Justin,
I'm with you and I'm glad to hear all the good things you say about developer recognition that Borland/CodeGear products are the class act in development. They were the first GUI development package to offer compiled code and opposed to P-Code, which all the MS products and Powerbuilder app devs were generating. Delphi apps flew!
I don't understand everything that CodeGear is doing, strategy-wise but it seems like the company is in capable hands.
Your comparison of Borland to Novell is right on the money. Unfortunately, this industry has gotten so comoditized that the best products rarely win out. Examples: Netware (the most efficent software I've ever seen), Scitor's Project Scheduler (no Sciforma's PSNEXT), Harvard Graphics.
Thank You for this article. ITs great to read som sense on these pages.
George
Discussion on:
View:
Show:
Do you have fond memories of Borland? Do you think that they can return to their former glory? Are you hoping that they can?
J.Ja
J.Ja
I'm still using Delphi on a regular basis and I'm still VERY happy with it. I wish it wasn't so hard to keep track of the new directions that "Borland/CodeGear/Whomever" head towards, but I'm still tagging along as long as they continue to produce the superior products that I rely on.
Yeah, I Borland has been on this name/strategy change kick since around 2000 or so, when they became "Inprise" (or maybe it was before that?). I get the impression that someone high up became convinced that their difficulties were something that mere re-branding would solve, since their products were so superior. I think that it was not *just* a marketing problem, but a full package problem. The market shifted quite suddenly from an IDE being an integrated code editor/compiler/debugger, to being the core of an entire ecosystem. Microsoft "got that" when VB 3 could suddenly shape how a developer saw the world, instead of picking an IDE that fit their style, and kept refining that formula. The latest versions of Visual Studio are so tuned to .Net development, I can't imagine trying to do anything but .Net in them.
J.Ja
J.Ja
Tony -
Take a look at comment #4 on this blog:
http://blog.marcocantu.com/blog/delphi_embarcadero_:2.html
It's from the gentleman that I interviewed. They are clearly changing their course on .Net, and re-emphasizing their native code stuff. I think that you'll appreciate that. They'll be passing me copies of the "next gen" Delphi & C++ Builder for me to look at.
J.Ja
Take a look at comment #4 on this blog:
http://blog.marcocantu.com/blog/delphi_embarcadero_:2.html
It's from the gentleman that I interviewed. They are clearly changing their course on .Net, and re-emphasizing their native code stuff. I think that you'll appreciate that. They'll be passing me copies of the "next gen" Delphi & C++ Builder for me to look at.
J.Ja
Pascal#, well OK, but I don't see what it offers. The VCL is nice but one of the most annoying things about it is the addition of completely non standard behaviour of apparently standard windows components. I'm now getting requests to implement things like tabbing between between buttons on a toolbar in .Net !.
It's too late, they've lost an enormous amount of their market share, and Delphi.NET offers very little (actually nothing !) to go non-standard and to accept the inherrent delays and potientally costly problems from attempting to keep up with a direct competitor.
The win32/64/128/256 (Please) by all means but delphi.net was/is and always will be a stoopid idea.
Lean, mean and mainly sever side would be my recomendation. Get them selves a market MS can only get by throwing hrdware at problems or by joining together applications, or of course unmanaged C++
It's too late, they've lost an enormous amount of their market share, and Delphi.NET offers very little (actually nothing !) to go non-standard and to accept the inherrent delays and potientally costly problems from attempting to keep up with a direct competitor.
The win32/64/128/256 (Please) by all means but delphi.net was/is and always will be a stoopid idea.
Lean, mean and mainly sever side would be my recomendation. Get them selves a market MS can only get by throwing hrdware at problems or by joining together applications, or of course unmanaged C++
I could never stand Borland, and it all started with Turbo Pascal. I was hot out of college (ok, dropped out, but still) and had just learned Pascal on a Dec VAX, VMS. I went down to a local software shop and bought TP, took it home and started to write a hot program I was going to sell. So I thought.
Turned out that TP was SO different from ANSI Pascal (or DEC's idea of it) that I might as well learn C. I got the text, thaught myself C, then bought QuickC 1.5. I was still getting junk mail from that first product registration card over a decade later. All I could do was laugh.
Then came Paradox. R:Base seemed to be in the dust, and so many of my friends were talking about Paradox like it was heaven. Spent a whole week converting all my databases to it only to find out that the ASK queries would let me link the tables together, but I'd never be able to edit (and save) the data back. (I spent the next week converting the databases to Access 1.1.)
