Hacker News Reader

Comment by bedobi | Hacker News Reader

> Apparently Southwest Airlines’ ingenious strategy of never upgrading from Windows 3.1 allowed it to remain unscathed.

this is pretty damning both ways

on the one hand, it's insane, unfathomable and inconceivable that anyone can run anything critical on windows 3.1 (!!!)

on the other hand, it's equally insane, unfathomable and inconceivable that those who do are actually better off - 30 years of "progress" is actually just bs? what are we as an industry "even doing here"???? is computing actually a solved problem and we're really just mostly reinventing the wheel and enshittifying perfectly already working systems?

View on HN

From my memory, this wildly circulating Windows 3.1 quote is inaccurate. The software they were running was compared to running something like Windows 3.1, but it wasn't actually running on Windows 3.1, as far as I understand.

Edit: https://kotaku.com/southwest-airlines-windows-3-1-blue-scree...

Obviously Southwest is not using Windows 3.1, and you should probably be thinking hard about the trustworthiness of any outlet or article that repeats that claim: https://www.osnews.com/story/140301/no-southwest-airlines-is...

Thirty years of progress is still progress for managing the additional complexity that modern software needs.

A lot of software doesn't need that additional complexity. Having thirty year old software, if properly sequestered (since there are security holes large enough to fly a 737 through), means that this is software that has been working for three decades. It has issues (as the mess they had previously showed), but Southwest appears to be able to be able to manage that to some degree without needing to incur the additional complexity of managing a modern software stack for application software that doesn't need it.

The ability to play minesweeper on critical computing equipment without impacting it isn't necessarily a desirable feature. Having the computer boot in five seconds and run the desired application is.

And there are a number of ways to handle that ... running old operating systems is one of the ways. Space Force S02E07 is not a desirable situation https://youtu.be/xDLvUqhwHZc . You could also have a kuberentes cluster with multiple replicas and load balancing and all of that additional complexity that takes more people to be able to manage without any real gains in what the application itself is doing.

> Having the computer boot in five seconds

There are certainly more modern options that allow that and it's highly doubtful that specifically is particularly relevant for Southwest.

Not that there is any evidence that they're actually using 3.1 for anything?

> that this is software that has been working for three decades

Or it's so buggy or designed (or more likely updated) so poorly that everyone is afraid to touch it. e.g. I doubt there are many (even any?) practical reasons for airlines to use GDS besides the cost and complexity involved in designing an entirely new system and somehow forcing all other airlines to switch to it?

> Not that there is any evidence that they're actually using 3.1 for anything?

Windows 3.1? No. I'd even say there's no evidence that windows 95 is being used but rather that they've got what appears to be some old software with older design.

https://www.dallasnews.com/business/airlines/2022/12/30/what...

> 2. The crew scheduling system is the main culprit.

> Southwest uses internally built and maintained systems called SkySolver and Crew Web Access for pilots and flight attendants. They can sign on to those systems to pick flights and then make changes when flights are canceled or delayed or when there is an illness.

> “Southwest has generated systems internally themselves instead of using more standard programs that others have used,” Montgomery said. “Some systems even look historic like they were designed on Windows 95.”

Screen shots of this are in http://www3.alpa.org/LinkClick.aspx?fileticket=IO7kd%2Bfm2Do...

Unfortunately, I don't know the nuances of Microsoft Windows UI well enough to be able to pick out which OS version is running the software in those screen shots.

---

> Or it's so buggy or designed (or more likely updated) so poorly that everyone is afraid to touch it.

That is a very common occurrence (I'm dealing with that now ... a .jar file that hasn't been rebuilt in 15 years). The big rewrite is something that comes with one part excitement (I can do it right this time!) and dread (oh my, that's how much code that I need to retest?!).

I was involved in the tail end of a 3 year project at one company with some software that replaced previously running DOS (and yes, it was DOS - they had an C and assembly guru employed who's job it was to remove / optimize code in the binary to get it to fit into 640k) to a Java Web Start (which was a neat technology) and the millions of lines of software that monstrosity had and needed to be debugged and fixed.

