/onboarding we needed. The issue was none of "onboarding" but was a lot of "Linux is a movement." In early times, Linux was an anti-corporate movement and some of that energy carried through to the 2010s. It's sadly ironic because Sun itself has roots in open standards and an anti-establishment mentality.
> zones, dtrace, smf, iscsi and zfs were very well integrated together and lightyears beyond what was available in the Linux space at the time.
True, but did you need that? These are all high-end (therefore low-volume) server stuff. I loved that, and their high-end systems with dozens of SPARCs, but I had one customer who had one, and everyone in my company knew they made a terrible decision - they needed a web server, and you could do more with a rack of Dell boxes for a fraction of the Ultra Enterprise [very large number] price tag. They also decided they should run an enterprise application server (they went with Dynamo) instead of Tomcat, so, again, they made a stack (maybe a directed acyclic graph) of bad decisions. I think I facepalmed so much that is the point my hair started falling.
We used it all to great success ... along with our Linux hosts. We had a stack of T1's and eventually a few thumpers (x4500's) which we ended up running a variant of OpenSolaris on. There was also some bigger iron that was humming along on Solaris 8 which provided a very stable environment (until I fat-fingered a command one day.)
Later on, we even ran an ipfilter based router from a T1 on OpenSolaris for decent IPv6 connectivity with link-aggregation+vlans when our Cisco 6509 setup was found to be inadequate for the task.
Solaris x86 launched in 1993. I had one on my desk at work in early 1997, running Common Desktop Environment.
Even back then, the CTO of our startup said Sun had missed the boat on dominating x86 Unix. They could have had a larger role in x86 Unix world if they had wanted to. It was definitely a misstep for Sun.
Having said that, throughout the 1990s Sun was amazing. They were even good right up to the Oracle purchase, although by that time RHEL on HP (or Dell) servers was eating their lunch.
As much as this resonates with me it smacks a bit too much of an Engineering vs Sales fight. Engineering blaming Sales for not selling, and Sales... where is the voice of Sales? By 2005 Sun's margins on hardware sales had to be under vicious attack from Dell and all the old incumbents. Google and Amazon were ushering in the era of commodified compute and soon offering it on demand on the cloud. Proprietary hardware platforms were under the most vigorous multi-front assault since the inception of computing.
Meanwhile I have to imagine Sun, like IBM and everyone else at the time, was noticing what incredible margins you get from software services. No factories or fabs. No supply chain logistics. Sun's proprietary hardware business should have been dying (or at least radically shifting: a product/engineering problem, not a sales one) by all market indications of the time.
So if I was a sales person and got a call from a startup for a colo's worth of machines... I probably don't have much incentive to pick up. The startup will probably fail, as most do, so the opportunity to build a relationship and sell higher margin services is probably nil. Startups hate paying for that stuff anyway.
I have incredible respect for Bryan Cantrill. He obviously knows far more about this situation and business than I do, so he's almost definitely more correct than I am...
However I would ask readers to at least be curious why sales didn't pick up the phone. Were they "bored?" Or was it a rational choice and much deeper problems with Sun's business model were the cause of their demise?
Well, definitely read Jason's blog entry that I linked to[0], which gets into the structural problems. To be clear, there were some extraordinary sales folks at Sun (as was said to me when I first came to Sun: "Remember that our sales folks are so good that they sold Solaris 2.3 on Viking!"), and I always personally got along very well with them. I was happy to be put in front of customers, and they were happy to use me -- and some of my fondest memories were with sales folks in some of the geographies which were particularly strong (shout out to Sun Canada, Sun Australia, and to the Wall Street and Itasca offices in the US).
So this isn't about the people at all (they will express the same frustration!) but about leadership that wasn't interested in running a business. I do want to stop short of overly personalizing it, but the simple truth is: Scott visited customers where they lived -- and Jonathan more or less didn't.
Ah, I love this framing of "visiting customers where they live." Definitely gets the bad taste of an eng-v-sales spat out of my mouth. Thanks for the details!
I feel like Sun's main problem was that that era was pretty much the steepest part of the Intel S curve as far as chip performance goes, and it was outpacing the older chips on fab progress alone. I'm pretty sure had something like TSMC been around then at current strength, and Sun fabbed their excellent designs there, we'd have a mirror of what Apple is doing but in server space.
> I probably don't have much incentive to pick up.
This is what Dell got right - for their sales team, this was an easy process - the product line was simple and easy to customise. Their sales reps wouldn't spend much time creating the proposals and could spend the remaining time building relationships with a lot more customers, some of which wouldn't fail.
> curious why sales didn't pick up the phone. Were they "bored?"
Unless you create incentives for bringing new customers, they'll leave the new names for last.
I don't know what Sun's strategy was in the 2000s, but doing all of the R&D to develop exotic new hardware and then focusing the sales force on software seems like they're getting the worst of both worlds.
Sure, I'm not saying that IBM's switch to services is some brilliant panacea for all proprietary hardware makers, just citing it as one of the market changes at the time that Sun seemed unprepared for.
What's interesting is that the other RISC outfit at the time, little old Acorn in the UK, did manage to adapt their proprietary chip business to the future of compute and now we all use many many ARM processors every day.
So clearly there was still a hardware route available, just again: a huge product/engineering shift.
Making workstations even in the dominant market position that Sun had, was a cottage industry compared to technology today, and it was managed that way. I wouldn't blame management for Sun being irrelevant now.
What everybody except Steve Jobs got wrong was not taking user interface seriously enough. Not even Xerox. I've used aa Xerox Alto, a Sun workstation, an LMI CADR, and a Symbolics 3600, and none of them were as carefully crafted as a Macintosh with a tiny memory and a floppy disk. Next inherited this ethic.
Jobs valued the person using the machine more than anything else about the machine.
My personal favourite UI was HyperNeWS built on Sun's NeWS.
Edit: Writing PostScript and having no discernible security probably didn't make NeWS a great fit for the world of the Internet - but it was a lot of fun.
> What everybody except Steve Jobs got wrong was not taking user interface seriously enough.
But Linux took over the world and it had a crappy UI. I was there, back when Xorg didn't even exist, running X and messing with modelines on Linux. All the window manager sucked big times.
Sun, just like SGI, got attacked by cheap beige PCs running either Windows or Linux.
I don't think their UI had much to do with their downfall and, arguably, SGI's UIs were better than the competition and they still disappeared too.
Also weren't most machines sold by Sun servers anyway? Their famous "We the dot in .com" tagline.
> But Linux took over the world and it had a crappy UI. I was there, back when Xorg didn't even exist, running X and messing with modelines on Linux. All the window manager sucked big times.
My first exposure to Solaris was in 2003 when Linux had Gnome 2.6 and Solaris had CDE. The UI on my Debian box seemed decades ahead of the Solaris machines I managed, which looked like Windows 3.1 and performed about as well.
Combined with Linux package managers being infinitely better, the GNU core utilities having so many more conveniences and options than the Solaris equivalents, I felt like Sun was this dinosaur that couldn’t die fast enough. I’ll still never understand why so many people praise Sun for its engineering culture when free software seemed so superior to me at the time.
My favorite anecdote exemplifying my opinion about Solaris at the time: you log into the console of your Sun workstation, running SunOS, with your Sun keyboard plugged in, and press backspace. It echoes ^H back to you. They couldn’t even figure out how to configure their own keyboards correctly out of the box.
Maybe in the 90s the standards were different. Maybe their tech was better. But by the time I ever used their software is was the butt of jokes.
> My favorite anecdote exemplifying my opinion about Solaris at the time: you log into the console of your Sun workstation, running SunOS, with your Sun keyboard plugged in, and press backspace. It echoes ^H back to you. They couldn’t even figure out how to configure their own keyboards correctly out of the box.
That's really an old-school UNIX convention that should have died with the move from teletypes to CRT terminals. The default behavior was that ERASE was # and KILL was @ (rather than backspace and Ctrl-C). It was typical for these to get set properly on logging in.
Ahh, ^H brings back memories. We had Solaris x86 machines in the computer lab at UC Santa Barbara when I was there from 97 to 02. They were deathly slow never mind that CDE was miserable too. I made it slightly less painful by loading the BSD versions of the utilities and using those. Pretty much gave up and ran FreeBSD on the desktop with Window Maker.
Sun got by because they had faster and more reliable hardware than what was available in the x86 space at the time. eBay's services ran on a handful of 64 CPU Starfire servers. And proprietary UNIX workstations had a bunch of specialized apps that weren't available on Linux at that point.
The vast majority of Linux installations don’t have a UI, they’re headless, and it was the same with Sun. I.E. the UI is irrelevant to the success or failure of Linux and Sun.
Linux took over the world because any licensing model that requires a per CPU license simply can’t work in a massively scalable situation like server farms and cloud.
At the end of the day, the UI on a Unix server was always shell based. That was just the best way to configure a server. (And, in some way Infrastructure as Code is an evolution of the shell scripts a unix admin would use.)
Linux won on the server side because they got just as right as Unix. The crappy UI (windows) was the one that didn't work with shell scripting, didn't work with ssh, etc.
I also recall that Linux on a PC often beat the performance of Unix equipment I could access in contemporary times. There was a lore around those machines, but my actual experience doing basic FOSS-adjacent dev work (bash, make, gcc, autoconf) was that the Linux PC was faster as well as inexpensive.
The exceptions were mostly hardware graphics accelerators in some labs, when my Linux PC had to do it in software. And I got myself a surplus DEC Multia (alpha) that was really fast with Linux for natively compiled code, beating my contemporary 486DX3-100. But, the PC was still faster for web browsing, because running Netscape on alpha meant doing x86 emulation!
Likewise, my ~90 MHz Pentium MMX laptop with Linux was faster at regular dev work on my work-supplied Sun Ultra workstation in the late 90s. (Edit: Hmm, I am not sure about this laptop anymore, it might have been 166 MHz??)
Due to my positive experiences, I was an aggressive early adopter (and frequent beta tester) of nearly every Linux ecosystem frontier for decades. E.g. 2D accelerators, 3D accelerators, SMP, large file support, gigabit ethernet, wireless ethernet, power management, virtualization, 64 bit architectures, SCSI/SATA/Firewire/e-SATA/NVME, hardware and MD-RAID, ext2/ext3/xfs/btrfs, ...
"Free", until mid 90s I guess most people bought Linux distros on CDs, later DVD (no downloads). For me the big, big plus on Linux (vs. Dos / Windows), had been all the free compilers.
I'd have to say that gcc is the unsung hero of FOSS. If you wanted compilers on your shiny new commercial-UNIX box, you either had to fork out $$$ or find a way to get the GNU toolchain into place. Even on PCs compilers tended to cost a lot, with Borland being a shining exception.
Somewhere, I still have a Red Hat 5 retail CD. I'm not sure, but I might have bought it at Borders Books rather than a computer shop. I know I bought a few CD-ROM collections at the bookstore.
to be fair, nobody in 1990 decided between NeXT and Linux, because (A) Linux didn’t exist yet and (B) they were / would have been completely different markets and totally different capabilities.
You're off by a year or two at best. Linus announced his kernel project on the Minix usenet group in Aug 1991.
It wasn't until mid 1992 that distros started to exist, and I'd say it took a bit of time for those to be complete enough to be usable for someone with a goal other than smoke-testing the kernel on their PC.
I can say for certain that I, a complete newbie, was able to boot SLS Linux floppies in early 1993, and within a couple months I was installing Slackware to HDD.
> But Linux took over the world and it had a crappy UI.
Linux took over in all the places where a good GUI didn't matter. In ran most of the Unix server software on hardware orders of magnitude cheaper than what was needed to run a commercial Unix. Windows ran on that same cheap hardware but was much more expensive for the same capability.
Despite Sun's marketing push Linux was really the dot in most of the dotcom darlings.
There's a vast gulf between Android and a Linux desktop environment. Android is a Linux kernel with the Android environment on top and is not just some Linux distro like anyone uses on a PC. ChromeOS is very similar.
I remember a web page put up by an SGI employee, saying that x86 wasn't going to destroy the Unix workstation business. The title was "Pecked To Death By Ducks".
Then Intel came out with dual-core processors, which was the new hotness. So he wrote a followup page, literally titled "Pecked To Death By Ducks With Two Bills".
At the time, I thought they were funny, and made reasonable arguments. But in retrospect, they were just trying to distract the customers from the fact that Intel was overtaking them.
Buying from Sun may have devolved into a direct-touch upsell disaster. Esp. compared to Dell which at one point was almost unique in having a working, easy-to-use web product configurator. Dell really nailed "getting products ordered and to the customer is part of the value equation"—a supply chain virtue no one else was really optimizing 1990–2005.
But selling to Sun Microsystems as a vendor was a joy. It paid bills like clockwork (net 35 IIRC). Completely showed up all the multi-$B IT competitors. They preached hyper-efficiency, but Sun delivered it in its own dealings. Unfortunately after the dot-com downfall Sun extended days payments outstanding (DPO) just like everyone else.
The Sun model of beefy server, ultra-thin client worked incredibly well and presaged the widespread adoption of virtualisation that eventually led to cloud computing and containerised software. Unfortunately that model made it very easy to undercut their prices using commodity hardware; I managed a Solaris/Sun Ray network c.2004 and it was cheaper to switch to Linux/Intel and buy spare hardware to keep in a cupboard “just in case” than to keep the Sun network going through its support contract.
I worked at Sun labs and used a Sun Ray daily for a whole summer. It worked...OK. Definitely not "Incredibly well". We had a network of a few hundred Sun Rays and two E10Ks backing it. More than a million for the servers and $1000k a piece for the Sun Rays was not cheap. Basically any $1000 laptop was more powerful, without the need for a crazy powerful Sun server. The experience of taking out my ID card, walking down the hall to another cubicle and popping it in and getting my session was magical. But I did that less than once a week, so I don't think it was overall worth it. The Sun Rays also had a pretty crappy LCD, about 1-2 years behind the SoTA.
It was a cool idea. It will keep coming back again and again. But at Sun's price points around the early 2000s, it was a loser, and I actually agree with the market in this instance. Sun Ray was a flop.
Citrix is still everywhere you need a client that can do things a web browser can’t—like run ancient VB4 software or talk to device hardware—deployed at scale.
Citrix was bought out by PE for $16.5B a few years ago, under the (sadly correct) assumption they hadn’t been charging their customers enough money for this essential tech.