My partner works for Borland now that they've but off all this stuff. It's always fun to tell these stories at their drinks outings after work. LOL
Turned out that TP was SO different from ANSI Pascal (or DEC's idea of it) that I might as well learn C. I got the text, thaught myself C, then bought QuickC 1.5. I was still getting junk mail from that first product registration card over a decade later. All I could do was laugh.
Then came Paradox. R:Base seemed to be in the dust, and so many of my friends were talking about Paradox like it was heaven. Spent a whole week converting all my databases to it only to find out that the ASK queries would let me link the tables together, but I'd never be able to edit (and save) the data back. (I spent the next week converting the databases to Access 1.1.)
My partner works for Borland now that they've but off all this stuff. It's always fun to tell these stories at their drinks outings after work. LOL
"Then came Paradox. ... Spent a whole week converting all my databases to it only to find out that the ASK queries would let me link the tables together, but I'd never be able to edit (and save) the data back."
Whoa there, cowboy! I've been using Paradox since 1988, eight versions ago. In Paradox, you shouldn't even *try* to edit data directly in tables (via queries or anything else), although Pdox does allow it. You link tables in forms, then place index-based filters on the forms. Takes a minute, and gives fully relational editing. Present in 1988, and very much improved since.
It strikes me that you may have disliked these products because they looked funny, not duplicating other products you were comfortable with. Others loved Borland for precisely this reason -- it broke out of the ruts of tired Old School thinking.
Whoa there, cowboy! I've been using Paradox since 1988, eight versions ago. In Paradox, you shouldn't even *try* to edit data directly in tables (via queries or anything else), although Pdox does allow it. You link tables in forms, then place index-based filters on the forms. Takes a minute, and gives fully relational editing. Present in 1988, and very much improved since.
It strikes me that you may have disliked these products because they looked funny, not duplicating other products you were comfortable with. Others loved Borland for precisely this reason -- it broke out of the ruts of tired Old School thinking.
It's the fastest and best way to scrub a long table of data. Fastest way to see some errors, such as happless data entry clerks entering their own city and state, but the correct zip code.
Sometimes old school is still the most effective way of doing things. Large procedures in OOP still need some top-down structured practice.
Sometimes old school is still the most effective way of doing things. Large procedures in OOP still need some top-down structured practice.
Turbo Pascal left great chunks of the language out of its implementation, and part of the reason why it ran fast (as well as compiling fast) was that it took shortcuts in the code it generated, which led to some nasty bugs sometimes.
Essentially, it wasn't Pascal, it was another language, one that just superficially resembled Pascal. There were other, much better implementations of Pascal around at the time, also compiling to native code (the UCSD P-system and the whole idea of P-code had pretty much died out by then) - but they cost ten times as much and were much more awkward to install and use.
Borland got it right with Delphi, though. The "Object Pascal" that underlies it is a superset of standard Pascal, not a new variant; its method of handling objects and classes and pointers is exceptionally clean and logical to use (you don't even notice you're using pointers to objects, let alone learn a special syntax to indicate that something's a pointer); the component library similarly clean and elegantly well-designed; and the RAD IDE quickly builds applications that are actually good enough to use.
I do wish the company would pick a name, and a focus, and stick to it.
Essentially, it wasn't Pascal, it was another language, one that just superficially resembled Pascal. There were other, much better implementations of Pascal around at the time, also compiling to native code (the UCSD P-system and the whole idea of P-code had pretty much died out by then) - but they cost ten times as much and were much more awkward to install and use.
Borland got it right with Delphi, though. The "Object Pascal" that underlies it is a superset of standard Pascal, not a new variant; its method of handling objects and classes and pointers is exceptionally clean and logical to use (you don't even notice you're using pointers to objects, let alone learn a special syntax to indicate that something's a pointer); the component library similarly clean and elegantly well-designed; and the RAD IDE quickly builds applications that are actually good enough to use.
I do wish the company would pick a name, and a focus, and stick to it.
Delphi is what got me singing Borland's praises. I was impressed with Paradox. It was the fastest Desktop database I had ever seen. It sounds like its still useful, too.
Wasn't it Borland that had another product that tried to advance another generation in development, Objectview or something like that where you graphically mapped things out? Maybe it wasn't Objectview but there was some other product that I just can't remember the name of at the moment.
Wasn't it Borland that had another product that tried to advance another generation in development, Objectview or something like that where you graphically mapped things out? Maybe it wasn't Objectview but there was some other product that I just can't remember the name of at the moment.