While they're in a better spot now (can use modern hardware), and its something that they can build in house (a major part of the reason to do it was to drop the external contractor who didn't like maintaining the C code) ... but that also came with the added complexity of the software that they licensed and the maintenance and deployment of that software. Before they could put the software on a floppy and have it shipped to each location ... now its a big bigger and more complex of a deployment (that was built with duct tape and chewing gum one night to do diff deployments of specific class files rather than trying to push the entirety down the pipe for each location).

My rambling point is that we are moving forward with complexity - and that allows us to manage more complex situations, but it comes at the cost of managing that additional complexity of the infrastructure and software it needs and that cost is ongoing and not always taken into account.

Those screenshots look more like WinXP to me with the rounded and shaded button elements. It’s the boring grey and buttons people presumably associate with the 90s.

Most times its less about a system being poorly designed and more about it being able to solve very hard problems which most existing employees today haven't even heard of. Institutional knowledge plays big time on these decisions.

It was a "troll tweet" says the guy who started the rumor:

https://x.com/ArtemR/status/1815408553131426179

You don't want to know what OS Sabre (backend for 30% of world's airlines) is using on their mainframes.

actually, I do :) is it DOS? some IBM mainframe OS? do tell

Was DOS actually every used on mainframes/servers on a significant scale? (genuine question, not saying it wasn't)

A mainframe OS called DOS was in fact quite popular, but it’s not the same thing as the DOS that was in PCs. (There were others, too, like Apple ][ DOS. As soon as your computer gets the capability of attaching a disk drive, somebody has to write a Disk Operating System.)

I do actually. My last job had a mainframe team maintaining (and adding to) an AS/400 application. They still had punchcard programs.

They had json apis. Each one had some variation on parsing http from a raw tcp connection with IBM RPG. I had to do some unspeakable things to a ruby library so I could control the order of the headers.

Looks like it was IBM System/360 mainframes but they've recently migrated to google hosted services.

Well, what actual new features does Windows 11 give you compared to Windows 3.1?

It will support a huge number of new chips, new peripherals, more memory, and so on.

If I'm running Southwest's crew scheduling software, how much of that do I care about? Do I care that it will now support the latest Bluetooth? Do I care that it now has the same UI as tablets? Do I care that it has better ads to display on the start menu? No, no, and no.

The only thing might be more memory. (I mean, the UI might not look like it belonged in the Stone Age, so that's something, I guess...)

There hasn't been a real fundamental improvement in the functionality of OSes since Windows 3.1. It's all been device support (including new classes of devices), new CPU support, and new UI styles. (The security improvements in Windows were a legitimately big deal, but those were fixing what was broken, not adding new functionality.)

And I'm sure that, having said this, someone is going to point out something really important that I forgot...

> If I'm running Southwest's crew scheduling software, how much of that do I care about?

Not a lot of if you can't/don't want to upgrade or replace that software to make sure it runs on modern OSes. I'm sure that for the most part that software is causing various unnecessary issues and decreasing potential productivity at least to some extent. Just look at GDS, they love to get rid of that, but that would require a coordinated effort and extensive collaboration between all major airlines which is somewhat tricky.

> There hasn't been a real fundamental improvement in the functionality of OSes since Windows 3.1.

Multitasking? A massive amount of other important features that matter if you want to build new software or significantly improve what you're using now.

Also you seem to be downplaying security a but too much? Those devices would need to be carefully isolated from everything else (not that as I understand there is any evidence that Southwest Airlines is actually using 3.1?).

> but those were fixing what was broken, not adding new functionality

It's like saying that every new feature in introduced in any type of software that wasn't entirely novel was actually fixing stuff that was broken and wasn't "new". I guess some would apply to the claim GUI/desktop wasn't something new but it was just "fixing" (inherently "broken") command line interfaces?