Nobody knows how the future will be. There are lots of options. There are self hostable open models today - they are not as good, but some report they are good enough. Right now I don't have the hardware to try them (I could - if I'm willing to accept a small number of tokens per minute - useful means more tokens per second as I can get in an hour). However I expect in a few years I'll be able buy consumer hardware with good enough GPUs (or perhaps an AI chip???). Will anyone other than a few hobbyists buy them? time will tell.
And space launches. The Big Names will lose a ton of money and disappear as the tech commoditises.
The moat is Being Able To Do Things No One Else Can and as both hardware and software commoditise that moat disappears.
Open LLMs are already only a few months behind the frontier models and as soon as someone works out that accessible and cheap domain-specific custom training has the potential to be a gamechanger the Big Names will die.
Math puzzles only get you so far.
LLMs are in their Big Compute phase, and it's only a matter of time before that ends - because the intelligence of the frontier models isn't increasing fast enough and providing big enough benefits to make the models more valuable than cheaper, smaller competition.
I think prices will go down not up eventually and everybody will have access to it in abundance. That might make most current AI companies go down of course.
if a support contract is 70% of the cost of something, then for every 2 of something a spare is cheaper.
For Aruba I think it was 1:1; a support contract was the same price as the device itself over 2 years or something.
Support contracts never cover a "site" they always cover individual devices, so your cost scales with the amount you have... so I'm not sure I follow what you're saying..
Unless you mean to imply that you're replacing units more than once every 2 years, in which case of course you're right... but then I might consider going with another vendor entirely because that is a pretty terrible reliability record.
Support contracts never cover a "site" they always cover individual devices
Wrong. I've signed many support contracts that were negotiated basically to "one flat fee for everything we sell that you use, including future purchases[1]" for some period of time. Some people think that they 'optimize' contracts by pushing for "I just want a service contract only on the SKUs I have". That's not always a real optimization.
[1] Often, this includes a tacit "including things you are using that you didn't tell us about".
Interesting, I've literally never seen that (though obviously not doubting your experience) and I have (at one point) been in charge of 1% of web traffic..
I'm wondering if this is something uniquely american, because we don't seem to have much negotiating power in the UK/SE.
There are lots of different support contracts. Some of them really do cover the entire site. You seem to not realize the scale of a large data center. If every device lasts 5 years you are replacing 10 per day is not unheard of.
the lab I was at got on this train right around the beginning of the sparc architecture. yellow pages. nfs based swap, nfs based boot. 10Mbit thinnet. we suffered for a few months and then moved to xterms for the cheap seats (which was a much better place to cut services) and went back to buying self-sufficient workstations.
the best was having everyone with a disk cross-mount each other. any failure snowballed into a global failure.
I worked on enterprise software around the turn of the century. There was no database/os combo more bullet proof than Oracle on Solaris. Solaris was a great OS but the speed and ferocity that Linux killed Sun was something I will never forget.
I went to the Menlo Park offices to see a demo of their 64-cpu (I think) system at the time but it didn’t take long for Linux on intel killed the entire market.
I was once SUN Microsystems Campus Ambassador at Oxford University and our first lecture was about dtrace! The guy from SUN was blown away by attendance so the higher ups wanted to do more & more events in bigger rooms. But as corporate weeks passed, I started telling them to postpone because everyone was entering the exam sessions. Corporate of course did not listen. Long story short, SUN sent 3 experts but we couldn’t get the crowds and the main lecture room was embarrassingly empty.
I remember the attitude was that solaris was for serious professionals and had the highest calibre of code. For the most part it seemed true and these were solid workhorses. But, i did end up finding concurrency bugs when they introduced their multi-core integer computing behemoth.
There is a simple rule they forgot to follow: never make it difficult for customers to give you money. They had excellent hardware and software but that wasn’t enough to save them.
First work computer was Sparc IPX, a brief dalliance with HP D and K series, then had various Ultra, and servers were all sorts including E10k then various SunFire, and the Fujitsu Prime Power.
Having missed this era of computing growing up, I have a completely different view of Sun. I see Sun as this legendary old power that invented cool things like ZFS, zones, and lament that they sold themselves to Oracle. It shocks me to read about them making IBM-style mistakes.
I worked at Sun from 1995 to 2004. My take is that Sun had major structural problems all along, but as long as money streamed in from the tech bubble, these issues could be ignored. When the tech bubble ended, it was too late to fix the problems.
One big problem was that there was no cooperation inside Sun. Groups in Sun kind of viewed HP as a competitor, kind of viewed SGI as a competitor, but the real competitor to be destroyed was that other group in Sun. The leaders of the project I worked on spent most of their time in political battles, trying to acquire or outmaneuver the other teams working on similar projects.
Another problem was that the main parts of Sun had conflicting strategies. The Solaris/SunSoft division wanted to run on SPARC, x86, and Power PC for the largest market. The SPARC division, of course, wanted none of that. The hardware division wanted to make expensive servers, but the other teams saw cheaper hardware as the right direction. The market ended up moving to cheap rack-mounted Linux x86 servers, so nobody at Sun won in the end.
Right before they fell face-first into the grave, they did finally ship the v20z, which was a completely reasonable way to run Solaris x86. Its management board ran Linux on PowerPC, for irony purposes.
Forget the hardware, I feel Sun could of continued their life as the business AWS became.
Later on, when Schwartz was the CEO, they had a datacenter in a shipping container concept selling to management/IT folks. Delivering that infra via APIs to developers being the key differentiator. I'd love to hear Bryan's take since he was involved in the cloud space after he left.
Part of my maturing as I've gotten older is realizing how much business success actually comes from Sales, BizzDev, Product, etc. Engineering is great, but if you don't have people actually getting things shipped, getting customers, and talking with customers, you will eventually fail. Sure sometimes some internet companies find a money printer and get around this, but I think they are the exceptions to the rule. Sun having great tech, but having a shit sales channel, is not going to lead to success. Dell, small or large, they are going to get you someone on the phone and sell you a server.
Part of my maturing with getting older in addition to the above is that luck is a larger factor than most people are willing to admit. So I'm with you that GTM is just as important as engineering, but execution on all fronts is what tilts the luck factor. And with luck you get exceptions which is why you can get companies that seem to succeed despite themselves.
Wild that around that same time one of my fellow students at university managed to get a free trial of a Sun Thumper shipped to him. It arrived to our campus workplace on a pallet. Just to mess around with. No way was he going to actually purchase the roughly $70,000 machine.
Sun couldn’t handle the cold. Obviously in terms of the dotcom bubble burst, inobviously in terms of the company’s character. The things people loved about Sun (giving Java and NFS away, treating engineers as the point of the company, McNealy's loyalty to his people) were exactly the things that kept it from doing what it needed to survive. Very warm, very caring.
Oracle expressed the opposite. Within a year or two Larry killed OpenSolaris, let the MySQL, Hudson and OpenOffice communities fork away, sued Google over Java, and watched Gosling and most of the engineers who embodied Sun walk out.
The part that rankles the most, and undermines many a retrospective I think, is that Larry the lawnmower turned out to be a better steward of Java & MySQL than Sun was. Coldbloodedness is a survival trait.
We had a similar company in Norway in the 1980ies. Norsk Data (literally means Norwegian Computers) made computers, a multi taskin operating, a word processor thar inspired MS Word, a relational database on the same level as Oracle, the firsr web server was made by Tim Berners-Lee on one of their computers. It all went down because the managers kept insisting that they were a hardware company. The software was supposed to help selling their hardware. They were not good at making hardware.
MS Word was a descendant of Bravo and Gypsy on the Xerox Alto, not ND-NOTIS. NOTIS is an interesting footnote in the history of software, but PARC's influence on early microcomputing is well-known and recognized.
Having read a lot of these things from ex-Sun employees, I dearly want to read one with the topic my department is the one that fucked up Sun. I don't think I have seen a topic like that yet.
I actually haven't read a ton of these, so would love to be pointed to what you have found. But yes, everyone will be biased, and I am too. I think that the organization that I was in fucked up plenty (when I joined the company in 1996, SunSoft was led by truly one of the most incompetent executives I have ever encountered, a bumbling oaf that Sun kept putting in charge of different business units, letting him asphyxiate them one by one), but the fact that everyone also sees something else wrong probably tells you a few things: everyone at Sun earnestly wanted the company to succeed; Sun did a bunch of stuff right; Sun did a bunch of other stuff wrong.
Ultimately, the leadership of the company needs to bear the responsibility for the outcome -- and I think it's pretty telling that Sun's executive leadership basically went on to do more or less nothing after Sun. I loved Sun -- but it was a breeding ground for engineers, not executive leadership.
I'm reluctant to call anyone out. I liked Sun a lot, as well as the people I've encountered who worked there. There was a thread between a couple of ex-Sun guys here on hn years ago - I can't find it right now - which made me feel there was a humorous and very human pattern in Sun postmortem comments: "man, that other team really killed Sun, we were doomed after that thing they did." I was pretty curious what the people on that other team (the SPARC microprocessor designers, in the case of that comment thread) might have said in their own defense, where they might have spread some of the blame, and so on.
(perhaps it doesn't do your rather full blog post justice to compare it to something considerably less complete like a comment thread)
> Ultimately, the leadership of the company needs to bear the responsibility for the outcome
I emphatically agree. Maybe it goes without saying but with Sun, multiple generations of leadership, I think.
I seriously don't know what Sun could have done to survive - once Linux was stable, running Linux on commodity hardware was just a better solution than anything Sun could offer.
I've never spent much time with Sun equipment, but the hardware and OS had features I haven't seen on commodity hardware with Linux. IIRC, Solaris could deal with an unrecoverable ECC error by killing the process that had the bad memory. And I think they had hot swap memory on some systems.
ZFS was pretty unrivaled for long term storage for a long time. Solaris zones were containers before they were cool. Dtrace is pretty neat. Etc.
Maybe they could have survived with their OpenSolaris push driving low end adoption and high end customers buying their hardware. Pretty sure they would have had to migrate to x86, but they were dabling in that.
Maybe make some low end commodity x86 servers. Sell enterprise a fully engineered big ass database server and a bunch of commodity web servers. Doesn't have to be exceptionally priced, just good enough to let customers have single vendor server purchasing instead of getting the big stuff from sun and the little stuff from dell or hp.
More systems consulting? Sun had great engineers, but you don't see too many war stories about them fixing stuff for customers (maybe they keep quiet though).
> IIRC, Solaris could deal with an unrecoverable ECC error by killing the process that had the bad memory. And I think they had hot swap memory on some systems.
Sun clearly suffered some from being a bit too early to what would become cloud computing —- didn’t help that their tagline was “the network is the computer”, which IMO was a bit too obtuse, because I didn’t really understand what it meant (sounded like a violation of the 7-layer OSI model lol) until I saw actual cloud computing come about.
But as some sibling comments indicate they already had a lot of the software infra for distributed computing the way it eventually came to be done today, with SOA/microservices, service discovery, or serverless lambdas etc. via frameworks like EJB or JXTA. They were admittedly pretty cumbersome, like the contemporary equivalents COM/DCOM, and made the fundamental mistake of trying to abstract away the fallacies of distributed computing, but also solved most of the common problems involved.
Potentially they could have pivoted to monetizing that software infra even if it got deployed on arbitrary hardware, which may have tided them over until they could spin up their own cloud service, even if it was a fast follow to AWS. However Sun seemed committed to making their expensive servers their primary business model and relegated the software to “loss leader” complements.
If they had just looked at how services were being developed on Linux + commodity servers, they’d have seen they already had offerings in that space that could be monetized.
They should have seen commodity hardware coming and built cheaper hardware themselves. They could have. Other companies were making clone Spark machines for much less so it would have been possible.
I'm willing to bet they couldn't have done that because of the mind-boggling array of 1990s UNIX standards Solaris seemed to support. By the mid 1990s it was clear that the (platform-independent) GNU command line tools were generally superior to what was on Solaris, but presumably they had to keep all the old fossils around to satisfy their long-standing supply / support contracts with industry, defence, and government.
That was my guess back then, and it's my guess now. Someone please correct me if I'm wrong and you know for certain why.
I'm not saying it would've been easy. They could've installed the GNU tools in a different path and left it up to the user to change their preference. They did something similar with putting the BSD tools in "/usr/ucb" (I may have that path wrong.)
They could've also just built a real package manager, left it up to the community to add GNU tools. I remember the Solaris package situation being annoying, having to download crap off FTP and install manually. Or often build from source.
Solaris was pretty bare bones, out of the box. I spent days customizing it. Even early Linux distros, like Slackware, were in better shape.
With the benefit of perfect hindsight, which is a bit silly?
Open source Solaris as early as possible.
Really compete with MIPS, ARM, Motorola, etc. in the embedded space. Don't even try to make it a huge financial win, just get the SPARC architecture and dev tools out there with good Linux support.
Consistently support Solaris x86 and frequently release new generations of Intel and AMD workstation-class hardware.
Partner with TSMC for SPARC production a lot earlier.
Since we're just having fun here: acquire Nvidia in 1999, at which time it was pretty obvious workstation graphics was going to be commodified, and Sun could have bought them with their pocket change.
That doesn't exist. No department is every completely at fault. No department has enough budget to sink a company alone. It is always a lot of departments that don't turn in expected results - often for reasons not in their control.
The fault is always upper management not doing their job. Even if one department really would be that bad (which again they are not), upper management allowed them to be that bad.
I'm perceiving a lot of enthusiasts of the race to the bottom here. Yes, Dell and Linux could offer prices and convenience that undercut frontier hardware companies like Sun and SGI. But now what do we have? The loss is palpable to those of us who used to work with varied hardware competing to awe us, not to mention filling niches much more satisfactorily than one-size-fits-all.
I know OP has an intimately inside perspective, whereas what I know about Sun is only from what I read online as an admirer and developer on its platforms… but “gotten bored of the mechanics of running a business” does not match what seemed to be more about having gotten “too enterprise-y” or “too out of touch with the market” and maybe also “too hubristic.”
I interned for Sun and then worked as an MTS in Sun Labs for a couple of years. I miss Sun; I think they were very much an engineer's company that was spoiled by two things: big success in the early workstation market, and the dumb luck of the dot com boom. Sun hardware was solid (even though they fell behind the curve because of mismanagement of a couple of CPU lines), and its kernel (Solaris) was one of the well-engineered kernels ever. While any business makes some bad business decisions from time to time, one thing that was fatal was a crippling enterprise focus that failed to understand the huge growth opportunity that OSS and Linux represented. It was growth in a direction they not only didn't understand, but refused to understand, study, or adapt to.