I didn't get deep enough into Pascal before heading to Turbo Pascal to see a difference, and coming from the VAX/VMS (or NCR mainframe, I can never remember which it was) with vi (or something like it) to Turbo Pascal was like waking up and discovering that the door in your room leads to the rest of the world. I know that TP was fairly non-standard, from what I have read. In college, I ended up being stuck with an awful "p2c" routine which was their version of a compiler... talk about non-standard! The clue was its case sensitivity.
J.Ja
J.Ja
I probably would have loved it. They could even call it 'Pascal-like' and I'd be fine. I would have expected a new language learning curve.
I guess if Borland had one vice that worked against me, it was lack of documentation when I needed it most. Unfortunately, MS took some of that on board. Now MS seems to 'forget' to document things in some up front way. Too many people in the past that would ignore the documentation when they did it.
I guess if Borland had one vice that worked against me, it was lack of documentation when I needed it most. Unfortunately, MS took some of that on board. Now MS seems to 'forget' to document things in some up front way. Too many people in the past that would ignore the documentation when they did it.
I can't vouch for their earlier documentation (it was a lot better than what I had available to me previously, which was "compiler errors and nothing else"). But you are right about the role of documentation. Microsoft had it nailed, but lately that seems to be slipping. Too many items in the .Net documentation seem to just give syntax reference (Visual Studio provides that quite conveniently, thank you, with IntelliSense), with some useless "explanation" ("DoOperation performs the operation.") and maybe a useless example showing the syntax "in action". I'm still in love with Perl's documentation, particularly the "perfaq" document set, which is a great collection of information in the form of, "what's the best way to acheive common task XYZ?" with answers that show the choices and explain the strengths and weaknesses of them (including some common, "a lot of people do it like this, it works, but it's wrong, and here's why..." stuff).
J.Ja
J.Ja
If someone at TR set up the framework, then the rest of us fill in descriptions and examples. Maybe someone at TR to make sure that it all works together.
This would be for all the tool sets so we could get used to one style, and move from toolset to toolset.
Hey, I can dream, right? You know another site where we might be able to do a framework like this? *chuckle*
This would be for all the tool sets so we could get used to one style, and move from toolset to toolset.
Hey, I can dream, right? You know another site where we might be able to do a framework like this? *chuckle*
Matt -
Not a bad idea. MSDN has that feature recently added, PHP's documentation has it too. I think what you're looking for is something built in to the F1 help, not merely a wiki, correct? That would definitely make an interesting Visual Studio plugin, at the very least.
J.Ja
Not a bad idea. MSDN has that feature recently added, PHP's documentation has it too. I think what you're looking for is something built in to the F1 help, not merely a wiki, correct? That would definitely make an interesting Visual Studio plugin, at the very least.
J.Ja
I'm fine with stand-alone systems. Of course it's nice to have a local copy of the lot since I'm often working off-line. I use the O'Reilly CD Bookshelves for this.
But, there is something to be said for a website help framework where each topic has a full article (or set of articles) with examples. Then maybe a wee message board of helps from that.
And, to complete my impossible dream, I'd love it in one place. Same style and all. But I'd forgo that if I could get links to the right base places. I've actually thought I'd put something like this on my site.
But, there is something to be said for a website help framework where each topic has a full article (or set of articles) with examples. Then maybe a wee message board of helps from that.
And, to complete my impossible dream, I'd love it in one place. Same style and all. But I'd forgo that if I could get links to the right base places. I've actually thought I'd put something like this on my site.
by using Borland C++. I also preferred TASM to MASM although I learned Assembly for 80x86 on MASM. I still have my old compilers which I use for some hobby programming. The other Borland product that I remember was Quatro Pro for Windows. I liked how it made making 3D graphs as cinch.
I believe it came bundled with Paradox and QuattroPro. (for the IT youth: database and spreadsheet apps).
In 1990 when CS became my major, Borland's books were my primary source of C/C++ cognition. Two years ago, a collage professor asked me for a good C++ reference. I gave him one of my books. He was impressed and thankful. Don't throw them away. Not yet.
In 1990 when CS became my major, Borland's books were my primary source of C/C++ cognition. Two years ago, a collage professor asked me for a good C++ reference. I gave him one of my books. He was impressed and thankful. Don't throw them away. Not yet.
The first Borland tool I work with was Turbo Pascal 4.5 when I went back to technical school. It was really good
Then version 5 came out and I really started to get interested in OOP. It was really nice to be able to work with the simple TP object model.
