[dmitrybrant]

Like so many other programmers of a certain age, I cut my teeth on Borland IDEs in the late 90s, building software for Windows 9x, which was a time when Borland was absolutely crushing it with tools that were light years ahead of Microsoft's own offerings at the time.

It makes me so sad that Borland tools are squarely in the dustbin of history, and are now maintained almost exclusively for legacy customers. I would love to read some kind of retrospective (a book, a series of blogs, etc) on the downfall of Borland and what exact business circumstances led to its irrelevance.

[tabtab]

There's always Lazarus, a more or less Delphi clone. Being open-source, it's harder for a greedy or incompetent vendor to yank the carpet out from under your org.

I've considered abandoning web-dev for Delphi/Lazarus. The web is a lousy fit for CRUD/GUI. React etc. are Spruce Gooses. (We need a dynamic over-http markup standard, kind of like a friendlier and XML version of QML.)

[dt3ft]

I’m considering this too. I tried writing some test using react testing library and jest, but in order to debug my test and see actual html that the test was rendering, turns out you either have to log the output or use jest-preview which is in alpha to say the least. Documentation, getting started… all of it is so immature and working half-ways that I can’t help but wonder what are we doing with our lives?

[dt3ft]

As someone who got started with Delphi 2.0, I can tell you what killed it for me: the release of .net and visual studio. I looked back a few times, but always went back to visual studio.

[snarfy]

Microsoft hired Anders to stop working on Delphi and work on C# instead. That was the beginning of the end.

[pjmlp]

That isn't how it went down, Anders got fed up with Borland management and decided to finally accept the offers from ex-Borland folks to join Microsoft.

When he joined, his first project was being part of J++.

[fm77]

I have read that quite often here and there and while the management at Borland certainly was a pain, I don’t think that for Anders Hejlsberg this was the main reason to leave.

I believe (and I have no prove) that during development of Delphi V1.0 Anders realized that his „baby", the source code of the compiler core written in 16-bit assembly became worthless. The new 32-bit compiler for Delphi V2.0 was written in C by Peter Sollich.

Anders new role at Borland was just to be an architect / manager / teamleader / whatever… He could have contributed to the new 32-bit compiler in a way he does today to TypeScript, but for whatever reason he didn’t wanted to. That, along with the Borland management and the nice $$$ signing bonus offered at Microsoft made him leave. Again, that's just my take.

Peter Sollich left Borland I think in 1998 in order to join Microsoft, to this day he is in the C#/.net team. https://youtu.be/LPcjSdob9AA

And I still wonder who has written Turbo Pascal for Mac in 1985/86…

[pjmlp]

It is the official reason told by himself in this interview,

"Anders Hejlsberg: A craftsman of computer language"

https://behindthetech.libsynpro.com/001-anders-hejlsberg-a-c...

Now if you want to tell he makes that up, I dunno.

[shrimp_emoji]

The shittiest etiology for an apocalypse possible XD

[lelanthran]

There was a critical time in the industry, if your tech didn't run on Linux it was dead in the water.

Delphi and C++ builder should have been ported to Linux and the BSDs by 2001.

They hitched their wagon to Windows and are now almost a footnote in history.

Come to think of it, almost all Windows only applications faced the same fate.

[slowmovintarget]

No. It wasn't Linux, it was web applications that did Delphi in commercially. GUI desktop software was suddenly unpopular and web applications were all the rage because OMG! just go live! No upgrade and multi-platform InstallShield build to deal with.

My first serious language was Turbo Pascal 3.0. When Delphi came out, I jumped on board. My first programming job was as a Delphi programmer, starting with Delphi 1, switching to Java after Delphi 5 Enterprise. Delphi, and even C++ Builder were IDEs geared toward producing desktop client-server applications. If you were building web apps Java was the place to be.

Kylix never mattered, because Linux desktop applications didn't matter, generally speaking, so it never became profitable. Linux as an alternative to Solaris on servers was a huge deal though, and is now nearly ubiquitous. But again, Delphi came very late to the web application game.

[coverband]

That was exactly it… Plus their clumsy rebranding over and over, changing licensing terms that stopped non-enterprise users from adopting, and the CS educators adopting Java and dropping Pascal… They tried to use Delphi as a cash cow when they needed innovation…

[heelix]

One of my early professional 'whoops' was with Delphi. We delivered a client (server) application that was sent out on like 20 floppy disks. I misspelled Pharmaceutical, which was part of the company name, on the splash screen. We had a lot of 'junk' floppies floating around the office after that.

