Slashdot Mirror
Borland Being Purchased By Micro Focus
An anonymous reader tips news that Micro Focus is in the process of buying Borland Software for $75 million. They also picked up Compuware's application testing and automated software quality business. Quoting ZDNet:
"The boards of both companies agreed to the deal, which is expected to complete around mid-2009. ... In 2008, Texas-based Borland made a pre-tax loss of $204m, almost four times the size of the previous year's loss. It had revenues of $172m, part of a consistent downward trend since at least 2004. ... Borland was one of the oldest software companies in the PC software business, having been founded in 1981. Its most successful era was in the late 1980s via massive sales of Sidekick, a DOS-based terminate-and-stay-resident personal productivity application, and development tool Turbo Pascal, which challenged Microsoft's dominance in the application-development market."
"Micro Focus Net Express® is the market-leading COBOL development environment"
So, a company that should've died off in the nineties is being bought by a company that noone has ever heard of that should've died off in the eighties. Weird.
It's too bad the company went under like that, but I would have to blame the executives for making such massively bone-headed business decisions.
Anybody remember Inprise? After about a year of incredible downturn, they decided, "You know what? Maybe Borland wasn't a bad name after all"
Idiots
Delphi *was* my favorite language
Sidekick, a DOS-based terminate-and-stay-resident personal productivity application
Aaah good old terminate-and-stay-resident programs, from the heydays of non-multitasking OSs. Anyone else remember Int 27h and the magic of hooking a subroutine to make it appear like your OS was actually multitasking? Hmph...kids these days..
An old-timer with old-timey ideas.
I remember, back in the day, when all malware was written in borland C/C++.
Er.. not that i wrote malware. >_>
Skiffy is Spiffy, but Ort is tort.
It's a shame that they are going under, because C++ Builder is he best C++ IDE for Rapid Application Development, by far.
You can design forms and controls in the same way as Visual Basic, but it is C++.
Man - between all that bullshit and bands like "A Flock of Haircuts" it was enough to make Max Headroom hurhurhur-HURL!
RS
Shoes for Industry. Shoes for the Dead.
First Sun now Borland? Very sad but in both cases you had good technology and poor management. I realize that IBM's funded free Eclipse made hurt Borland JBuilder sales but to sell off the development tools division? Really?
You can design forms and controls in the same way as Visual Basic, but it is C++.
I thought that was called Visual C++.
https://www.eff.org/https-everywhere
I'm not sure what surprised me more when I read this: that Borland still exists, or that Micro Focus still exists.
"You cannot simultaneously prevent and prepare for war." -- Albert Einstein
Let us not forget that Borland had a pretty dominate position in the programming C/C++ IDE market way
back in the early 90s.
http://en.wikipedia.org/wiki/Turbo_C%2B%2B
I remember all of the C programming college courses in my area all used Turbo C as the preferred IDE.
I remember that many folks claimed Microsoft sabotaged Borland's product by integrating their Visual Studio with windows in ways that Borland just could not do. This was years before the Netscape lawsuit! I even seem to recall reading that Microsoft was accused of preying on Borland's staff and hiring them away. Perhaps someone with more knowledge than I can provide some more information on those bygone days.
Being a Delphi programmer I noticed this news and thought wtf, but it must be noted that Delphi along other programming stuff was moved under CodeGear a few years ago and wasnt included in the purchase.
Still, I have high respect for Borland and the fact they provided early Delphi's for free on my teenage years when noone else did. I still enjoy Delphi as the most rapid programming tool, because it nicely integrates easy of GUI design but still powerful and fast code.
http://www.networkworld.com/news/1997/0512borland.html
In the end, Microsoft strategy of simply throwing obscene salaries at the Borland talent ultimately worked. It was systematic, it was effective.
Now go suck on Visual Studio.
I believe they sold more Delphi licenses than turbo pascal. Furthermore I think Delphi was the the impetus at Microsoft for things like the MS developing a true IDE, J++/visual J and finally C# which btw was architected by the very same guy that did Delphi.
The biggest shame was when at the end Borland tried to sell their compiler business for roughly $1b no one wanted it, eventually some veritably unknown company called Embarcadero made an offer for $24m for the business and that was the end of that.
Lesson of the day: Regardless of how good/essential the products you deliver may have been, bad management and poor future insight can make you crash and burn.