I got into my "compiler collecter" phase were I would buy anything affordable. I bought a copy of Quick Pascal (on sales). It was interesting but did not have the professional feel of TP.
I eventually got a copy of the excellent Turbo Assembler. Started writting little libraries for my Turbo Pascal work using that. I got to perfect my understanding of how to setup and teardown the stack for Pascal procedure calls and other such fun stuff.
The big event at work was when we bought the superb Borland C++ 3.1 giant box set. We used that for a lot of development. Microsoft tools in those days simply looked and felt amateuriss compared to Borland product. This included the excellent Turbo Vision application framework that started the tradition of superior OO Application framework by Borland. This also happened to be OOP 301 for us because the quality of the code was really good (level of abstaction, cohesiveness, atc...). Even though most of the programming that we did in those days was graphic (we had our own graphic libraries) we did use Turbo Vision to build a shell for our application suite. The good old days...
We used subsequent version of the C++ compiler and OWL to build a few applications. We also have used Delphi for a few things.
For our first years of Java development of course we used JBuilder. Eventually the competition of Eclipse got too strong and we switched.
I hope they survive and wish them the best in their new incarnation. I will keep an eye on them and if I can use anything from Borland I will.
JS
Then version 5 came out and I really started to get interested in OOP. It was really nice to be able to work with the simple TP object model.
I got into my "compiler collecter" phase were I would buy anything affordable. I bought a copy of Quick Pascal (on sales). It was interesting but did not have the professional feel of TP.
I eventually got a copy of the excellent Turbo Assembler. Started writting little libraries for my Turbo Pascal work using that. I got to perfect my understanding of how to setup and teardown the stack for Pascal procedure calls and other such fun stuff.
The big event at work was when we bought the superb Borland C++ 3.1 giant box set. We used that for a lot of development. Microsoft tools in those days simply looked and felt amateuriss compared to Borland product. This included the excellent Turbo Vision application framework that started the tradition of superior OO Application framework by Borland. This also happened to be OOP 301 for us because the quality of the code was really good (level of abstaction, cohesiveness, atc...). Even though most of the programming that we did in those days was graphic (we had our own graphic libraries) we did use Turbo Vision to build a shell for our application suite. The good old days...
We used subsequent version of the C++ compiler and OWL to build a few applications. We also have used Delphi for a few things.
For our first years of Java development of course we used JBuilder. Eventually the competition of Eclipse got too strong and we switched.
I hope they survive and wish them the best in their new incarnation. I will keep an eye on them and if I can use anything from Borland I will.
JS
They WERE a GREAT company. I owned many of their compiler products, still do. Did anybody else read the manuals that came with the compilers? Who can tell me what the parameters "GNU" will do in a search?*
What turned me against them? The two biggest things were Sprint and CodeWright. Editors that they took over and KILLED! Sprint could have easily surpassed WordStar, WordPerfect, and even WORD to become the standard word processor, if Borland would have kept it going. And, if you're programming on various platforms, the BEST programmer's editor I've ever found is still CodeWright. Extensible, works with any language, any source control program, any compiler, across platforms, built-in FTP to transfer files anywhere... until Borland bought it and threw it away.
I don't regret purchasing any tools from Borland, but I need to know the tools I use will be maintained and supported for as long as I need to use them. Borland just doesn't offer that anymore, so I'll take my business elsewhere.
* In one version of C++, the manual actually said that using the parameters "GNU" in a search will "find small antelope-like creatures..." I was ROFLMOA for hours!
What turned me against them? The two biggest things were Sprint and CodeWright. Editors that they took over and KILLED! Sprint could have easily surpassed WordStar, WordPerfect, and even WORD to become the standard word processor, if Borland would have kept it going. And, if you're programming on various platforms, the BEST programmer's editor I've ever found is still CodeWright. Extensible, works with any language, any source control program, any compiler, across platforms, built-in FTP to transfer files anywhere... until Borland bought it and threw it away.
I don't regret purchasing any tools from Borland, but I need to know the tools I use will be maintained and supported for as long as I need to use them. Borland just doesn't offer that anymore, so I'll take my business elsewhere.
* In one version of C++, the manual actually said that using the parameters "GNU" in a search will "find small antelope-like creatures..." I was ROFLMOA for hours!
Still using Delphi 2005, compared to earlier versions it's crap.
At one point they were the best, then they threw their success in the bin and tried to out Microsoft Ms pursuing .net.