The way the web changed customer facing software was amazing at the time.

[mellosouls]

There was a critical time in the industry, if your tech didn't run on Linux it was dead in the water.

No. I can't remember that ever being a thing, I've no idea where you got that idea from.

[eitland]

I guess it was more Mac than Linux, and it happened around peak Rails.

[pjmlp]

Except outside wealthy countries, macOS doesn't really matter and has about 20% market share across the globe.

[eitland]

True, and on a global scale 20% is even a much larger percentage than I would have expected.

My point however is that when I started working in IT I was the only one at the places where I worked who used something other than Windows whenever possible and also advocated for it.

Then there was a change. And as far as I can see that change started with screencasting and as far as I saw, the first widely popular screencasts that existed was Rails, created mostly by 37 Signals (now Basecamp I guess) and other Rails devs/enthusiasts.

I might be wrong about this but I feel fairly certain these guys drove a lot of the change in peoples and organisations attitude to alternative OSes.

Also: for a long time it seemed Linux adoption was also lower in less wealthy countries as using unlicensed versions of Windows are much more accepted there.

[pjmlp]

I would say that its adoption was mainly driven by iOS development, the meme of Year of Desktop Linux hardly coming true as the desktop story keeps being rebooted every couple of years, and the fact that many folks don't really care about GNU/Linux per se, any POSIX environment will do just fine.

As it turns out, the remaining 78% users (taking out the usual 2% from Steam surveys) that cared about POSIX tooling, were happy to use SUA, cygwin, mingw, Virtual Box, VMWare for their needs, and now have WSL in the box anyway.

[Zuider]

Borland released a Delphi compiler for Linux in the late 90s named Kylix, discontinued in 2011. However, it was not successful for a number of reasons, notably that there were better, free IDEs available, there were many compiler errors, and poor integration with GDB. For this reason, developers found it frustrating to use, and it had poor uptake. Even so, Borland had spent so much trying to make Kylix work that it contributed significantly to the company's ultimate collapse.

[IgorPartola]

I cut my teeth on Delphi and I don’t remember Linux ever coming up despite me using it quite a bit. Delphi’s bug strength was the really polished UI builder. The language itself was almost secondary to that. But the UI builder was married to Windows UI and I don’t honestly know how well it might have been translated to something like GTK or QT. Also in the early 2000s both GTK and QT were going through some major changes and so were moving targets.

Besides, the early 2000s were the era of the “this is the year of Desktop Linux” tech articles. Of course that never materialized and Linux desktop is still pretty niche. So I don’t really see how lack of Linux support had anything to do with it.

[elzbardico]

More than this: Borland underestimated the role of web apps, and bet that enterprise users would never move from native client server apps to web apps.

[marcus_holmes]

They weren't the only ones making this mistake at the time.

I remember having a "ruh-oh!" moment when a junior dev showed me his latest DHTML web page that did something we'd previously been doing in VB. The sudden realisation that this was significant and I'd better get on with learning it or be left behind was painful. The urge to continue dismissing the web as a good information source but not somewhere you did things was strong.

[comrad]

Well, they did in 2003. Kylix was the attempt to offer Delphi 6 RAD for Linux.

https://wiki.freepascal.org/Kylix

[lelanthran]

I don't remember why, but kylix didn't work for me.

And C++ builder was never ported.

[robocat]

Crappy features - features that developers didn't want and didn't need. Management driven? The headline features on every release were just rubbish that you didn't care about or never used - very few headline features held up over time. This started in Turbo Pascal days - remember OWL?

Of course they were struggling to compete with Gorilla Microsoft, but Delphi had a huge community, commercial components, and most components were source available (C# was wayyyy behind on that I thought).

Borland fucked up, and after that it was dead and sold. 64-bit super was released years late, and the IDE still ran 32-bit (slow and crashy on high-end machines).

I would summarise it as lack of 'taste' for what developers want. Microsoft later followed the same path: Microsoft had the intense love of many developers but has squandered that love over the years. Balmer's "developers, developers, developers" rant was actually brilliant (a perhaps is why it is so memorable).

[yobbo]

Probably technology shift; the apps that used to be delphi/visual basic became web apps in java, dotnet, or php, and those were free.