Arash Partow's Philosophy: Be a person who knows what they don't know, and not a person who doesn't know.
C++ Builder and Delphi were sold off some time ago (to Embarcadero in 2008, according to wikipedia), so I'm not sure what Borland actually does these days, but it should have no effect on any of the CodeGear stuff. I still use Delphi, it's a great IDE, but not as nice a language as c# imho, maybe there'll be a C# Builder in RAD studio at some point.
Oh no... it's the future.
Turbo C++ came years after Borland's original product: Turbo Pascal.
I started with Turbo Pascal with version 1.0. At the time, it was a revelation because it cost $49.00 in the days when PC development tools typically cost many hundreds of dollars, and because of its speed. It could compile a several thousand line Pascal program in just a few seconds. Other compilers of the time, such as Microsoft Pascal, took many minutes to compile the same code. It was limited, however, to 64K of code because the compiler created .COM files.
The compiler was so fast that Turbo Pascal was the rapid development tool of the 1980s on the PC. Nothing else could approach its speed.
While Phillipe Khan always maintained that he was the developer of the Turbo Pascal code, it was actually Anders Hejlsberg, the architect of C#, that actually wrote the code.
Only with .NET has Microsoft finally caught up with RAD form design. .NET is over 7 years old now... You might as well be railing against Windows 98.
https://www.eff.org/https-everywhere
COBOL may not have much mindshare among slashdotters, but there's a lot of COBOL code out there. Most of those boring apps that do nothing but apply simple business logic, like the one that cuts your paycheck, are written in COBOL. Remember the Y2K crisis? That was mostly about COBOL apps.
Which isn't a defense for the continued existence of COBOL. I only disagree with your statement that it should've died off in the 80s because I think it never should have been invented, with its stupid pseudo-English syntax. But like Fortran and RPG, it's too well established to be disposed of.
Assuming that Borland still does IDEs and compilers (weren't they trying to spin off that business?) this is a really good fit. Borland's tools are really kewl, but they've never gained serious mindshare, and survive only because of a lot of diehard users. Not, strictly speaking, legacy tools, but really the same kind of marketplace.
Incidentally, I used to work for Convergent Technologies, which back in the early 80s sold a MicroFocus COBOL compiler for its 68010 UNIX boxes. This compiler was, weirdly enough, written in COBOL. Somebody once explained to me why this made sense, but I've forgotten the explanation.
All the developer tools were shipped off to CodeGear a few years ago. They are now owned by Embarcadero who are starting to invest more heavily in R&D. Delphi and C++ Builder 2009 are a vast improvement on the previous offerings.
My favorite Borland product was Paradox for Windows, a RDBMS engine and GUI with IDE. The engine was available as a C++ library for embedding. It brought together programming and data techniques from spreadsheets, databases, languages and GUIs that made "Windows" a complete and consistent platform.
Borland, or somebody, could do exactly that with existing OSS code today. The software world could use such a tidy tool, and especially a competent company to market it. Maybe that's Oracle now, but the game is just getting rebooted again.
--
make install -not war
It was blinding fast for a compiler of its day, running on a 1 MHz Z-80. There was no debugger, but if a Turbo Pascal program halted with an error at a given location (which it would politely print out before quitting), you could run the compiler to find out which line of code that location represented. It was cheap, too -- fifty bucks or so at a time when other compiler makers were charging $300 or more.
I wrote a computer game in Turbo Pascal that got me my first job in the game industry. VERY fond memories.
I piss off bigots.
All the people remembering Borland's language wars with Microsoft, and came up on the other side, should know that all of those tools were sold to Embarcadero some time ago. The Borland that we knew has already been gone for quite some time. Turbo C++, C++ Builder, Turbo Pascal, JBuilder, etc, all live on at Embarcadero. In fact, I think Embarcadero even got the Borland database...
This is my sig.
BTW Delphi 2009 supports C# style generics, anonymous methods, inferred typing and deferred execution all in native code w/o .NET
Delphi is still very much a viable platform for new software!
Two very popular Borland products back in the day were the Quattro Pro spreadsheet and the Paradox relational database. Quattro Pro had WYSIWYG and three dimensional features running on DOS way before Lotus. Paradox was a huge advance over dBase III in ease of use and report writing.