Sun's heyday was in the boom. The fish were jumping in the boat. So naturally they were building bigger and bigger boats.
There was always the one-off ideas that might have gone here or there, but that core server cow just seemed like an endless source of cash. Thus they neglected the workstation market and ironically, the market for anything bigger than one single server rack. They didn't understand horizontal scaling, and the rise of cheap-but-replaceable consumer hardware like how Google did things was alien to them. They poo-poo'ed Linux, they killed and then unkilled Solaris x86, and a number of other things.
I do miss Sun though. As far as engineering goes, they were a very serious company full of serious engineers. Basically the polar opposite of 10X tech bros and startup culture.
Oof, the killing and subsequent unkilling of Solaris x86 in 2002 brings back very bad memories -- Sun management at its worst. It also highlighted the relative power of Sun's engineering in that everyone in Solaris knew it was the wrong decision, and we more or less disobeyed orders: during that period (January 2002 to August 2002, when it was reversed), we didn't change our treatment of x86. (And in particular, breaking x86 was grounds for a putback to be backed out.)
When Solaris x86 was unkilled (and then became increasingly important as x86 eclipsed SPARC), there was damage done that I don't think we ever fully repaired.
Agreed. When I was in the labs (2007 - 2010), my workstation was an AMD with Solaris on it, with a modern window manager and all, with IntelliJ and a great development environment. I almost didn't know it was Linux. But by then the ship had long sailed.
I worked at sun (sunsoft) right out of school near the peak. It was great. I was very excited to have all of the sources for sunos easily available to read.
Later, in grad school I went to work at a startup… google. For me it recaptured everything I liked about Sun. And had the same CEO.
What of Jini, JavaSpaces / Linda, JXTA, "grid computing"?
Were they too early? Dismissed as (Bill Joy's) hobby projects?
At the time, I was very enthusiastic. Especiall grid computing. So much so that when cloud computing arrived, I couldn't recognize the sea change. Square peg, round hole.
We even POC'd some Jini support. Demo'd for hardware manufacturers, customers, our dealer channel. Customers loved the idea.
But like UPnP, Jini failed to get traction.
As for JXTA (peer-to-peer), although I didn't gave any skin in that gane, I really expected it to vecome the norm. eg My brother (for his ISP) POC'd video delivery in anticipation. But that got mooted pretty quick.
Arguably, AWS Lambda (et al) was is "dumb" grid (run anywhere in asandbox) and the OpenClaw agentic (madness) is the "evil" incarnation. (I'm aware of efforts like Emerald AI (et al) to make a "smart" grid.)
RE: JXTA, right around the turn of the century, it felt like there was an opportunity for a globe-spanning peer-to-peer application layer, based on content-addressable storage.
DHTs—Chord paper was 2001—started popping up. The big software vendors looked at the architectural (if not legal) success of Napster & the other P2P networks and were invested in finding legitimate applications. Open standards were seeing real adoption left and right.
But it didn’t happen. Maybe simultaneously too late in the Internet’s evolution and too early.
Too late because the big kids were involved. Especially Microsoft, who would never allow any standard they couldn’t EEE to emerge. Too early because the players like Google who later could muscle through core infrastructure standards weren’t there yet.
And now it’s likely too late, with mobile embedding a client-server model with the weight of 10B devices.
when I was a student in the waning days of Sun, the SPARC/Solaris machines were impenetrable Big Iron guarded closely by the greybeards -- not to be touched by the likes of mere undergrads. Never really did get beyond noodling around with cast off gear from eBay, though I did get to enjoy a bit of it via the Sun DNA in SmartOS years later.
Our Sun rep served us well at an academic library. I had no problems provisioning a server with a nice storage array. When things fell apart at Sun our sales engineer from Sun wound up working for us.
I did get to sit in front of the consoles that served the whole school (via telnet) and even hard rebooted them a few times and I wasn't even in a techy major. Just had good relations with the head of school IT. That much exposure was enough to crave a similar computing experience and that's when I installed Linux at home right after college.
The good thing about Sun (when working at a corporate customer) was that they were very good at getting their distinguished tech people into give talks - Tim Bray, James Gosling etc.
While we're reminiscing, I never did understand why Sun paid a billion dollars for MySQL in 2008.
So far the best theory I've heard was that it was bait to make them tastier to IBM and Oracle - but even that seems kind of weak; had they really given up at that point? Does anyone have a better theory?
Heh, yeah, we're hearing that a lot. We'll find a way of getting these out to people, but we're still debating the best way to do that (running an online store isn't zero work -- but it's also not an unsolved problem!). Stay tuned!
Solaris lack of a package manager and crappy patches where patches replaced things wholesale, caused us a lot of overtime, but hey, got me a bigger house! Thanks Sun!
Oh btw their hardware was way ahead of their shit software, which wasn’t as bad as shittier IBM or Digital or HP
Yes. The software really wasn't all that great. I played a lot with Sun gear and had SGI stuff as well, the polish of the SGI machines vs the 1980's vibe that the Sun gear gave off was on a completely different level. Sun was quality hardware but the software was really a time capsule of what Unix workstations had looked like in the past.
But yeah, the lack of package manager hurt a lot, and there were competing third-party package managers which had totally different pro's/con's and packages and were incompatible with each other... what a mess.
Compared to apt/yum it was kinda comical... but the first party software really was excellent otherwise.
Not to mention the horrible installer. At that time, Debian and Red Hat installs were super-easy compared to Solaris. And Sun controlled their own hardware stack too!
I liked Sun Microsystems. And they were not that wrong because
in 2009, Oracle acquire Sun for 7.4 billion$. Granted, Sun was
perhaps more worth at that time; and most definitely before that
but it is also not a total disaster either. Except, of course,
that Sun died. And I do not like Oracle at all. It is a very
different company. Like a slow behemoth - it keeps on bulldozering
over but nobody is excited about it while it does so.
I think the big US mega-mega-corporations, while powerful and
dominating, also became super-boring. Nobody mentions how epic
Google allegedly should be, or is; in 2012 that was much more
common. People fatigued from corporate propaganda. And now the
AI slop propaganda - that one accelerates disinterest and decay.
I think Sun should probably have pushed more on its key technologies.
Simplify java, simplify Solaris while pushing on it hard - who knows
what could have happened. It is idle to want to talk about it now
though.
I had a similar experience. I forget the exact timing, but it was probably around 2002 - 2003 that Linux felt "good enough." I was at a startup that finally considered putting its "enterprise" stack on Linux.
At my employer in 1999 we were maintaining home-grown binary modification to the Solaris kernel so that TCP segment send and receive were not O(N^2) where N was the number of addresses on the interface :-/ It was, as far as I could tell, an operating system that nobody had ever tried to use before. For anything where we could avoid Sun we used FreeBSD. One of the nice things about the free Unixes of that time was that process and thread creation or destruction were orders of magnitude faster than on Sun. To me, Sun was obviously in a lower category than Linux/FreeBSD/NetBSD.
The linux kernel didn't even gain SMP until 1996. Solaris was running circles around it both before and after that point. Solaris was vastly more mature and stable.
Of course, after that, Solaris just kept running in very expensive circles while linux went on to make steady progress.
it was called "Slow arse". This was because it was rather reluctant in handing out resources (memory fe).
Having fought it, I remember the system APIs being really well defined and picky (which was good). At one point, the C compiler had some identity issues "am I 32 bit, or 64 bit?"
Yep. Solaris was the Unix for the 90's. I worked at a couple of early ISPs and users demanded Sun Solaris shell servers. At a couple of startups, we built everything on Solaris because Linux wasn't accepted in the enterprise yet.
By the early 2000's, 1) Linux was solid, why bother with Solaris? 2) ebay and other channels were flooded with used Sun hardware from failed dot-com companies anyway. No need to buy new.
I was asked to benchmark some Solaris/SPARC vs Linux/Intel hardware in around 1997, and the price-performance of the Linux gear was easily 3+ times better. There wasn't really any comparison, especially when you consider how you could buy white box PCs from a local shop, grab a Red Hat ISO for nothing, and have an installed Linux server in 24 hours, compared to weeks to get something more expensive and worse for the Sun option. Even if Sun had been more reliable, it wouldn't have mattered because you could buy 3x the PCs and keep two in a cupboard, or more likely just replace everything with more numerous and better performing Linux PCs which was what we did.
Things were going great at Dell until its wildly popular advertising campaign suddenly came tumbling down over an unfortunate indiscretion on the Lower East Side that surprised absolutely nobody...
>Monday, February 10, 2003 Posted: 3:55 PM EST (2055 GMT)
>NEW YORK (CNN) -- Benjamin Curtis, better known as the "Dell Dude" from the computer company's television commercials, was arrested Sunday while he allegedly attempted to buy marijuana on Manhattan's Lower East Side.
Steve The Dell Dude Commercial Compilation (DUDE WE'RE GETTING A DELL!)
As someone who was buying hardware in the late 1990s, it is hard to overstate the difference in buying experience between Sun or Digital (DEC) and someone like Dell. The former forced you into a live sales meeting, endless quote revisions, it was a nightmare. I remember calculating the server rails and power cords for a new Alpha server were going to cost more than a shipped/delivered Dell server that I could get the next day.
For us dell has become that. Rep changes all the time, what appears to be random pricing and long unclear delivery dates. Instead I can for example go to Thomas Kren and configure my serves online for a fraction of the cost. They have what they list in stock, delivered a few weeks later. Everything is clear and hasle free.
At this point I have the feeling dell just charges you what they think they can get out of you. You push back and the price drops, I don't want to negotiate.
The iron law is that vendors want to become like the old Sun. The only reason they don't is if they are trying to compete their way up the hill. Take away the old Sun and SGI and IBM and now Dell is Sun.
Difference is Sun was providing a "full stack" proprietary product including CPUs running a proprietary ISA, Unix OS, compiler, desktop environment and productivity apps. All that Dell does is assembling standard, commoditized parts and putting a free OS on it. Dell's business is incredibly easy to replicate, while Sun's business is extremely hard to replicate. Of course, Sun's competitive advantage turned into a curse when the components of that vertical integration become uncompetitive both features and price wise vs. the GNU/Linux-on-Intel stack. But regardless, Dell could never replicate the vertical integration Sun had.
Dell puts a fair amount of engineering into their servers (as do other main brands like HPE and Lenovo) so it's unfortunate they've lost their roots from the old mail order sales company.
If someone else is paying I'd prefer one of these big OEMs for established firmware lifecycle, spare parts and long gray market availability, shock (shipping) and thermodynamics engineering. If I'm paying, Asus or MiTAC (Tyan) are good enough.
> a "full stack" proprietary product including CPUs running a proprietary ISA, Unix OS, compiler, desktop environment and productivity apps.
this is apple today.
Really? When I first met Sun machines they were selling 68k-based Sun machines that were better specced than Apple and Amiga and such but left them dependent on Motorola’s architecture that had indirect address and thus no future.
Sure they had an in house chip development ARM team right at the time when RISC economics looked good but CPU architecture had to work harder and harder to beat the memory wall and SPARC was not competitive by 2005 and “Hail Mary” projects like CoolThreads were difficult and expensive but did not deliver for customers.
Hypothetically customers could benefit from that proprietary stuff but they benefit even more from rapid progress is competitive, commodity markets.
When I was working at the library circa 2005 we were thinking about replacing all the Windows computers that patrons used to access the online catalog (and other services) with a web browser with Sun Rays, we probably would have bought a few hundred of them and we would have bought some SPARC servers to back them up.
We could have gotten Netscape to run under Solaris but after a week of effort I wasn't able to get Mozilla to build, so no modern web browser. So no Sun Rays. No servers. DTrace and ZFS were great innovations but without the magic of "can run the software you need to run" they are just Blub. Had Sun done the work to get Mozilla running on their machines and forgotten about DTrace and ZFS people might still be using Solaris... but instead DTrace and ZFS found an audience by supporting... the commodity OS!
(arguably at that time supporting a web browser was the #1, #2 and all the way through #99 task that the Solaris team should have been working on at that time and everything else from Java to Tcl to ZFS was a dangerous distraction... and that's how companies in a position like Sun die)
They should reexamine that wish.
https://companiesmarketcap.com/sun-microsystems/stock-price-...
DELL was an innovator in this "incomparable pricing" model for normal consumers.
You'd price out a machine and then go look at comparable machines (difficult) from other vendors, then the next day your machine wouldn't be "the" configuration anymore. Instead there'd be half the ram and double the harddrive space for an extra $50, and if you swapped back to what it had been yesterday it'd cost you an extra $200.
IIRC they won business awards for this.
Where's Cartman when you need a vigorous "SHENANNIGANS!!!!"?
We sold web traffic analysis software to Dell Europe, around 1998 or so. They never kept their data more than a week. As the traffic and purchase behaviour they were analysing in the store changed so much every week they said (if I remember correctly) it didn’t make any sense to compare the data over time. I took from that that what they sold every week changed all the time. Of course, data storage and processing was then not as cheap as it is today, so maybe there was more to it.
try adding wifi, you don't get DVD. Remove DVD, you don't get wifi. It's ridiculous. Dell is a joke of a company
Did they learn this from the airlines ?
> At this point I have the feeling dell just charges you what they think they can get out of you. You push back and the price drops, I don't want to negotiate.
This is precisely the role of the sales person for something like a service, but for a physical product it is pretty egregious. There is no greater sin than leaving money on the table. There's an old story about someone offering a quote that was apparently pretty low based on reaction, so they quickly add "that was for my assistant" to give way for pushing the number higher.
I like this version: The optician quotes the price of new glasses by saying "$100", waits to see if the customer flinches, "... for the lenses", waits again, "... each.".
> so they quickly add "that was for my assistant" to give way for pushing the number higher
I don’t understand what this means? I understand that they bumped the price. But I du not understand what is meant by it being for the assistant? Care to explain?
Not OP but the way I read it...
Electrician: "Okay, so my services start at $100/hr"
<nobody bats an eye>
Electrician: "... for when my assistant can handle things."
Replace Electrician by whatever skilled professional you prefer.
Pretty much, except the assistant is never going to handle things and the updated fees go up to include you and the assistant. If you only ever send out your assistant, you never make money
I think you miss the point.