Being able to design significantly objectively better (based on how much it could increase productivity) is I guess is not strictly tied to the OS at least in some cases. But it certainly make it a lot cheaper/easier.

Multitasking. Yeah, I'll give you that. That is actually a huge step up.

I'm presuming that Southwest's internal software backends are not internet exposed, which is why I'm downplaying security.

Well all you'd really need to do to avoid this outage is not run auto-updating proprietary kernel modules in the early anti-malware environment. Bare Windows 11 would have been fine - the problem was Crowdstrike.

> is computing actually a solved problem and we're really just mostly reinventing the wheel and enshittifying perfectly already working systems?

On a whim I tried playing Solitaire on windows the other day. You know, that game that’s shipped with windows since forever. Well, it’s horrible now. When I tried firing it up, it first spent several minutes downloading software updates. Then it loaded in some horrible “casual games bundle” app which felt laggy like a web app - complete with Xbox cloud sync for my progress, and daily achievements and other junk.

The game used to run flawlessly on my old 486. My computer now is orders of magnitude more powerful - but solitaire feels laggy. I bet the entirety of windows XP is smaller than the “update” it performed to install solitaire.

I have a personal theory that there’s always something that gets the attention of the best engineers. Decades ago it was human interface guidelines and UI toolkits. Today it’s LLMs and AAA game engines. Most of the rest of the software in the world is worked on by the B team. And they don’t blink an eye at the idea of rewriting solitaire for windows on top of electron. If JavaScript is all their team knows, so be it. Heaven forbid we have to learn how to properly build software for windows.

>Heaven forbid we have to learn how to properly build software for windows

They already are properly building software for Windows, and your Solitaire app is a good example of it. It's much, much better than it used to be: it's laggy and slow, and downloads a bunch of extra crap, which has the potential of getting you to spend more money. In short, changing from the old and simple Solitaire to this bloated mess makes Microsoft more profitable, and customers like you keep using Windows regardless, so why shouldn't they do this?

Meanwhile, on my Linux box, my simpler games haven't changed substantially in years, if not decades (perhaps updates for the newer GUI toolkits being used), and still work great. But people prefer to stick with Windows and then complain about things being too bloated.

> which has the potential of getting you to spend more money.

I don't think any of the extra stuff is monetised. I don't know; I didn't look closely at it.

> customers like you keep using Windows regardless ... But people prefer to stick with Windows and then complain about things being too bloated.

"Stick with windows"? I run windows, macos, linux and freebsd at home. I want all of them to be better. It pains me to see any of the great operating systems of our time fall slowly toward mediocrity.

Of course I complain about it. Windows should be better than this.

>I don't think any of the extra stuff is monetised. I don't know; I didn't look closely at it.

If they weren't making money out those changes somehow, they likely wouldn't be dedicating resources to making those changes.

>Of course I complain about it. Windows should be better than this.

Why should it? It's not yours, it's Microsoft's, and it exists solely so their shareholders can extract money from customers. It's not there to be whatever you think an OS should be. They're doing things to it that they believe will improve their profits, and that's all Windows needs. As long as MS makes changes to Windows that improve profits, that by definition makes Windows "better".

It’s obviously not that simple.

Microsoft makes more money when the broader community wants to use their products. Users are stakeholders.

You’re right in part - Microsoft’s incentives aren’t perfectly aligned with mine. But they aren’t in opposition either. We’re both better off when Microsoft makes their products better for me such that I give them more money. Microsoft will try things on - of course! But if the broad community hates what they’re doing, they will stop doing it. They must, if they want to remain in business.

Which is a round about way of saying, yes of course windows isn’t “mine”. But as I’m a paying customer, my opinion still matters - or at least, the opinion of their customers in aggregate.

When you’re a customer in a situation like this, you have at least 2 choices: Voice and Exit. In this case, exit means dumping windows. And voice - at least how I’m exercising it - means making a bit of a stink about it in a forum like HN where microsoft engineers sometimes visit.