If you had 2 MB of system RAM, they could both exist in system memory at the same time and swap back and forth. Not quite multitasking, but innovative at the time. Using DR DOS made the memory tricks easier. Ah... memories.
Some mornings it's hardly worth chewing through the restraints to get out of bed.
I worked in Borland, when it was indeed Borland. Great company, you could not find another place with so many fine minds.
What is called Borland today is not the company that people knew. The management stole the name, connected it with mindless, buzzword-rich nonsense and moved the headquesters from Scotts Valley to Texas. They were selling nothing and that's what MicroFocus is buying: nothing.
The core of Borland's business, compilers and IDEs was spun off as CodeGear, recently purchased by Embarcadero Software. CodeGear is still located in Scotts Valley with many of the original developers in the group. Great people with a passion for tool development.
It's not a coincidence that Borland, the travesty, has been losing money at incredible speed after CodeGear was gone. The only part of the business that made sense, that generated revenues, was let go by a management simply unable to understand what a compiler is.
That the name Borland, which was synonym of innovation and "barbarian" spirit, is now associated with the leading name in a technology that was an embarrassment in the 80s, COBOL, is a shame that makes me cringe to no end.
Remember, this is not Borland, the real Borland, the one that brought us such gems as Turbo Pascal, C++ builder, Paradox, JBuilder etc, and that in general taught Microsoft how to write IDEs, is called CodeGear.
The company mentioned in this article, is a travesty and a sham.
The Borland Museum has the old Turbo series of Turbo Pascal, Turbo C, and Turbo C++ for MS-DOS downloadable for free.
Turbo Pascal and Delphi got replaced by Free Pascal, and Turbo C++ got replaced with GNU C++ and MinGW C++ for Windows which are open source alternatives to them. Which I think is why the Borland Museum got opened and why the command line version of Borland C++ was given away for free.
While people were waiting for the Borland Museum to release Delphi 1.0 the Lazarus Project was developed based on Free Pascal to replace Delphi.
Remember, Slashdot does not have a -1 disagree moderation, and no, troll, flamebait, and overrated are not substitutes.
Buying Ashton-Tate, maker of dBASE, was their downfall. Huge outlay and the migration to windows was a massive failure.
That wasn't their downfall. Their downfall was the same thing that made WordPerfect an also-ran, that virtually destroyed Novell, that ended Netscape, and heavily contributed to the end of Sun: Microsoft.
Love them or hate them (and at Slashdot it's usually the latter), Microsoft is single-handedly responsible for the deaths of many tech companies. In Borland's case, they simply couldn't survive against MS Visual Studio. Everything else they did or did not do pales against that fact.
Life is hard, and the world is cruel
They killed of Brief ... now it is their turn to rot. not that i hold a grudge or anything ...
Yes, in my experience 2009 is rock solid. 2007 was already OK. I agree 2005 was crap.
Never send a human to do a machine's job.
Other than frequent crashes, s l o w r e f r e s h e s, a Java client made by saboteurs, a couple hundred millions interface inconsistences(in most screens, pressing enter or escape do nothing), a braindead labeling schema, tectonic-speed checkouts, rude and intrusive dialogs that steal focus from whatever the fuck you're doing, the fucking JVM that eats oodles of RAM, the goddamn interface that suddenly frozes for a minute or ten, the motherfucking MPX, the metric fucktons of annoying as hell bugs that send your goddamn motherfucking productivity down the tubes, the incredibly shitty compare and merge tool and a couple other things that spring readily to mind no, there's nothing bad
Yes, I'm a CM and use Starteam daily for my sins. Why do you ask?
Um... it should also be noted that CodeGear was purchased by Embarcadero last year
I concur. I use it on Windows so it might be different elsewhere, but...
Network timeouts after disuse get annoying "Lost connection" messages - there's no way to recover. It's faster to CTRL_ALT_DEL and open task manager and kill javaw.exe than to attempt to close it using the X and battle the "i can't close the application because the server isn't responding" logic.
"Update Status" gets stuck sometimes and you can't even "compare files" to see if you are up to date. "File is locked"... have to kill ST and restart, which gets slower with every version. The "file compare" function gets stuck on "Finding differences" and the only solution is a) find the temp file that is causing problems and delete it b) when that doesn't work, because it won't, reinstall ST completely.