You price it at what you think works. There is no push back, so you make room to push the price higher.
You were happy to do it at the original price.
I didn't miss the point, I'm the one that posted the comment. I know precisely how it's used.
If you play poker, it is the equivalent of pushing 'portions' of your bet out at a time. You push the portion, observe their face and body language, and maybe push the rest.
The 'that was for my assistant' is just something they say to explain away the earlier quoted price. They are pretending that what they said wasn't the quote for this conversation, it was for another, different conversation, between the salesman and their assistant. This only works when the assistant is not in the call.
I wonder how effective it really is, over the long term, to generate short-term profit by consuming trust like this.
> For us dell has become that. Rep changes all the time, what appears to be random pricing and long unclear delivery dates.
I think this would be expected from a company that has acquired and absorbed EMC and Perot.
That’s exactly what enterprise sales is. Call for pricing, tell the absolute upper limit you can spend, and pay.
A nicer way to frame it is: “value based pricing”.
hire a salesperson to give them fake sales numbers and low ball them so you pay less?
> At this point I have the feeling dell just charges you what they think they can get out of you
This is how license compliance shakedowns work. Make sure your DRM is more flexible than theicensing terms, find an overage and see what you can get the customer to agree to.
I have seen senior Oracle sales reps deliberately misreading their license doc to inflate what is owed.
I think this is a case of AI ruining the hardware market. We used to almost exclusively use Dell, but now we've been forced to look elsewhere and go through validating new vendors. The lead-times are ridiculous, and these guys now only pay attention to the really big guys. Not sure I can blame them though, you've gotta do what's best for your business.
Dell was very much moving this way before modern AI explosion. The inconsistency on pricing alone depending on what rep was helping you also became a problem. I could often times call in direct and order via business line cheaper than dealing with my account rep. No reason to jump through so many hoops.
I prefer the "This is the price" model and just make it consistent. I can't wait until AI helps us remove multiple layers of middle men across Sales in general.
Just let me buy it and move on.
> I can't wait until AI helps us remove multiple layers of middle men across Sales in general.
Based on how things are moving, I think it's more likely that AI will _replace_ the layers of middle men (with layers of "middle AI"), then that those layers will be _removed_.
Of course, the process (of squeezing the customer) is the point. The implementation of that process, with either human account reps or sales bots, is irrelevant.
Not to the account reps who's pay put food on their table, vs the bots who don't have children to feed.
> I can't wait until AI helps us remove multiple layers of middle men across Sales in general.
Monkey's paw - now you have to deal with a layer of bots.
And bots don't buy lunch.
that's great as an attitude, the problem with it is that money is made up and so are prices. Business have an inventory and expenses, and hopefully they can sell enough product before their debtors come calling. There's no right price for something. Underpriced, Overpriced, the only qualifier is: is the company still in business? Sony could give Playstation 6's away for free, so long as people bought games for it, but culturally, that would look desperate, so that will have to charge money for it, when they announce it. The only thing something is worth, is what someone else is willing to pay for it.
Periodic component shortages have been a fact of life in the server business forever. Like a key supplier's factory burns down or something.
The key thing is being transparent with customers on price, availability, and delivery schedules. In the short term server vendors can use tricks and personal relationships to extract higher profit margins but eventually the senior executives at major customers figure out what's happening and start looking for other options.
> The lead-times are ridiculous, and these guys now only pay attention to the really big guys. Not sure I can blame them though, you've gotta do what's best for your business.
I don't blame them in the sense that I wish for them to be imprisoned or anything. But I wouldn't do business with them, either. I want business partners whom I can count on, not ones who will stop returning my calls as soon as a more lucrative opportunity arises.
Yeah, and then Dell more-or-less turned into that. Companies get big, customers and sales get more complicated, process gets more calcified.
SUNW would probably still be around if the whole x86 commodity pizza box revolution hadn't happened OR if they had pivoted away from SPARC earlier and went heavily into x86. Example: buying from IBM doesn't sound too different from buying from Sun, but they're still around because mainframes will be around until the Earth implodes (probably).
(I think that their channel/partner programs along with their professional services org are big reasons why Dell is still a key player in the datacenter space. You can buy Dell direct, and the sales process is still "good" as far as enterprise sales goes, but bigger players usually get their Dell kit through distributors/vendors with Dell professional services handling architecture, build-out, enablement, etc. I'm not sure how much energy SUNW invested in this, as they were before my time, but this combo is a major revenue driver for them.)
We have a couple mainframes at work, had an incident last year when someone forgot to apply a new license to one. p1, no payments happening.
I was kinda stunned; there’s a level of balls involved in selling an x millions machine and having it fail closed that I’m sort of impressed by but also utterly appalled they’d treat their big customers like that..
I mean surely to fuck it could have been an automated reminder email to the mothership and not just: sorry folks no banking for you today.
Also thanks to virtualization being originally invented in mainframes, and the OS design done by IBM (and Unisys as well), they get to run the old workloads on modern hardware, and also support various UNIX flavours.
You can even run most cool Python, Go, Java, Rust, PHP among others, on systems like z/OS.
> Sun or Digital (DEC) and someone like Dell.
In a time frame like 1999 to 2002, much more broadly than that, just the increasing gap in cost vs what you would get between buying any hardware+OS package from a legacy Unix-like vendor (Sun, DEC, SGI) and doing your own thing on a FreeBSD, OpenBSD or Linux on x86 hardware.
Dell was obviously a top choice, but by 2000 or so, Linux or FreeBSD had become good enough that you could do exactly the same thing for a fraction of the cost and run it on x86 hardware you had purchased from either Dell or one of the top 5 or 6 Dell-competitors.
As compared to, as you say, sitting through sales meetings, price quotes, revisions, a long slow and arduous process to arrive at a figure where you come to a conclusion "yeah you can spend $60,000 on these Sun servers or we can buy $20,000 of x86 servers and get creative with our own software stack to do the same thing in house".
At a certain point 1U and 2U server hardware got fast enough on boring standard x86 single or dual socket motherboards from Taiwan, in the era of like the 1.2 GHz Pentium 3 with 512KB cache, that the performance was pretty damn good and the machine cost a fraction of what the equivalent Sun or SGI rackmount machine would cost. This is in the era before a multi core x86 CPU existed. And you could be sure that its BIOS and motherboard were already OK for use with a current Linux kernel, we're talking about motherboards from Supermicro, Tyan, intel whitebox reference server boards, etc.
It's easier to see in hindsight but servers have eaten a lot of macro economic headwinds.
1999 and 2002 correspond roughly to telecom and dotcom bubbles popping. So open source operating systems were also pulled up into these positions through a lot of work and by the early 2000s it was kind of obvious this was the way forward.
2008 financial crisis really knocked the high end (RISC/UNIX/Itanium) and started the long migration to renting vs owning. In hindsight it's remarkable how much money used to just be there in the economy of all this. But I guess it still is, considering how expensive public cloud is.
> But I guess it still is, considering how expensive public cloud is.
It's not like there's less servers being sold, they're just bespoke designs from companies like the big Taiwanese ODMs that go straight into hyperscaler facilities, and we don't get to see public marketing websites and product info about them. The open compute platform stuff that's been publicly shown is like the tip of an iceberg.
I'm referencing the premium you pay to get some paltry VM in a public cloud with bizarre storage and networking that primarily make sense for the market maker versus the actual cost of the hardware. There are ways to make cloud economics work as a user, but only where staffing costs would dwarf spend (i.e. small businesses) and I've never seen them in practice.
> The open compute platform stuff that's been publicly shown is like the tip of an iceberg.
I find it a real shame it's not easier to get OCP hardware for our homelabs. Would be so much easier to operate, upgrade, and so on. And where would you even get decommissioned OCP gear?
Some of the ethernet switching top of rack 1U stuff used to show up on ebay. I think large other portions of it go into an industrial shredder/recycling process.
> other portions of it go into an industrial shredder/recycling process.
This is just criminal. Reusing is much better than recycling.
One of the first "expert systems" was one that helped you spec your DEC computer. It was easy to order a VAX that was missing necessary hardware (or had incompatible hardware).
https://en.wikipedia.org/wiki/Xcon
If you click through to the anecdote[0] mentioned in the article, the problem was not with having to contact a sales rep. In fact the startup wanted a sales rep. Sun specifically asked them to use the website:
> More than once though, the Sun people did suggest I just buy the servers off the web site.
Dell was just the only one to callback.
I don't doubt your experience but that was at the height of Sun's business, not it's nadir as this article is about.
[0] - https://fullhoffman.com/2006/03/20/the-sun-doesnt-shine-on-m...
Sun commercial sales reps, at least in Europe, were astoundingly slimy.
I had a colleague where almost his whole job consisted of beating them in sheer sliminess, and at this he was good. I don't think we ever paid 30% of the prices they'd start at, but the process of achieving this was disgusting, and there was still the sense everyone was being screwed.
So when we could move to Dell we did.
Yeah, and it wasn't just the "call for pricing" game, it was having to deal with monthly "touching base" calls which were mostly an opportunity to push corporate hospitality and other not-quite-inducements.
Whereas Dell just gave us a flat 30% discount on list pricing, and were clear in advance about what level of annual spend would trigger an increase to 35%. We'd order online and our sales rep would phone a few hours later to confirm. Other than that, they'd initiate contact maybe once a year when Dell refreshed their server or storage product lines and would offer to either send some bumpf or arrange for some demo kit.
Dealing with them was so much less hassle all round, you never got covered in slime, and you never felt like they were on the verge of offering you bribes.
I'm sure I have a memory of someone at Sun UK telling me about how they managed to run up a £30K bar bill at an event....
Yeah that fits. I was in London (though not the City) and it was definitely that sort of scene.
Thinking about it, we moved to Dell as part of the run up to IPO, and I suspect had we not done so Sun would have been much harder to deal with afterwards.
What I recall from that time is Sun were very happy they'd sold a huge pile of servers to be the core of Interlink, and a whole load of people went off to work on that.
I’ve had that experience with dell consistently over the last 6+ years. While with other vendors I can do everything over email from start to finish, Dell insists on meetings, follow-ups on topics that I said I am not interested in talking about, services that I do not need, meetings with teams I am not interested in meeting, reps that straight up skip meetings, so I have to console a junior rep that got stuck in a meeting that is above their pay grade. Different reps same slimy bs.
Forcing customers to talk to sales is actually a failure mode.
Yeah it was so much more accessible to play around with configs and pricing and then all you needed was a credit card.
Even now I think enterprise companies should be more open about what's possible and what the price and delivery date would be without having to talk/chat/email anybody. Even for something like a Juniper $100k+ device it would be kind of cool to configure it.
ordered small item from Dell 7/31, still not received.
And that is the "thing". Every failed mini company had the same exact issue, seems customers had to beg and/or go through hell to purchase equipment. Never mind that the hardware was more expensive than what you cold buy from companies like Dell.
IBM is not much better than what I hear, but somehow they keep surviving. Probably because of their mainframe business.
And Redhat[0], and professional services a la Accenture[1].
[0] https://en.wikipedia.org/wiki/IBM#2010s%E2%80%93present
[1] https://en.wikipedia.org/wiki/Accenture
Their old services arm was Kyndryl which they spun out as its own thing
I seem to remember IBM had a way to buy their stuff online, but not anymore apparently. I'm sure this was back in the System x days.
> IBM is not much better than what I hear, but somehow they keep surviving. Probably because of their mainframe business.
Imagine every bank you know going through tech modernization and maybe 70% of those are now using Red Hat OpenShift.
They own Red Hat (OpenShift kubernetes, Red Hat Enterprise Linux, Ansible Automation Platform) AND HashiCorp (Terraform, Nomad, Consul, Packer, Vault, etc.) now, on top of their previous mainframe, services gigs, global financing arm, and calling everything they sell "watson".
On the other hand, Sun and DEC hardware from then still works today, if you care to power it on. How many dell devices do?
I needed to add a bunch of capacity running Sun/Solaris back in the late-90s. Same thing: pointless meetings with sales reps without clarity in pricing. Finally, I found the gear we were looking for on eBay, bought it, drove over and picked it up in my Ford Explorer, and had it up-and-running in one day. To my surprise, people in my company were PO'ed. But I had air cover because the CEO found out and loved it.
I've written several times here about all the mistakes that Sun made during the 00s that doomed it. Among the many mistakes it made are:
- Cancelling -if briefly- Solaris on x86 in 2002. This killed Solaris in the minds of many who didn't want to be locked into Sun for SPARC.
- Failing to make a deal with Google in 2002. Apparently Sun insisted on knowing how many servers Google had, something that Google considered a high-value secret, so Sun failed to make a deal with Google, so Google ended up using Linux and contributing to Linux. This was a tremendous mind-share disaster -- it's hard to overestimate the damage done by this.
- Closing Sun PS (professional services). Bad bad move, possibly the worst of them.
- Not giving up on J2ME earlier -- it's not the sort of thing that could last forever, and Steve Jobs killed it with the iPhone. This was a case of vendor lock-in clouding Sun's decision making.
- Failure to recognize that Sun needed to become a systems company, not a CPU company.
- Failure to respond to Active Directory. This was yet another case of vendor lock-in clouding Sun's decision making: the Sun DS product team was milking their existing customers more than they wanted to go after more business with a sustainable strategy.
- UltraSPARC was more than a decade too late to make up for SPARC falling way behind x86_64. Sun needed to give up on SPARC, but again, vendor lock-in sounds sweet but turns out to be poison.
- Failure to make a deal with Apple for it to use ZFS in OS X.
- The MySQL purchase. WTF, this was horrible and stupid. The only interesting effect of this was to make Sun a target of acquisition for Oracle. But of course, it turns out that Oracle -a company built on building mind-share- had become too blinded by vendor lock-in just like Sun, so...
There were numerous other mistakes along the way. These are the most salient, for me anyway.
What's shocking is how long it took Sun to fail under those circumstances!
Also shocking is how much amazing stuff came out of Solaris engineering and the systems division!
> - UltraSPARC was more than a decade too late to make up for SPARC falling way behind x86_64. Sun needed to give up on SPARC, but again, vendor lock-in sounds sweet but turns out to be poison.
UltraSPARC came out in 1995, and Athlon 64 in 2003. I'm confused on how it could be late?
> - The MySQL purchase. WTF, this was horrible and stupid. The only interesting effect of this was to make Sun a target of acquisition for Oracle. [..]