If you don’t think this is the forum for that, feel free to downvote my comments. But in the meantime, it seems mighty strange of you to try and convince me that my opinions don’t matter and I should… what exactly? Be quiet and take it instead? Provide no feedback and quietly migrate to another equally imperfect ecosystem?

I’m not sure how I would be helped by being smaller in the world.

I think your analysis is highly flawed.

>Microsoft makes more money when the broader community wants to use their products. Users are stakeholders.

>But if the broad community hates what they’re doing, they will stop doing it. They must, if they want to remain in business.

This mostly isn't true. MS customers are going to use Windows no matter what, unless MS somehow makes it so completely unusable they're forced to abandon it. So they can make whatever annoying changes they want, and their customers aren't going anywhere.

However, this isn't true of everything MS makes, so you have to be careful not to conflate Windows with anything else. If people hate Teams enough, they might be convinced to switch to Zoom or whatever. This just isn't the case with Windows. They're free to piss people off as much as they want here.

>But as I’m a paying customer, my opinion still matters - or at least, the opinion of their customers in aggregate.

No, it really doesn't. You aren't paying for a Windows subscription, you paid for a Windows license. You probably paid it when you bought your PC. Or if you're a big company, you probably have some sort of site license, because you made the decision to use MS products across your organization. And if you're BigCo, you probably don't have Solitaire installed on employee computers anyway. So MS making the solitaire game slow and shitty isn't going to change the amount of money they get from you: they already got your money when you bought your PC. Even worse, you probably didn't have much choice: there aren't a lot of no-OS computers unless you assembled yours from parts, and those that exist (or come with Linux pre-installed) usually cost more, because MS isn't making money from you so much, but from kickbacks from all the crapware that's pre-installed.

>...a forum like HN where microsoft engineers sometimes visit. >it seems mighty strange of you to try and convince me that my opinions don’t matter and I should… what exactly?

What makes you think MS engineers have any sway at all over the user experience? That stuff is decided by upper management, product managers, etc. And they don't care about you; they only care about the company's profitability. Adding more crapware and ads into Windows makes it more profitable, so that's what makes sense for them to do. Making a bloat-free, high-performance OS without any annoying ads or other crap doesn't make them more money, because you and everyone else are going to keep using Windows (and buying new computers with it pre-installed) no matter what.

>it seems mighty strange of you to try and convince me that my opinions don’t matter and I should… what exactly? Be quiet and take it instead?

Basically yes: you're wasting your keystrokes complaining about Windows enshittification, because your opinion really doesn't matter. It's not like MacOS, where someone has to actively want to use a Mac to go buy one, or Linux where you not only have to actively be willing to buck the trend, but also choose which of dozens of distros you want to use. With Windows, it's just the default choice for 90% of users, and they're not going to switch to something else, which MS has found out over the course of decades now.

>Provide no feedback and quietly migrate to another equally imperfect ecosystem?

Other ecosystems are far more likely to take your complaints seriously and to worry about your user experience.

I hear what you're saying; I just think your worldview is too cynical. The world isn't full of middle managers micromanaging engineers and laughing all the way to their overflowing bank accounts. People want money - sure. But people also want to do good work. If the world was really as dismal as you think it is, microsoft would destaff the entire windows team and just let the ecosystem rot while reaping in their locked in revenue.

Thats not what happens.

Windows is worked on constantly. And most of the changes are clearly done with the intent of making the product better. Things like IO completion ports. C# and the .NET ecosystem. Support for big/little (P/E) cores in the kernel. Edge replacing IE. ARM support. And so on. Of course like any big company, they make plenty of bad choices - like telemetry, the "watch everything you do" AI assistant, and so on. But its too simple to paint any large company with the "evil megacorp" paintbrush. Microsoft has 200 000 employees. I'm sure some of them are exactly who you think they are. But some - I suspect most - of them really want to do a good job and make products their customers want to use.