Big mistakes all round.
It would take something pretty significant for me to go back to them, and PHP is not it.
At one point they were the best, then they threw their success in the bin and tried to out Microsoft Ms pursuing .net.
Big mistakes all round.
It would take something pretty significant for me to go back to them, and PHP is not it.
... you'd chime in. I thought of you while writing this one.
J.Ja
J.Ja
And nobody is sadder than I (Well OK may be Borland shareholders), that I have to go in the against column.
I started commercially with Paradox 2.0, did Turbo Pascal, and then used Delphi from version 1.0. A sad demise.
I started commercially with Paradox 2.0, did Turbo Pascal, and then used Delphi from version 1.0. A sad demise.
Micropsoft evangelists always had a healthy respect for Delphi and might use it as a prototyper. Do readers realise that, because of the technical superiority of Delphi over VB3, that the VB3 tutorial (part of the VB3 package) was unashamedly written using Delphi? (If you don't believe me and you have access to VB3, run DOS Debug over the executable!
But in fact it is not easy. Borland needs something realy new. They need something like Delphi in the middle of 90th.
My team would like to offer them a new way. But we don't know how contact them :(((( May be someone can help us?
My team would like to offer them a new way. But we don't know how contact them :(((( May be someone can help us?
My entire company still develops exclusively in Delphi, without doubt the best RAD tool out there. Long may CodeGear prosper.
Nice to know that some shops still get to use Delphi and didn't get forced to choose between Eclipse and Visual Studio.
J.Ja
J.Ja
My experience with Borland goes back to the TP 3 days and I am still addicted despite regularly working in many other languages and tools. Ok, maybe I wish for some of the excellent productivity features in Eclipse, but Delphi is still my tool of choice for many things short of web development.
I would like to point out that the power of Delphi goes well beyond the VCL and GUI applications. In my previous job I used Delphi extensively with great success to rapidly build many types of applications including complex mission-critical data processing solutions in support of some of the largest US corporations. To get a sense of scale, imagine an object-oriented data processing platform with 600K+ LOC (95% Delphi) that generates and fulfills one-to-one personalized products in nine languages and implements marketing and business rules for a marketing-obsessed company. Delphi not only delivered with ease, but made it possible to manage a recurring twice-a-month release cycle for changes. The solution also put an equivalent mainframe application to shame by cutting execution time from 4+ hrs to 45 min
That is only the tip of the iceberg of what can be done with a little creative thinking!
I would like to point out that the power of Delphi goes well beyond the VCL and GUI applications. In my previous job I used Delphi extensively with great success to rapidly build many types of applications including complex mission-critical data processing solutions in support of some of the largest US corporations. To get a sense of scale, imagine an object-oriented data processing platform with 600K+ LOC (95% Delphi) that generates and fulfills one-to-one personalized products in nine languages and implements marketing and business rules for a marketing-obsessed company. Delphi not only delivered with ease, but made it possible to manage a recurring twice-a-month release cycle for changes. The solution also put an equivalent mainframe application to shame by cutting execution time from 4+ hrs to 45 min
That is only the tip of the iceberg of what can be done with a little creative thinking!
Boy, do I ever miss Borland, and Delphi in particular.
My 1st job out of college was using Delphi I (thanks Anders Hejlsberg) and I used it and it's successors for many years in many different jobs.
However, one day I saw the writing on the wall and decided that I needed to diversify my skill set in order to be marketable.
I still have a copy of Delphi 5 that I use occasionally when I need a desktop app for personal use, and it's still a lot of fun to use.
My 1st job out of college was using Delphi I (thanks Anders Hejlsberg) and I used it and it's successors for many years in many different jobs.
However, one day I saw the writing on the wall and decided that I needed to diversify my skill set in order to be marketable.
I still have a copy of Delphi 5 that I use occasionally when I need a desktop app for personal use, and it's still a lot of fun to use.
I'm looking forwards to trying out their new products then! I'll be taking a look at their next generation of Delphi and C++ Builder, and I am looking at 3rd Rail and Delphi for PHP now.
J.Ja
J.Ja
Glad you're looking forwards to them! So far, I've got D4PHP2 and 3rdRail installed, which is a major leap for me. It also helps that I am a few weeks ahead of schedule on my MSDN article, I just have a few paragraphs of "commentary" to add and some revisions to make...