I disagree with this assessment strongly. I believe Oracle wanted Java and only Java. All other Sun properties were collateral damage of the acquisition. It might be weird that Oracle now owns MySQL, but Oracle never cared about it. (Of course, both your and my assessment are mere opinions.)
> UltraSPARC came out in 1995, and Athlon 64 in 2003. I'm confused on how it could be late?
Ah, sorry, I was referring to the CMT architecture.
I sold my Sun stock at the internet bubble top for $70 a share. A few months later it was $7. I think of that when I see PE ratios in the hundreds for Tesla, SpaceX and other AI stocks.
Scott McNealy had a few choice words in retrospect[1].
[1] https://news.ycombinator.com/item?id=40950311
Good timing. But on the other hand if you had sold AAPL, MSFT, NVDA, ORCL, AVGO, ADI, KLAC, LRCX, etc at the Internet bubble top then you would have missed out on huge gains since then.
Your point is correct, but let's not gloss over that most of those had 80% drawdowns or more. The nasdaq took 15 years to recover and it's only been 25 years total, meaning 60% of the time since then you'd have been living with a paper loss. Timing is difficult but powerful. (Just not as powerful as 25 years of waiting.)
> […] etc at the Internet bubble top then you would have missed out on huge gains since then.
Only if you never repurchased.
Or you could DCA out: have a standard policy that no single stock can be more that x% of your portfolio, and as any one gets above that limit you crystallize profits. It's no different than having a 60/40 portfolio and rebalancing your equities if they get >60%.
I lean more towards passive and (total) market index funds, and just riding the roller coaster, as anything else is fairly either impossible or too time consuming:
* https://ofdollarsanddata.com/even-god-couldnt-beat-dollar-co...
Since we're sharing, I have fond memories of my using my the Sun thin clients at university. They were slow to start up and use for the most part, but once you got to the terminal... it was great.
Being a hacker-type, I did everything with it.
Email? Forget waiting for some bloated email client, I'm using pine.
Want some html on my personal page? Use vi.
I think they didn't take away my access for years after college, and I would sometimes login and poke around just for fun.
I do blame them, along with the rise of Java and the awful gui toolkit motif for convincing me to drop out of computer science and switch to Computational Math for my bachelors degree. I was so frustrated that our grades were dependent on learning something that I saw as completely worthless in light of the dotcom boom happening at the time. I wanted them to teach us web technologies, or at least some good system code, but all I got was broken x11 and awful performance.
I think motif was fine for its time, but definitely not much better than OpenLook.
However, 30 years ago, developing a standard SunOS app with devGuide was the most enjoyable and productive programming I've ever experienced. Simply layout your GUI, link actions to methods, generate code, fill in the abstracted methods, compile and run. I didn't have to learn X or any other bloated low level graphics library.
AFAIK, nothing has been easier, especially XCode.
I used a machine that had 8MB of RAM and a huge b&w monitor. One day I found out it would crash if your process forked more than 127 times. These things were so cool at the time but are nothing compared to a Raspberry Pi.
It's absolutely crazy to realize that even a Pi 400 running from MicroSD would smoke classic workstations of yesteryear, and yet by today's standards it's slow.
I appreciate this comment for reviving memories of pine and lynx and all of that
xtalk!
> I do blame them, along with the rise of Java ...for convincing me to drop out of computer science and switch to Computational Math for my bachelors degree.
I avoided studying CS for this reason: I saw my girlfriend's Java textbooks and thought if I had to appreciate this, I wasn't cut out to be a programmer, and stuck to network engineering. Years later, after 'Gang of Four' style OOP started dying, I have some code in a new things people use here everyday.
I was spoilt by SGI and their wonderful interface, where every detail was awesomely cool. The Sun machines were just old-fashioned at this superficial but important level of how it all looked, but only if you knew better, and that meant experience of SGI. In the world of IRIX, we had CPU-eater desktop wallpaper, meanwhile, on Sun, you had some badly drawn window decorations.
I imagine this to be the awful GUI that you speak of.
I say look and feel is superficial, however, when you have to learn a variant of UNIX with slightly different command line switches to the UNIX you know, with an interface worse than even Windows (3.11), for it to run dog slow, a lot of copium is needed. A poor interface gets in the way of what you are trying to do, a cool interface just lets you get on with it.
Back in the early 90s, I ran several procurements for university computer labs. We were under severe budget constraints, and, as a computer science department, we had constraints not normally found in such labs (for example, we needed to use certain compilers, so they had to support the workstation architectures). We wrote RFPs in accordance with university policies, and, while a number of potential bidders couldn't meet them, the companies in question all understood the constraints, and unsuccessful bidders were constructive, on the grounds of “consider us next time, ok?” All of them were like this, except one. The only time I have ever had someone threaten my job was the Sun Microsystems sales rep, after they lost a competition.
> Sun had become bored with the mechanics of running a business.
I would contend that Sun was never interested in running a business. They always cared more about building amazing technology and suffered the sales to make money.
You can't deny that Sun hardware was top tier and always ahead of its time (until Oracle bought them).
Edit: I had on rose colored glasses. I was thinking of late 90s SUN. They did in fact start getting crushed in the early 2000s.
I think this sorta forgets how slow the cpus were - I was running mostly ASIC design workloads on big Sun hardware in 2001 and we eventually got some x86 boxes that were only 32 bit but could run some of our jobs - and the single-threaded perf of these commodity machines was amazing compared to the Sparc boxes. When x86-64 came around, this was the death of all the Sun hardware in that group.
Also sometime in 2000 I had interviewed at Compaq - the old Alpha group who I think later worked on Itanium at Intel. They were constantly frustrated because they couldn't even get a build from Synopsys of the design tools they needed to run on Alpha and were forced to use Sun hardware they described as "gasping and wheezing" to design their crazy fast cpu.
Sun lost on a bunch of fronts, but I think the amazing thing is they let happen to them what they did to the other big UNIX vendors before - they were eaten from the bottom by faster cheaper hardware.
Yes, I had the same experience while using various Sun computers between 1995 and 2005. During all that decade the Sun computers were reliable, but excessively slow.
Around the beginning of 2006, I had a laptop with a 64-bit AMD Turion CPU and I could run on it Cadence and Mentor tools for ASIC design and the simulations were much faster than on a big Sun multi-processor server. Later that year our team finally got some servers with AMD Opteron, which accelerated dramatically the work in comparison with the traditional Sun servers.
I recall a fork/exec heavy shell script at a previous job that ran about an order of magnitude (~25s vs 3s) slower on Solaris/SparcV9 vs Linux/Intel P3 circa 2003. The standard retort back in those days was to rewrite it in Perl.
I can deny it. It was OK, but not amazing. I had to make pretty frequent repair calls on their gear--much more frequently than the less-technologically-amazing-but-solid-as-a-rock Dell and Compaq gear. IBM, Digital, and HP (yes, that HP, before they lost the Way) also had very good UNIX gear, but Sun had the mindshare during the '90s.
In terms of performance, even Linux 0.x on a 33Mhz 486-class processor could outperform a uniprocessor Sun/SPARC box at a fraction of the price.
I also question the sanity of anyone who thinks Sun's choice of 13W3 for a video connector was a good one. :-)
I don't know, the clustering on Solaris was pretty amazing, if we are just judging "ahead of time"
I loved Sun hardware but I'll deny that they were ahead of their time after 2003 or so. I had a boss who loved his pizza box but Intel/Linux crushed them in the 2000s on a price/performance basis.
Sun's various divisions all cared very much about their respective vendor lock-in, so they didn't innovate fast enough. The divisions that didn't have that innovated. On the way downward Sun cannibalized itself (think Sun PS getting closed).
That's the whole gist.
One of the things that Larry Ellison found utterly baffling when he acquired Sun was the way that sales commission was calculated. It was all on the value of the contract, with no requirement for the contract to be profitable.
At least in the later stages of life Sun sales folks were literally selling the company into bankruptcy. They'd find out what competitors were offering, undercut it to the degree they'd make a loss, just to get the contract and their sweet commission.
To not realise you've got this going on really speaks to how little they were interested in the actual running of the business.
> You can't deny that Sun hardware was top tier and always ahead of its time (until Oracle bought them).
From Brian Cantrill's original HN comment in 2011(https://news.ycombinator.com/item?id=2287033): "I never hesitated to tell anyone that was listening that our x86 boxes were starting to smoke the hell out of UltraSPARC". SPARC stopped being ahead of its time way before being bought by Oracle.
>> You can't deny that Sun hardware was top tier and always ahead of its time (until Oracle bought them).
> SPARC stopped being ahead of its time way before being bought by Oracle.
You seem to be focusing strictly on performance, but there were other factors: e.g., remote management is still a bit of a mess with x86. With SPARC everything can be done on the CLI (serials consoles or SSH), with x86 quite often you're still clicking on "open KVM window" for so many things. And of course many/most vendors make you purchase an 'advanced license' to use the remote KVM functionality of the LOM system.
Yeah, that was amazing. And so, I learned how to provide LILO + agetty on COM1 on every server I was admining :) So, unless there was HW fault, I could just connect to server via serial to check things out.
Why is that a property of the ISA?
It's path dependence. If you built a Sun-syle x86 server with a serial console and inserted a random Linux installer CD, the installer would probably only output VGA and thus it wouldn't work.
> You seem to be focusing strictly on performance,
Just like the vast majority of buyers do. SPARC was abandoned by customers due to poor performance.
> but there were other factors: e.g., remote management is still a bit of a mess with x86
That's a property of the systems integrator, not the ISA. Dell's iDRAC works quite well.
I think the most important thing Sun got wrong was having no credible onboarding path for new clients. Why would someone invest in SPARC and Solaris when a generic x86 with Linux would do 99% of what the SPARC box would do, for far less? Why would someone who is developing their software on an x86 box with Linux care about deploying it to SPARC+Solaris? Or SPARC+Linux? Or x86+Solaris? There is a huge advantage in deploying to the same platform you develop for, which means, not having a credible desktop will hurt you in the long run.
The same applies to IBM's POWER and Z - how many green field projects are being built for them? Apart from weirdos (such as myself) who is even interested in that? And even then, I will not suggest building anything new on AIX or z/OS, or Linux on either, unless there is a really compelling (such as "we got the hardware for free", or "we really need instruction-cycle latencies to run inference and we can't match that with any other NPU") reason.
Don't get me wrong - I love using uncommon hardware with uncommon OSs (which expose bugs that are affected by endianness, or Linuxisms, or blindly assuming a string is ASCII, or that a file stays put while you use it).
> Why would someone who is developing their software on an x86 box with Linux care about ... Solaris?
There was a time when ZFS would have been a good answer for this. Of course, now it's available on FreeBSD and Linux.
At one point also there was zones, but again, containerization is pretty standard elsewhere now.
I guess also if you go further back, like the Linux 2.6 era, which is close to when Solaris was open sourced ... the 2.6 release involved a concerted effort to improve SMP support. So maybe you'd be interested more in commercial Unix in the 2.4 era, for better SMP. But again, Linux surpassed this ~20 years ago.
I’m interested in oxide, but I feel like you could level the same criticism. The startup cost of using azure or aws is almost nothing. The cost to use oxide is 6 figures?
I feel like they need to run a public cloud just to let people tinker with lightweight stuff and scale up to owning the thing.
That said, I’m also a moron about certain things, and this might be a moronic take.
> but I feel like you could level the same criticism.
Exactly. Oxide builds machines for the few companies that won't go with Azure or AWS, or that got so big Azure and AWS are no longer competitive (and it takes A LOT to be in that space, because pricing is ridiculously flexible). This is a niche market.
The question I always ask is "why would I deploy my software on your platform?". The answer is, most of the time, I wouldn't. I don't want to play platform bingo with company money.
I mean, flashback to 2010 ... zones, dtrace, smf, iscsi and zfs were very well integrated together and lightyears beyond what was available in the Linux space at the time.
Zones were lightweight, more akin to the Linux vserver project or lxc. You could make a zone on its own zfs filesystem, clone the zone (which creates a zfs clone), launch the clone, run a throw-away task, tear the zone down which also discards the zfs snapshot+clone. It was quick and cheap but also half a decade before Docker was released or popularized.
zfs integrated into other parts of the system just as well. eg: iSCSI volumes backed by a zpool and export properties configured in zfs. All of these technologies were built with the idea of being "self-healing." Meanwhile, Linux has a disparate iSCSI daemon, lvm+ext4/reiserfs, initd, and probably 20 different tools to learn if you want to grab metrics across the stack like dtrace on Solaris was capable of.
The out of band management was amazing too. Yes, some x86 hardware had ilom but like everything else in the space, it's bolted on and not truly integrated.
Sun made really nice stuff that outpaced Linux in a lot of ways. The ecosystem came from a time of competing BSDs and looked for paying customers for the engineering marvel they had created. They were used to customers figuring out how to use Unix, and the Linux crowd at the time still fit that rubric. The OS was not optimized for low-latency terminal interaction and thus "felt slow" (aka: Slowaris) but a well tuned DB on Sparc really cooked.
Finally, I went to a few Sun events and had Sun reps come and look at what we were doing. They were definitely happy to help with any kind of outreach/onboarding we needed. The issue was none of "onboarding" but was a lot of "Linux is a movement." In early times, Linux was an anti-corporate movement and some of that energy carried through to the 2010s. It's sadly ironic because Sun itself has roots in open standards and an anti-establishment mentality.
> zones, dtrace, smf, iscsi and zfs were very well integrated together and lightyears beyond what was available in the Linux space at the time.
True, but did you need that? These are all high-end (therefore low-volume) server stuff. I loved that, and their high-end systems with dozens of SPARCs, but I had one customer who had one, and everyone in my company knew they made a terrible decision - they needed a web server, and you could do more with a rack of Dell boxes for a fraction of the Ultra Enterprise [very large number] price tag. They also decided they should run an enterprise application server (they went with Dynamo) instead of Tomcat, so, again, they made a stack (maybe a directed acyclic graph) of bad decisions. I think I facepalmed so much that is the point my hair started falling.
We used it all to great success ... along with our Linux hosts. We had a stack of T1's and eventually a few thumpers (x4500's) which we ended up running a variant of OpenSolaris on. There was also some bigger iron that was humming along on Solaris 8 which provided a very stable environment (until I fat-fingered a command one day.)