Just like any big company.

> Other ecosystems are far more likely to take your complaints seriously and to worry about your user experience.

Are they though?

Parts of Apple are great. And other parts are obviously horrible, extractive and greedy. Did you know companies can't use the NFC chip in Apple phones - in my phone, which I've already paid for - without being paid millions of dollars for the privilege? I'd love it if I could pay for public transit in my city using my phone, which I own, but my government doesn't want to pay the extortionate price apple is charging to bless them with that capability. Horrible.

Companies aren't good guys or bad guys. They're just big groups of people. And people are complex, and they have a wide variety of incentives and drives. Calling companies out for bad behaviour and sloppy work is important - even though it often has no effect. "Killed by google" is infamous inside google. Microsoft rolled back their AI assistant thing after the bad press.

If you're looking for great products made by companies who are ethically spotless, well, there aren't a lot options. Linux on the desktop is pretty decent these days. But I still use windows for gaming. And I'm typing this on a mac.

>The world isn't full of middle managers micromanaging engineers and laughing all the way to their overflowing bank accounts.

Now this isn't what I think either. It's the top executives mainly that drive this stuff, since they set the direction for the company. Middle managers are just pawns, except for the ones who are empire-builders trying to work their way up. Also, I'm not claiming that all companies are this bad, but I think MS is a unique company, and Windows a unique product, because of its monopoly status. Other companies really can't afford to piss off their customers too much because they'll just jump ship. MS doesn't have this problem with Windows. Remember how paranoid they were about Linux back in the early 00s? Now they don't seem to care at all, and I think it's because they figured out they had nothing to be worried about after all: almost no one was going to stop buying or using Windows (or even if they did stop using it, they still bought a license with their PC anyway; the addition of pre-loaded crapware also changed the economics here so they weren't worried about actual license fees from individuals).

>If the world was really as dismal as you think it is, microsoft would destaff the entire windows team and just let the ecosystem rot while reaping in their locked in revenue.

Of course, they can't just go to this extreme. They'd keep taking in money, but things would go down eventually, and they wouldn't grow revenues either, and remember, Wall Street wants never-ending growth. And as you point out, the ecosystem is important (as Apple has shown); they make a lot of money from all that other stuff too. Windows itself might not even be that much of a money-maker these days.

>Of course like any big company, they make plenty of bad choices - like telemetry, the "watch everything you do" AI assistant, and so on.

MS can afford to do a lot more bad stuff like this, because Windows users aren't going anywhere. It's why they can bake ads into the OS.

>I'm sure some of them are exactly who you think they are. But some - I suspect most - of them really want to do a good job and make products their customers want to use.

I'm sure there's some of them too, but they're not pulling the strings. They do get to work on cool stuff now and then of course, like the interesting kernel features you listed, but those are in support of their long-term goals (like being prepared for a post-x86 world).

>Did you know companies can't use the NFC chip in Apple phones - in my phone, which I've already paid for - without being paid millions of dollars for the privilege? I'd love it if I could pay for public transit in my city using my phone, which I own, but my government doesn't want to pay the extortionate price apple is charging

No, I didn't know that. It doesn't surprise me though; Apple was like this with Firewire too, and it's why Firewire is dead now. But that's weird too, because here in Japan Apple users routinely use their phones for the public transit, and I kinda doubt the IC card companies paid that much, though I guess it's possible. (Android users use their phones too, but only if their phones are Japanese models. iPhones however all have the Felica NFC chip needed for Japanese transit systems.)

FWIW, that newer Windows Solitaire app isn't actually using Electron now; it's a UWP app using XAML. I guess that just goes to show that the problem isn't Electron specifically.

In the long run the hardware that can still run Window 3.1 will become harder and harder to find and they'll be forced to upgrade, but currently enjoying the benefits of "if it ain't broke don't fix it". Plus, there were literally millions of systems made that can run Windows 3.1 so it will be many many years before the hardware is too hard to find.