When you check in a file, and "compare differences" to make sure everything is entered in the check-in message, you can either type in the commit box, or scroll through the differences in the compare tool - but not both. You could in the old version, so they intentionally cocked this up looks like.
Enter a commit message and forget to lock the file first (for those 5-second changes) or if there is some conflict on the local drive and it errors, it throws away the entire message and you have to re-type if you want it back. That's when I learned the art of selecting all and copy before hitting enter. And sometimes that doesn't work, so select all, cut, paste, and then you're sure it's on the clipboard so hit OK.
I use the keyboard as much as possible, but it's shite for keyboarding as mentioned.
"View Manager" is a travesty. I have had so many conflicts and such I couldn't make sense of, and it took 4 hours to sync a vew!!!! I resorted to copying the files manually from one view to another and then manually re-check them in to bring the base view up to date. I finally gave up and used labels/tags instead.
You open it, and it takes several minutes to get to a usable state, so let it run in the background. Well it calls SetForegroundwindow() every time it accomplishes something, so just start it and read slashdot and ALT+TAB back to slashdot a few times. when you no longer notice that it's annoying you, it's been ready for a while and you just lost half your workday. It's not a tool for the impatient.
I did ask the server admin where one of my files went and they couldn't find it (it's in the view, but can't do anything with it), but that was maybe 2005 version and we have updated.
It's easy to have duplicate files with the same name, especially when merging views. So one is "unknown" and the other up to date, then you update the other file and they switch. One is locked, the other isn't... Depending on the situation and how it happened you can actually lose commit comments and history when this happens, just choose the most useful one and remove the other one.
The view settings by default go to the first one in the list, so I have someone else's filtering on every folder and have to change it. I can add my own view type, and if it's alphabetically before everyone else's, the ENTIRE USER BASE for that server gets my new view as default. Name it something dirty with a leading underscore or less-than and hilarity ensues. Maybe they fixed that in 2008, but that's seriously broken. It's actually very easy for a simple user to make changes that affect all other users, like re-locating the local repository location for individual files or whole folders.
You can program it using COM objects, and it's fairly easy to find samples on like codegear, if that still exists, but documentation is crap. I wrote a VB6 app that would check stuff in and out and lock it and do simple stuff, just so I didn't have to load this virus into memory. I used it to select files for review documentation, since it's very difficult to get information out.
I asked for a simple "Copy folder location" or "Open containing folder" function - you have to right-click, properties, and then it's hard to sele
And VB has been killed by .net as well so what did you gain?
Java is still doing great so you bet wrong.
See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
I think Borland underestimated two things: {...} and the way businesses used development platforms - to talk to databases.
Perhaps you should google around about Why Delphi was called Delphi.
Delphi was envisioned from the beginning as a platform to communicate with databases.
"Sufficiently advanced satire is indistinguishable from reality." - [Tips: 1DrYakQDKCQ6y52z6QbnkxHXAocMZJE61o ]
Almost ten years ago, in my early twenties, I asked in a forum which language to learn for relatively simple, Windows applications. I am not a professional programmer, just a hobbyist.
Most people replied that the best language for RAD was Delphi. A few said go with Java. I didn't choose any of these, I preferred Visual Basic to have the peace-of-mind of Microsoft.
Delphi died when the .NET and C# arrived, Java will probably lose its mojo now that Oracle leads the development. I don't know, we may hate Microsoft but most of the times is the last player standing.
If you've learned Delphi, you wouldn't have trouble switching to C# when that arrived. The language at 1.0 embodied many of the same design concepts (not surprising, since lead designer is the same), and the UI library (WinForms) had that definite VCL'ish smell.
And VB? I mean, that VB6 -> VB.NET migration was a major change, the languages are only vaguely similar syntactically, but semantically they're significantly different.
Anyway, 10 years ago when you asked, Delphi was definitely the right tool for that job. And part of the job of software developer is the ability to pick up new technologies as time goes on. You can't realistically expect to stay in this business for long, since languages and frameworks and approaches change quite radically every decade, and there doesn't seem to be an end to this; for example, right now we're clearly in the middle of a shift from strictly statically typed, limited strict-OO languages such as the original Java or C# 1.0, to mixed static/dynamic typing, mixed OO/functional languages such as Scala).