Later on, we even ran an ipfilter based router from a T1 on OpenSolaris for decent IPv6 connectivity with link-aggregation+vlans when our Cisco 6509 setup was found to be inadequate for the task.
This. Sun behaved like it had vendor lock-in. Who wants a vendor like that?!
Solaris x86 launched in 1993. I had one on my desk at work in early 1997, running Common Desktop Environment.
Even back then, the CTO of our startup said Sun had missed the boat on dominating x86 Unix. They could have had a larger role in x86 Unix world if they had wanted to. It was definitely a misstep for Sun.
Having said that, throughout the 1990s Sun was amazing. They were even good right up to the Oracle purchase, although by that time RHEL on HP (or Dell) servers was eating their lunch.
As much as this resonates with me it smacks a bit too much of an Engineering vs Sales fight. Engineering blaming Sales for not selling, and Sales... where is the voice of Sales? By 2005 Sun's margins on hardware sales had to be under vicious attack from Dell and all the old incumbents. Google and Amazon were ushering in the era of commodified compute and soon offering it on demand on the cloud. Proprietary hardware platforms were under the most vigorous multi-front assault since the inception of computing.
Meanwhile I have to imagine Sun, like IBM and everyone else at the time, was noticing what incredible margins you get from software services. No factories or fabs. No supply chain logistics. Sun's proprietary hardware business should have been dying (or at least radically shifting: a product/engineering problem, not a sales one) by all market indications of the time.
So if I was a sales person and got a call from a startup for a colo's worth of machines... I probably don't have much incentive to pick up. The startup will probably fail, as most do, so the opportunity to build a relationship and sell higher margin services is probably nil. Startups hate paying for that stuff anyway.
I have incredible respect for Bryan Cantrill. He obviously knows far more about this situation and business than I do, so he's almost definitely more correct than I am...
However I would ask readers to at least be curious why sales didn't pick up the phone. Were they "bored?" Or was it a rational choice and much deeper problems with Sun's business model were the cause of their demise?
Well, definitely read Jason's blog entry that I linked to[0], which gets into the structural problems. To be clear, there were some extraordinary sales folks at Sun (as was said to me when I first came to Sun: "Remember that our sales folks are so good that they sold Solaris 2.3 on Viking!"), and I always personally got along very well with them. I was happy to be put in front of customers, and they were happy to use me -- and some of my fondest memories were with sales folks in some of the geographies which were particularly strong (shout out to Sun Canada, Sun Australia, and to the Wall Street and Itasca offices in the US).
So this isn't about the people at all (they will express the same frustration!) but about leadership that wasn't interested in running a business. I do want to stop short of overly personalizing it, but the simple truth is: Scott visited customers where they lived -- and Jonathan more or less didn't.
[0] https://fullhoffman.com/2006/03/20/the-sun-doesnt-shine-on-m...
Ah, I love this framing of "visiting customers where they live." Definitely gets the bad taste of an eng-v-sales spat out of my mouth. Thanks for the details!
I feel like Sun's main problem was that that era was pretty much the steepest part of the Intel S curve as far as chip performance goes, and it was outpacing the older chips on fab progress alone. I'm pretty sure had something like TSMC been around then at current strength, and Sun fabbed their excellent designs there, we'd have a mirror of what Apple is doing but in server space.
> I probably don't have much incentive to pick up.
This is what Dell got right - for their sales team, this was an easy process - the product line was simple and easy to customise. Their sales reps wouldn't spend much time creating the proposals and could spend the remaining time building relationships with a lot more customers, some of which wouldn't fail.
> curious why sales didn't pick up the phone. Were they "bored?"
Unless you create incentives for bringing new customers, they'll leave the new names for last.
I don't know what Sun's strategy was in the 2000s, but doing all of the R&D to develop exotic new hardware and then focusing the sales force on software seems like they're getting the worst of both worlds.
Sure, I'm not saying that IBM's switch to services is some brilliant panacea for all proprietary hardware makers, just citing it as one of the market changes at the time that Sun seemed unprepared for.
What's interesting is that the other RISC outfit at the time, little old Acorn in the UK, did manage to adapt their proprietary chip business to the future of compute and now we all use many many ARM processors every day.
So clearly there was still a hardware route available, just again: a huge product/engineering shift.
Making workstations even in the dominant market position that Sun had, was a cottage industry compared to technology today, and it was managed that way. I wouldn't blame management for Sun being irrelevant now.
What everybody except Steve Jobs got wrong was not taking user interface seriously enough. Not even Xerox. I've used aa Xerox Alto, a Sun workstation, an LMI CADR, and a Symbolics 3600, and none of them were as carefully crafted as a Macintosh with a tiny memory and a floppy disk. Next inherited this ethic.
Jobs valued the person using the machine more than anything else about the machine.
My personal favourite UI was HyperNeWS built on Sun's NeWS.
Edit: Writing PostScript and having no discernible security probably didn't make NeWS a great fit for the world of the Internet - but it was a lot of fun.
> What everybody except Steve Jobs got wrong was not taking user interface seriously enough.
But Linux took over the world and it had a crappy UI. I was there, back when Xorg didn't even exist, running X and messing with modelines on Linux. All the window manager sucked big times.
Sun, just like SGI, got attacked by cheap beige PCs running either Windows or Linux.
I don't think their UI had much to do with their downfall and, arguably, SGI's UIs were better than the competition and they still disappeared too.
Also weren't most machines sold by Sun servers anyway? Their famous "We the dot in .com" tagline.
> But Linux took over the world and it had a crappy UI. I was there, back when Xorg didn't even exist, running X and messing with modelines on Linux. All the window manager sucked big times.
My first exposure to Solaris was in 2003 when Linux had Gnome 2.6 and Solaris had CDE. The UI on my Debian box seemed decades ahead of the Solaris machines I managed, which looked like Windows 3.1 and performed about as well.
Combined with Linux package managers being infinitely better, the GNU core utilities having so many more conveniences and options than the Solaris equivalents, I felt like Sun was this dinosaur that couldn’t die fast enough. I’ll still never understand why so many people praise Sun for its engineering culture when free software seemed so superior to me at the time.
My favorite anecdote exemplifying my opinion about Solaris at the time: you log into the console of your Sun workstation, running SunOS, with your Sun keyboard plugged in, and press backspace. It echoes ^H back to you. They couldn’t even figure out how to configure their own keyboards correctly out of the box.
Maybe in the 90s the standards were different. Maybe their tech was better. But by the time I ever used their software is was the butt of jokes.
> My favorite anecdote exemplifying my opinion about Solaris at the time: you log into the console of your Sun workstation, running SunOS, with your Sun keyboard plugged in, and press backspace. It echoes ^H back to you. They couldn’t even figure out how to configure their own keyboards correctly out of the box.
That's really an old-school UNIX convention that should have died with the move from teletypes to CRT terminals. The default behavior was that ERASE was # and KILL was @ (rather than backspace and Ctrl-C). It was typical for these to get set properly on logging in.
Ahh, ^H brings back memories. We had Solaris x86 machines in the computer lab at UC Santa Barbara when I was there from 97 to 02. They were deathly slow never mind that CDE was miserable too. I made it slightly less painful by loading the BSD versions of the utilities and using those. Pretty much gave up and ran FreeBSD on the desktop with Window Maker.
Sun got by because they had faster and more reliable hardware than what was available in the x86 space at the time. eBay's services ran on a handful of 64 CPU Starfire servers. And proprietary UNIX workstations had a bunch of specialized apps that weren't available on Linux at that point.
The vast majority of Linux installations don’t have a UI, they’re headless, and it was the same with Sun. I.E. the UI is irrelevant to the success or failure of Linux and Sun.
Linux took over the world because any licensing model that requires a per CPU license simply can’t work in a massively scalable situation like server farms and cloud.
At the end of the day, the UI on a Unix server was always shell based. That was just the best way to configure a server. (And, in some way Infrastructure as Code is an evolution of the shell scripts a unix admin would use.)
Linux won on the server side because they got just as right as Unix. The crappy UI (windows) was the one that didn't work with shell scripting, didn't work with ssh, etc.
IaC is really just declarative shell scripts. It's a different language, but it's often just kicked off by some bash in a git action.
Once you have git and actions, I don't think that the language really matters all that much.
Linux was free, and people will put up with a lot of shit for that.
I also recall that Linux on a PC often beat the performance of Unix equipment I could access in contemporary times. There was a lore around those machines, but my actual experience doing basic FOSS-adjacent dev work (bash, make, gcc, autoconf) was that the Linux PC was faster as well as inexpensive.
The exceptions were mostly hardware graphics accelerators in some labs, when my Linux PC had to do it in software. And I got myself a surplus DEC Multia (alpha) that was really fast with Linux for natively compiled code, beating my contemporary 486DX3-100. But, the PC was still faster for web browsing, because running Netscape on alpha meant doing x86 emulation!
Likewise, my ~90 MHz Pentium MMX laptop with Linux was faster at regular dev work on my work-supplied Sun Ultra workstation in the late 90s. (Edit: Hmm, I am not sure about this laptop anymore, it might have been 166 MHz??)
Due to my positive experiences, I was an aggressive early adopter (and frequent beta tester) of nearly every Linux ecosystem frontier for decades. E.g. 2D accelerators, 3D accelerators, SMP, large file support, gigabit ethernet, wireless ethernet, power management, virtualization, 64 bit architectures, SCSI/SATA/Firewire/e-SATA/NVME, hardware and MD-RAID, ext2/ext3/xfs/btrfs, ...
Exactly. Free vs. $8k (in 1990 dollars) for a NeXT computer was a pretty easy choice for most people.
"Free", until mid 90s I guess most people bought Linux distros on CDs, later DVD (no downloads). For me the big, big plus on Linux (vs. Dos / Windows), had been all the free compilers.
I'd have to say that gcc is the unsung hero of FOSS. If you wanted compilers on your shiny new commercial-UNIX box, you either had to fork out $$$ or find a way to get the GNU toolchain into place. Even on PCs compilers tended to cost a lot, with Borland being a shining exception.
Somewhere, I still have a Red Hat 5 retail CD. I'm not sure, but I might have bought it at Borders Books rather than a computer shop. I know I bought a few CD-ROM collections at the bookstore.
Linux books that included a CD-ROM with the subject distro and software were pretty common in that era.
to be fair, nobody in 1990 decided between NeXT and Linux, because (A) Linux didn’t exist yet and (B) they were / would have been completely different markets and totally different capabilities.
True. I think linux maybe had just been released but there was nothing like a "distro" that was easy to install.
Minix existed at the time too, I know because I ran it on an 8088 PC.
But within a few years, linux started gaining mindshare and was starting to replace proprietary unix especially on individual PCs.
You're off by a year or two at best. Linus announced his kernel project on the Minix usenet group in Aug 1991.
It wasn't until mid 1992 that distros started to exist, and I'd say it took a bit of time for those to be complete enough to be usable for someone with a goal other than smoke-testing the kernel on their PC.
I can say for certain that I, a complete newbie, was able to boot SLS Linux floppies in early 1993, and within a couple months I was installing Slackware to HDD.
Yeah, 1993 is when things really took off. First SLS, then Slackware, and using Slackware as the build environment: Debian.
You also had SuSE, which really started as German-localized Slackware.
> and it had a crappy UI.
That wasn't even true back then.
> But Linux took over the world and it had a crappy UI.
Linux took over in all the places where a good GUI didn't matter. In ran most of the Unix server software on hardware orders of magnitude cheaper than what was needed to run a commercial Unix. Windows ran on that same cheap hardware but was much more expensive for the same capability.
Despite Sun's marketing push Linux was really the dot in most of the dotcom darlings.
> Linux took over in all the places where a good GUI didn't matter.
like smartphones ;-)
There's a vast gulf between Android and a Linux desktop environment. Android is a Linux kernel with the Android environment on top and is not just some Linux distro like anyone uses on a PC. ChromeOS is very similar.
I remember a web page put up by an SGI employee, saying that x86 wasn't going to destroy the Unix workstation business. The title was "Pecked To Death By Ducks".
Then Intel came out with dual-core processors, which was the new hotness. So he wrote a followup page, literally titled "Pecked To Death By Ducks With Two Bills".
At the time, I thought they were funny, and made reasonable arguments. But in retrospect, they were just trying to distract the customers from the fact that Intel was overtaking them.
Buying from Sun may have devolved into a direct-touch upsell disaster. Esp. compared to Dell which at one point was almost unique in having a working, easy-to-use web product configurator. Dell really nailed "getting products ordered and to the customer is part of the value equation"—a supply chain virtue no one else was really optimizing 1990–2005.
But selling to Sun Microsystems as a vendor was a joy. It paid bills like clockwork (net 35 IIRC). Completely showed up all the multi-$B IT competitors. They preached hyper-efficiency, but Sun delivered it in its own dealings. Unfortunately after the dot-com downfall Sun extended days payments outstanding (DPO) just like everyone else.
The Sun model of beefy server, ultra-thin client worked incredibly well and presaged the widespread adoption of virtualisation that eventually led to cloud computing and containerised software. Unfortunately that model made it very easy to undercut their prices using commodity hardware; I managed a Solaris/Sun Ray network c.2004 and it was cheaper to switch to Linux/Intel and buy spare hardware to keep in a cupboard “just in case” than to keep the Sun network going through its support contract.
I worked at Sun labs and used a Sun Ray daily for a whole summer. It worked...OK. Definitely not "Incredibly well". We had a network of a few hundred Sun Rays and two E10Ks backing it. More than a million for the servers and $1000k a piece for the Sun Rays was not cheap. Basically any $1000 laptop was more powerful, without the need for a crazy powerful Sun server. The experience of taking out my ID card, walking down the hall to another cubicle and popping it in and getting my session was magical. But I did that less than once a week, so I don't think it was overall worth it. The Sun Rays also had a pretty crappy LCD, about 1-2 years behind the SoTA.
It was a cool idea. It will keep coming back again and again. But at Sun's price points around the early 2000s, it was a loser, and I actually agree with the market in this instance. Sun Ray was a flop.
"It will keep coming back again and again."
It lived on in the world of Windows Terminal Services and things like Citrix - maybe still does in some places.
Citrix is still everywhere you need a client that can do things a web browser can’t—like run ancient VB4 software or talk to device hardware—deployed at scale.
Citrix was bought out by PE for $16.5B a few years ago, under the (sadly correct) assumption they hadn’t been charging their customers enough money for this essential tech.