We're talking about a problem on the scale of 4,000 flights per day. Assuming you avoid O^2 complexity computations that's the sort of thing even 90s computers could handle easily.

You can basically run Windows 3.1 in dosbox on a potato now, so the hardware really isn't even a problem. If any of this was actually true...

>> Plus, there were literally millions of systems made that can run Windows 3.1 so it will be many many years before the hardware is too hard to find.

Two of my first contractor roles I had as a developer really opened my eyes to a lot of this.

We were building an inventory management system for a large company that built farm equipment. We started building it and once we got to the browser and mobile requirements, one of the VPs spoke up and asked if it would run on IE6 since they had not one, but THREE of their inventory legacy systems that still ran on Win98. This was in 2014, a full 6 years after many companies had stopped supporting it. And another 4 years since websites stopped supporting it.

The other one was for a very large, regional construction company. Same thing, we were building a web app for them and in one of the conference calls, one of the VP's was asking how this would run on Windows95 for the same reason. They had several legacy ERP systems that were running on Win95 and had specific requirements for stuff to run on that OS.

As a developer who was used to working with somewhat current tech - it was a real eye opener. It was crazy to think how many massive companies just didn't have the constitution to upgrade their stuff, and then by not doing so, had now dug themselves into an even deeper hole.

Once I started hearing stories about the details of this scenario, it made perfect sense to me since I had seen it multiple times. And not from little companies who didn't have the money or resources to upgrade, but massive Fortune 500 companies who just neglected their stuff until was too late.

productivity wise 70's and 80's peaked with those thin terminals with every action carried out by keyboard... workers tapped at light speed due to muscle memory, didn't look fancy, but it got the job done. GUI is sexy but like short videos ultimately did nothing for the user

> productivity wise 70's and 80's peaked with those thin terminals with every action carried out by keyboard

Computers only started showing up in nationwide productivity figures in the 80s to 90s.

> GUI is sexy but like short videos ultimately did nothing for the user

That's a stretch.

I guess it increased the productivity (measured in amount of "work" done, not necessarily something useful) expectations for most workers which effectively did nothing for them because they still need to work as much even if modern software allows them to accomplish much more in the same amount of time. So t might make sense in that regard.

> workers tapped at light speed due to muscle memory

That's great if we're mainly talking about robotic tasks than can be mostly automated to only require a fraction of those clicks but wasn't for some unclear reasons.

Almost every industrial embedded system I've ever used runs Windows XP at the absolute newest, and it is not uncommon in the slightest to see stuff as old as 95/3.1. These are computers that operate machinery that costs 6 or 7 figures. If it ain't broke, don't fix it.

Don't get me wrong, my Macbook is an absolute beast for all of my work tasks, and my gaming PC is an utter joy to use for my recreation time, but at the end of the day, for a ton of applications, a computer doesn't need to do shit beyond sending a lot of signals out of a parallel/RS232 port to control systems to operate... I mean Christ, anything. CNC mills, building lighting/security systems, packing machines, or to do things like issue tickets to people parking in a ramp. Like... a lot of this stuff just does not benefit at all from a modern software stack. Stick a crappy PC inside instead, load it up with the same image it had before which includes firewall rules that shut down every last port and connection apart from whatever needs to manage it, and you're done.

Don't fix what ain't broke.

It’s like a calculator, if you need to do basic math, you can use an old calc.

> is computing actually a solved problem and we're really just mostly reinventing the wheel and enshittifying perfectly already working systems?

80% of the work is json bureaucracy

The other 80% is adapting to new requirements

And if you’re lucky maybe 0.1% of the time you get to build something new.

Fear not, a lot of this stuff was perfectly solved with pen and paper long before us computer nerds came to play in the big boy sandbox

It is BS. Continuous updates for security notion, especially so. That said, the barrier to entry for programmers did come down significantly.

Reply on news.ycombinator.com