It lives on in the world of LLMs in a way.
Stadia? GeForce Now? Xbox Cloud Gaming? :-)
Chromebooks…
$1000k ?
Is this how the whole LLM thing will go down once providers raise prices?
Maybe - ask me in 20 years...
Nobody knows how the future will be. There are lots of options. There are self hostable open models today - they are not as good, but some report they are good enough. Right now I don't have the hardware to try them (I could - if I'm willing to accept a small number of tokens per minute - useful means more tokens per second as I can get in an hour). However I expect in a few years I'll be able buy consumer hardware with good enough GPUs (or perhaps an AI chip???). Will anyone other than a few hobbyists buy them? time will tell.
And space launches. The Big Names will lose a ton of money and disappear as the tech commoditises.
The moat is Being Able To Do Things No One Else Can and as both hardware and software commoditise that moat disappears.
Open LLMs are already only a few months behind the frontier models and as soon as someone works out that accessible and cheap domain-specific custom training has the potential to be a gamechanger the Big Names will die.
Math puzzles only get you so far.
LLMs are in their Big Compute phase, and it's only a matter of time before that ends - because the intelligence of the frontier models isn't increasing fast enough and providing big enough benefits to make the models more valuable than cheaper, smaller competition.
I think prices will go down not up eventually and everybody will have access to it in abundance. That might make most current AI companies go down of course.
it is often cheaper to buy a spare of something and throw it in a cupboard rather than have a support contract.
I have a bunch of Aruba switches and this is what we do, because it's a lot cheaper.
If you are only buying a single spare that is the case. Support contracts make sense when you would want several dozen spares.
if a support contract is 70% of the cost of something, then for every 2 of something a spare is cheaper.
For Aruba I think it was 1:1; a support contract was the same price as the device itself over 2 years or something.
Support contracts never cover a "site" they always cover individual devices, so your cost scales with the amount you have... so I'm not sure I follow what you're saying..
Unless you mean to imply that you're replacing units more than once every 2 years, in which case of course you're right... but then I might consider going with another vendor entirely because that is a pretty terrible reliability record.
Support contracts never cover a "site" they always cover individual devices
Wrong. I've signed many support contracts that were negotiated basically to "one flat fee for everything we sell that you use, including future purchases[1]" for some period of time. Some people think that they 'optimize' contracts by pushing for "I just want a service contract only on the SKUs I have". That's not always a real optimization.
[1] Often, this includes a tacit "including things you are using that you didn't tell us about".
Interesting, I've literally never seen that (though obviously not doubting your experience) and I have (at one point) been in charge of 1% of web traffic..
I'm wondering if this is something uniquely american, because we don't seem to have much negotiating power in the UK/SE.
I'm guessing it is the size of your company or datacenter. You have to be throwing around millions of dollars to get the opened ended contracts.
There are lots of different support contracts. Some of them really do cover the entire site. You seem to not realize the scale of a large data center. If every device lasts 5 years you are replacing 10 per day is not unheard of.
Remember, "The Network is the Computer".
Which is why it is kind of ironic having folks discussing microservices as something groundbreaking.
Counterpoint: it actually sucked. Entire universities lost a day's work when their editors were all fatally stuck in an rpc.lockd bug.
the lab I was at got on this train right around the beginning of the sparc architecture. yellow pages. nfs based swap, nfs based boot. 10Mbit thinnet. we suffered for a few months and then moved to xterms for the cheap seats (which was a much better place to cut services) and went back to buying self-sufficient workstations.
the best was having everyone with a disk cross-mount each other. any failure snowballed into a global failure.
Java applets!
I worked on enterprise software around the turn of the century. There was no database/os combo more bullet proof than Oracle on Solaris. Solaris was a great OS but the speed and ferocity that Linux killed Sun was something I will never forget.
I went to the Menlo Park offices to see a demo of their 64-cpu (I think) system at the time but it didn’t take long for Linux on intel killed the entire market.
I was once SUN Microsystems Campus Ambassador at Oxford University and our first lecture was about dtrace! The guy from SUN was blown away by attendance so the higher ups wanted to do more & more events in bigger rooms. But as corporate weeks passed, I started telling them to postpone because everyone was entering the exam sessions. Corporate of course did not listen. Long story short, SUN sent 3 experts but we couldn’t get the crowds and the main lecture room was embarrassingly empty.
As an outsider at the time, when Linux was taking off, there seemed to be a lot of people asking why they wouldn't Open Source Solaris.
It seemed to me like they couldn't decide whether they were a hardware company or a software company.
Mind you, I was young at the time so possibly didn't have the best understanding of the situation. Seems like they left the decision too late though.
Dismissing Linux as a hobbiest OS definitely turned out badly.
I remember the attitude was that solaris was for serious professionals and had the highest calibre of code. For the most part it seemed true and these were solid workhorses. But, i did end up finding concurrency bugs when they introduced their multi-core integer computing behemoth.
There is a simple rule they forgot to follow: never make it difficult for customers to give you money. They had excellent hardware and software but that wasn’t enough to save them.
HN lift of James Gosling's story about the radioactive UltraSparc II
https://news.ycombinator.com/item?id=10638929
First work computer was Sparc IPX, a brief dalliance with HP D and K series, then had various Ultra, and servers were all sorts including E10k then various SunFire, and the Fujitsu Prime Power.
They were quite fun.
But can confirm, buying Sun was PITA.
Having missed this era of computing growing up, I have a completely different view of Sun. I see Sun as this legendary old power that invented cool things like ZFS, zones, and lament that they sold themselves to Oracle. It shocks me to read about them making IBM-style mistakes.
I worked at Sun from 1995 to 2004. My take is that Sun had major structural problems all along, but as long as money streamed in from the tech bubble, these issues could be ignored. When the tech bubble ended, it was too late to fix the problems.
One big problem was that there was no cooperation inside Sun. Groups in Sun kind of viewed HP as a competitor, kind of viewed SGI as a competitor, but the real competitor to be destroyed was that other group in Sun. The leaders of the project I worked on spent most of their time in political battles, trying to acquire or outmaneuver the other teams working on similar projects.
Another problem was that the main parts of Sun had conflicting strategies. The Solaris/SunSoft division wanted to run on SPARC, x86, and Power PC for the largest market. The SPARC division, of course, wanted none of that. The hardware division wanted to make expensive servers, but the other teams saw cheaper hardware as the right direction. The market ended up moving to cheap rack-mounted Linux x86 servers, so nobody at Sun won in the end.
Right before they fell face-first into the grave, they did finally ship the v20z, which was a completely reasonable way to run Solaris x86. Its management board ran Linux on PowerPC, for irony purposes.
Forget the hardware, I feel Sun could of continued their life as the business AWS became.
Later on, when Schwartz was the CEO, they had a datacenter in a shipping container concept selling to management/IT folks. Delivering that infra via APIs to developers being the key differentiator. I'd love to hear Bryan's take since he was involved in the cloud space after he left.
Part of my maturing as I've gotten older is realizing how much business success actually comes from Sales, BizzDev, Product, etc. Engineering is great, but if you don't have people actually getting things shipped, getting customers, and talking with customers, you will eventually fail. Sure sometimes some internet companies find a money printer and get around this, but I think they are the exceptions to the rule. Sun having great tech, but having a shit sales channel, is not going to lead to success. Dell, small or large, they are going to get you someone on the phone and sell you a server.
Part of my maturing with getting older in addition to the above is that luck is a larger factor than most people are willing to admit. So I'm with you that GTM is just as important as engineering, but execution on all fronts is what tilts the luck factor. And with luck you get exceptions which is why you can get companies that seem to succeed despite themselves.
Wild that around that same time one of my fellow students at university managed to get a free trial of a Sun Thumper shipped to him. It arrived to our campus workplace on a pallet. Just to mess around with. No way was he going to actually purchase the roughly $70,000 machine.
Sun couldn’t handle the cold. Obviously in terms of the dotcom bubble burst, inobviously in terms of the company’s character. The things people loved about Sun (giving Java and NFS away, treating engineers as the point of the company, McNealy's loyalty to his people) were exactly the things that kept it from doing what it needed to survive. Very warm, very caring.
Oracle expressed the opposite. Within a year or two Larry killed OpenSolaris, let the MySQL, Hudson and OpenOffice communities fork away, sued Google over Java, and watched Gosling and most of the engineers who embodied Sun walk out.
The part that rankles the most, and undermines many a retrospective I think, is that Larry the lawnmower turned out to be a better steward of Java & MySQL than Sun was. Coldbloodedness is a survival trait.
I worked there right before they sold to Oracle for a bit—my impression was a lot of brilliant engineers but no unified product story.
Sun was just never really setup to sell lots of small servers. You have different pipeline when your goal is to sell million dollar E10K servers.
I got to see a server installed via crane because they couldn't go up in that building's elevator.
We had a similar company in Norway in the 1980ies. Norsk Data (literally means Norwegian Computers) made computers, a multi taskin operating, a word processor thar inspired MS Word, a relational database on the same level as Oracle, the firsr web server was made by Tim Berners-Lee on one of their computers. It all went down because the managers kept insisting that they were a hardware company. The software was supposed to help selling their hardware. They were not good at making hardware.
MS Word was a descendant of Bravo and Gypsy on the Xerox Alto, not ND-NOTIS. NOTIS is an interesting footnote in the history of software, but PARC's influence on early microcomputing is well-known and recognized.
[1] https://en.wikipedia.org/wiki/Xerox_Alto#Software
[2] https://en.wikipedia.org/wiki/Bravo_(editor)
[3] https://en.wikipedia.org/wiki/Gypsy_(software)
[4] https://en.wikipedia.org/wiki/ND-NOTIS
Berners-Lee made the first web server true but it was on a NeXT not Norsk Data
Having read a lot of these things from ex-Sun employees, I dearly want to read one with the topic my department is the one that fucked up Sun. I don't think I have seen a topic like that yet.
I actually haven't read a ton of these, so would love to be pointed to what you have found. But yes, everyone will be biased, and I am too. I think that the organization that I was in fucked up plenty (when I joined the company in 1996, SunSoft was led by truly one of the most incompetent executives I have ever encountered, a bumbling oaf that Sun kept putting in charge of different business units, letting him asphyxiate them one by one), but the fact that everyone also sees something else wrong probably tells you a few things: everyone at Sun earnestly wanted the company to succeed; Sun did a bunch of stuff right; Sun did a bunch of other stuff wrong.
Ultimately, the leadership of the company needs to bear the responsibility for the outcome -- and I think it's pretty telling that Sun's executive leadership basically went on to do more or less nothing after Sun. I loved Sun -- but it was a breeding ground for engineers, not executive leadership.
I'm reluctant to call anyone out. I liked Sun a lot, as well as the people I've encountered who worked there. There was a thread between a couple of ex-Sun guys here on hn years ago - I can't find it right now - which made me feel there was a humorous and very human pattern in Sun postmortem comments: "man, that other team really killed Sun, we were doomed after that thing they did." I was pretty curious what the people on that other team (the SPARC microprocessor designers, in the case of that comment thread) might have said in their own defense, where they might have spread some of the blame, and so on.
(perhaps it doesn't do your rather full blog post justice to compare it to something considerably less complete like a comment thread)
> Ultimately, the leadership of the company needs to bear the responsibility for the outcome
I emphatically agree. Maybe it goes without saying but with Sun, multiple generations of leadership, I think.
I seriously don't know what Sun could have done to survive - once Linux was stable, running Linux on commodity hardware was just a better solution than anything Sun could offer.
I've never spent much time with Sun equipment, but the hardware and OS had features I haven't seen on commodity hardware with Linux. IIRC, Solaris could deal with an unrecoverable ECC error by killing the process that had the bad memory. And I think they had hot swap memory on some systems.
ZFS was pretty unrivaled for long term storage for a long time. Solaris zones were containers before they were cool. Dtrace is pretty neat. Etc.
Maybe they could have survived with their OpenSolaris push driving low end adoption and high end customers buying their hardware. Pretty sure they would have had to migrate to x86, but they were dabling in that.
Maybe make some low end commodity x86 servers. Sell enterprise a fully engineered big ass database server and a bunch of commodity web servers. Doesn't have to be exceptionally priced, just good enough to let customers have single vendor server purchasing instead of getting the big stuff from sun and the little stuff from dell or hp.
More systems consulting? Sun had great engineers, but you don't see too many war stories about them fixing stuff for customers (maybe they keep quiet though).
> IIRC, Solaris could deal with an unrecoverable ECC error by killing the process that had the bad memory. And I think they had hot swap memory on some systems.
Both of these are supported on Linux now. The kill-on-ECC is handled by hwpoison: https://www.kernel.org/doc/html/v5.1/vm/hwpoison.html and the hot-swappable RAM has been present for quite a bit longer.
zfs + solaris zones were amazing
Sun clearly suffered some from being a bit too early to what would become cloud computing —- didn’t help that their tagline was “the network is the computer”, which IMO was a bit too obtuse, because I didn’t really understand what it meant (sounded like a violation of the 7-layer OSI model lol) until I saw actual cloud computing come about.
But as some sibling comments indicate they already had a lot of the software infra for distributed computing the way it eventually came to be done today, with SOA/microservices, service discovery, or serverless lambdas etc. via frameworks like EJB or JXTA. They were admittedly pretty cumbersome, like the contemporary equivalents COM/DCOM, and made the fundamental mistake of trying to abstract away the fallacies of distributed computing, but also solved most of the common problems involved.
Potentially they could have pivoted to monetizing that software infra even if it got deployed on arbitrary hardware, which may have tided them over until they could spin up their own cloud service, even if it was a fast follow to AWS. However Sun seemed committed to making their expensive servers their primary business model and relegated the software to “loss leader” complements.
If they had just looked at how services were being developed on Linux + commodity servers, they’d have seen they already had offerings in that space that could be monetized.
They should have seen commodity hardware coming and built cheaper hardware themselves. They could have. Other companies were making clone Spark machines for much less so it would have been possible.
sure, they shoulda woulda coulda, but isn't that the innovator's dilemma?
Yes. There were lots of other things they could have done.
Early Linux distros were very rough. Remember messing around with modelines in X11 configs?
Sun could've been the "Apple" of x86 Unix: it just works.
I'm willing to bet they couldn't have done that because of the mind-boggling array of 1990s UNIX standards Solaris seemed to support. By the mid 1990s it was clear that the (platform-independent) GNU command line tools were generally superior to what was on Solaris, but presumably they had to keep all the old fossils around to satisfy their long-standing supply / support contracts with industry, defence, and government.
That was my guess back then, and it's my guess now. Someone please correct me if I'm wrong and you know for certain why.
I'm not saying it would've been easy. They could've installed the GNU tools in a different path and left it up to the user to change their preference. They did something similar with putting the BSD tools in "/usr/ucb" (I may have that path wrong.)
that would have required them to swallow their pride, and admit GNU tools were in demand, which didn't fit the marketing narrative.
They could've also just built a real package manager, left it up to the community to add GNU tools. I remember the Solaris package situation being annoying, having to download crap off FTP and install manually. Or often build from source.
Solaris was pretty bare bones, out of the box. I spent days customizing it. Even early Linux distros, like Slackware, were in better shape.
With the benefit of perfect hindsight, which is a bit silly?
Open source Solaris as early as possible.
Really compete with MIPS, ARM, Motorola, etc. in the embedded space. Don't even try to make it a huge financial win, just get the SPARC architecture and dev tools out there with good Linux support.
Consistently support Solaris x86 and frequently release new generations of Intel and AMD workstation-class hardware.
Partner with TSMC for SPARC production a lot earlier.
Since we're just having fun here: acquire Nvidia in 1999, at which time it was pretty obvious workstation graphics was going to be commodified, and Sun could have bought them with their pocket change.
Exactly, I used desktop Sun machines and administered many very large Sun machines from 1997 through to 2010.
I saw the end coming for Sun when I had a dx2/66 at home running Linux in the late 90s. It was brilliant.
That doesn't exist. No department is every completely at fault. No department has enough budget to sink a company alone. It is always a lot of departments that don't turn in expected results - often for reasons not in their control.
The fault is always upper management not doing their job. Even if one department really would be that bad (which again they are not), upper management allowed them to be that bad.
I'm perceiving a lot of enthusiasts of the race to the bottom here. Yes, Dell and Linux could offer prices and convenience that undercut frontier hardware companies like Sun and SGI. But now what do we have? The loss is palpable to those of us who used to work with varied hardware competing to awe us, not to mention filling niches much more satisfactorily than one-size-fits-all.
I know OP has an intimately inside perspective, whereas what I know about Sun is only from what I read online as an admirer and developer on its platforms… but “gotten bored of the mechanics of running a business” does not match what seemed to be more about having gotten “too enterprise-y” or “too out of touch with the market” and maybe also “too hubristic.”
I interned for Sun and then worked as an MTS in Sun Labs for a couple of years. I miss Sun; I think they were very much an engineer's company that was spoiled by two things: big success in the early workstation market, and the dumb luck of the dot com boom. Sun hardware was solid (even though they fell behind the curve because of mismanagement of a couple of CPU lines), and its kernel (Solaris) was one of the well-engineered kernels ever. While any business makes some bad business decisions from time to time, one thing that was fatal was a crippling enterprise focus that failed to understand the huge growth opportunity that OSS and Linux represented. It was growth in a direction they not only didn't understand, but refused to understand, study, or adapt to.
Sun's heyday was in the boom. The fish were jumping in the boat. So naturally they were building bigger and bigger boats.
There was always the one-off ideas that might have gone here or there, but that core server cow just seemed like an endless source of cash. Thus they neglected the workstation market and ironically, the market for anything bigger than one single server rack. They didn't understand horizontal scaling, and the rise of cheap-but-replaceable consumer hardware like how Google did things was alien to them. They poo-poo'ed Linux, they killed and then unkilled Solaris x86, and a number of other things.
I do miss Sun though. As far as engineering goes, they were a very serious company full of serious engineers. Basically the polar opposite of 10X tech bros and startup culture.
Oof, the killing and subsequent unkilling of Solaris x86 in 2002 brings back very bad memories -- Sun management at its worst. It also highlighted the relative power of Sun's engineering in that everyone in Solaris knew it was the wrong decision, and we more or less disobeyed orders: during that period (January 2002 to August 2002, when it was reversed), we didn't change our treatment of x86. (And in particular, breaking x86 was grounds for a putback to be backed out.)
When Solaris x86 was unkilled (and then became increasingly important as x86 eclipsed SPARC), there was damage done that I don't think we ever fully repaired.
Agreed. When I was in the labs (2007 - 2010), my workstation was an AMD with Solaris on it, with a modern window manager and all, with IntelliJ and a great development environment. I almost didn't know it was Linux. But by then the ship had long sailed.
I worked at sun (sunsoft) right out of school near the peak. It was great. I was very excited to have all of the sources for sunos easily available to read.
Later, in grad school I went to work at a startup… google. For me it recaptured everything I liked about Sun. And had the same CEO.
It indeed was a fantastic place, doors nailed to the walls, trying out the newly fledged Mapreduce and Charlie's. Good times. Good times.
Great perspective to share, thanks.
> They didn't understand horizontal scaling
What of Jini, JavaSpaces / Linda, JXTA, "grid computing"?
Were they too early? Dismissed as (Bill Joy's) hobby projects?
At the time, I was very enthusiastic. Especiall grid computing. So much so that when cloud computing arrived, I couldn't recognize the sea change. Square peg, round hole.
We even POC'd some Jini support. Demo'd for hardware manufacturers, customers, our dealer channel. Customers loved the idea.
But like UPnP, Jini failed to get traction.
As for JXTA (peer-to-peer), although I didn't gave any skin in that gane, I really expected it to vecome the norm. eg My brother (for his ISP) POC'd video delivery in anticipation. But that got mooted pretty quick.
Arguably, AWS Lambda (et al) was is "dumb" grid (run anywhere in asandbox) and the OpenClaw agentic (madness) is the "evil" incarnation. (I'm aware of efforts like Emerald AI (et al) to make a "smart" grid.)
RE: JXTA, right around the turn of the century, it felt like there was an opportunity for a globe-spanning peer-to-peer application layer, based on content-addressable storage.
DHTs—Chord paper was 2001—started popping up. The big software vendors looked at the architectural (if not legal) success of Napster & the other P2P networks and were invested in finding legitimate applications. Open standards were seeing real adoption left and right.
But it didn’t happen. Maybe simultaneously too late in the Internet’s evolution and too early.
Too late because the big kids were involved. Especially Microsoft, who would never allow any standard they couldn’t EEE to emerge. Too early because the players like Google who later could muscle through core infrastructure standards weren’t there yet.
And now it’s likely too late, with mobile embedding a client-server model with the weight of 10B devices.
when I was a student in the waning days of Sun, the SPARC/Solaris machines were impenetrable Big Iron guarded closely by the greybeards -- not to be touched by the likes of mere undergrads. Never really did get beyond noodling around with cast off gear from eBay, though I did get to enjoy a bit of it via the Sun DNA in SmartOS years later.
Our Sun rep served us well at an academic library. I had no problems provisioning a server with a nice storage array. When things fell apart at Sun our sales engineer from Sun wound up working for us.
I did get to sit in front of the consoles that served the whole school (via telnet) and even hard rebooted them a few times and I wasn't even in a techy major. Just had good relations with the head of school IT. That much exposure was enough to crave a similar computing experience and that's when I installed Linux at home right after college.
The good thing about Sun (when working at a corporate customer) was that they were very good at getting their distinguished tech people into give talks - Tim Bray, James Gosling etc.
Very expensive and valuable information to come by, thanks for sharing.
I wonder if bcantrill will ever get back into cloud hosting ... to revive his Joyent days (which he alludes to without naming directly in the post).
I mean in a way isn't that what Oxide is doing? Aren't they just providing you the hardware to do cloud hosting yourself on prem?
On-prem vs cloud though, is two very different things.
In the later, I don't have to worry about hardware/power/networking/etc at all.
As someone that buys hardware today, all the hardware companies are now in the "we don't want to do business with you" territory.
How timely, I was just thinking about Sun after doing some office rearranging last week and running into a bunch of sun branded ethernet cables.
While we're reminiscing, I never did understand why Sun paid a billion dollars for MySQL in 2008.
So far the best theory I've heard was that it was bait to make them tastier to IBM and Oracle - but even that seems kind of weak; had they really given up at that point? Does anyone have a better theory?
> all of these homage shirts are amazing
You can't just say that without sharing those too!
Brian, I know you lurk and post here but i'd love to see those designs make their way to the store so the general public can purchase them.
Heh, yeah, we're hearing that a lot. We'll find a way of getting these out to people, but we're still debating the best way to do that (running an online store isn't zero work -- but it's also not an unsolved problem!). Stay tuned!
Awesome. I'm not in a position at my company to move the needle much in terms of buying, but I would love to support you guys any way I can.
Solaris lack of a package manager and crappy patches where patches replaced things wholesale, caused us a lot of overtime, but hey, got me a bigger house! Thanks Sun! Oh btw their hardware was way ahead of their shit software, which wasn’t as bad as shittier IBM or Digital or HP
Yes. The software really wasn't all that great. I played a lot with Sun gear and had SGI stuff as well, the polish of the SGI machines vs the 1980's vibe that the Sun gear gave off was on a completely different level. Sun was quality hardware but the software was really a time capsule of what Unix workstations had looked like in the past.
I still feel like we're catching up to SMF.
But yeah, the lack of package manager hurt a lot, and there were competing third-party package managers which had totally different pro's/con's and packages and were incompatible with each other... what a mess.
Compared to apt/yum it was kinda comical... but the first party software really was excellent otherwise.
Not to mention the horrible installer. At that time, Debian and Red Hat installs were super-easy compared to Solaris. And Sun controlled their own hardware stack too!
Yea he’s wrong. Plenty of people switched to dell cause of Linux and open source . In our case the switch was from alpha and vms
Where can one buy the Oxide t-shirt
I liked Sun Microsystems. And they were not that wrong because in 2009, Oracle acquire Sun for 7.4 billion$. Granted, Sun was perhaps more worth at that time; and most definitely before that but it is also not a total disaster either. Except, of course, that Sun died. And I do not like Oracle at all. It is a very different company. Like a slow behemoth - it keeps on bulldozering over but nobody is excited about it while it does so.
I think the big US mega-mega-corporations, while powerful and dominating, also became super-boring. Nobody mentions how epic Google allegedly should be, or is; in 2012 that was much more common. People fatigued from corporate propaganda. And now the AI slop propaganda - that one accelerates disinterest and decay.
I think Sun should probably have pushed more on its key technologies. Simplify java, simplify Solaris while pushing on it hard - who knows what could have happened. It is idle to want to talk about it now though.
Solaris was worse than Linux, and SPARC was worse than Intel. A lot worse.
Even ZFS was way better integrated with the Linux kernel even while Solaris was supposedly maintained.
Not in 2000, at my employer back then, our Linux server was only good as MP3 storage exposed via Samba/NFS.
All development and production deployments was being done across Solaris, Aix and HP-UX in regards to UNIX flavours.
I had a similar experience. I forget the exact timing, but it was probably around 2002 - 2003 that Linux felt "good enough." I was at a startup that finally considered putting its "enterprise" stack on Linux.
Later in 2003, when I joined CERN, there were still Sun workstations being used, and I was an earlier user of Scientific Linux in 2004.
Afterwards Nokia Networks were still deploying into HP-UX, and the first big Linux project started around late 2005, to Red-Hat.
At my employer in 1999 we were maintaining home-grown binary modification to the Solaris kernel so that TCP segment send and receive were not O(N^2) where N was the number of addresses on the interface :-/ It was, as far as I could tell, an operating system that nobody had ever tried to use before. For anything where we could avoid Sun we used FreeBSD. One of the nice things about the free Unixes of that time was that process and thread creation or destruction were orders of magnitude faster than on Sun. To me, Sun was obviously in a lower category than Linux/FreeBSD/NetBSD.
The linux kernel didn't even gain SMP until 1996. Solaris was running circles around it both before and after that point. Solaris was vastly more mature and stable.
Of course, after that, Solaris just kept running in very expensive circles while linux went on to make steady progress.
it was called "Slow arse". This was because it was rather reluctant in handing out resources (memory fe). Having fought it, I remember the system APIs being really well defined and picky (which was good). At one point, the C compiler had some identity issues "am I 32 bit, or 64 bit?"
Eh, depends when. In 2005, sure. Not so much in the 1990s
Yep. Solaris was the Unix for the 90's. I worked at a couple of early ISPs and users demanded Sun Solaris shell servers. At a couple of startups, we built everything on Solaris because Linux wasn't accepted in the enterprise yet.
By the early 2000's, 1) Linux was solid, why bother with Solaris? 2) ebay and other channels were flooded with used Sun hardware from failed dot-com companies anyway. No need to buy new.
I was asked to benchmark some Solaris/SPARC vs Linux/Intel hardware in around 1997, and the price-performance of the Linux gear was easily 3+ times better. There wasn't really any comparison, especially when you consider how you could buy white box PCs from a local shop, grab a Red Hat ISO for nothing, and have an installed Linux server in 24 hours, compared to weeks to get something more expensive and worse for the Sun option. Even if Sun had been more reliable, it wouldn't have mattered because you could buy 3x the PCs and keep two in a cupboard, or more likely just replace everything with more numerous and better performing Linux PCs which was what we did.
Assuming that the PC was actually supported.
I had to survive with 800x600 on a 1024x768 monitor for quite some time.
lol. at first i thought this is going to be about what Sun Tzu got wrong and i was ready to refute.
Speaking of another Steven, the Dell Dude:
Things were going great at Dell until its wildly popular advertising campaign suddenly came tumbling down over an unfortunate indiscretion on the Lower East Side that surprised absolutely nobody...
https://edition.cnn.com/2003/LAW/02/10/deli.dude.arrest/
>'Dell Dude' arrested for marijuana possession
>Monday, February 10, 2003 Posted: 3:55 PM EST (2055 GMT)
>NEW YORK (CNN) -- Benjamin Curtis, better known as the "Dell Dude" from the computer company's television commercials, was arrested Sunday while he allegedly attempted to buy marijuana on Manhattan's Lower East Side.
Steve The Dell Dude Commercial Compilation (DUDE WE'RE GETTING A DELL!)
https://www.youtube.com/watch?v=oJxTHMygpDY
Love that t-shirt. Does it have the Meta logo on the back of it? /s
that right
Sun? Sun Microsystem?