J.Ja
It also helps that I am a few weeks ahead of schedule on my MSDN article, I just have a few paragraphs of "commentary" to add and some revisions to make...J.Ja
"3rd Rail" is their new IDE for working with Ruby + Ruby on Rails. I would imagine that it can also edit Ruby all by itself, but they told me that it is really oriented towards Rails development.
J.Ja
J.Ja
Market wise Delphi as a marketable skill turned niche and has waned rapidly. Fortunately switching to VS and C# was vey easy (Thank you Anders Hejlsberg).
Doesn't matter how good it is or was, muscling VS and .NET out of the windows development market will only happen if MS choose to make the sort of catastrophic mistakes Borland did by trying to do Delphi.Net.
Anybody want to hold their breath for that?
Doesn't matter how good it is or was, muscling VS and .NET out of the windows development market will only happen if MS choose to make the sort of catastrophic mistakes Borland did by trying to do Delphi.Net.
Anybody want to hold their breath for that?
I have to agree with your assessment if you need to sell your skills on the open market. I guess I am lucky being able to make my own decisions.
From my experience, however, hitching your wagon to the current favorite comes with its perils. I have seen many mainstream products come and go and leave their loyal followers high and dry. Striving to have the same skills and using the same tools as everyone else also does not leave much room for innovation or differentiation in a competitive market place.
From my experience, however, hitching your wagon to the current favorite comes with its perils. I have seen many mainstream products come and go and leave their loyal followers high and dry. Striving to have the same skills and using the same tools as everyone else also does not leave much room for innovation or differentiation in a competitive market place.
HR types want experience with Tool Y version X, some of them feel it's important.
Delphi is going / gone in the eyes of the marketplace, staying with it isn't going to do me any good. If I want to be marketable, then I have to do something marketable.
A quick glance a my cv reveals I am not a one trick pony though. I still get hits for C and Fortran on VMS. I've even had some for paradox.
I'm a programmer, you pick a language and a task and short of a hilarious mismatch, I'll do it. Anyone who can't isn't a programmer, they are a cookie cutter.
Delphi is going / gone in the eyes of the marketplace, staying with it isn't going to do me any good. If I want to be marketable, then I have to do something marketable.
A quick glance a my cv reveals I am not a one trick pony though. I still get hits for C and Fortran on VMS. I've even had some for paradox.
I'm a programmer, you pick a language and a task and short of a hilarious mismatch, I'll do it. Anyone who can't isn't a programmer, they are a cookie cutter.
I lost a little respect for Anders when he defected to M$. C# isn't a bad language, but it is still a cheap knockoff of java.
BTW, when I stopped doing Delphi development, I didn't move to VS & C#, but rather server-side java web development. I still miss doing Delphi development, and it really was "fun".
As for "catastrophic mistakes", M$ doesn't have a threat like M$ was to Borland. Delphi is still better than VS, in my opinion.
BTW, when I stopped doing Delphi development, I didn't move to VS & C#, but rather server-side java web development. I still miss doing Delphi development, and it really was "fun".
As for "catastrophic mistakes", M$ doesn't have a threat like M$ was to Borland. Delphi is still better than VS, in my opinion.
And it's a flakey piece of crap.
VS2005 and VS2008 are way better.
I've never been a big fan of Java and so I've barely used it.
Aside from the differences managed environments imposes, I've got to say C# compares very well with win32 Delphi. There are a few things I miss such as polymorphic class (static) methods and the native set type, but other than that functionally I have no qualms at all.
It was a pleasant surprise to be quite honest, if something says MS on the front of it, I generally fear the worse.
As for Anders jumping ship, that actually increased my respect for him. Not even someone as clever as him could have made a success of Delphi.net. That required their largest competitor to cooperate. Baldric comes up with better plans than that.
I'm still not convinced VS and .NET are a technological step forward, but from an employability point of view.....
VS2005 and VS2008 are way better.
I've never been a big fan of Java and so I've barely used it.
Aside from the differences managed environments imposes, I've got to say C# compares very well with win32 Delphi. There are a few things I miss such as polymorphic class (static) methods and the native set type, but other than that functionally I have no qualms at all.
It was a pleasant surprise to be quite honest, if something says MS on the front of it, I generally fear the worse.
As for Anders jumping ship, that actually increased my respect for him. Not even someone as clever as him could have made a success of Delphi.net. That required their largest competitor to cooperate. Baldric comes up with better plans than that.
I'm still not convinced VS and .NET are a technological step forward, but from an employability point of view.....
Tony, I totally agree with you on Delphi.net, that was a bad idea. Borland should have concentrated on native Delphi code/performance/improvements, and they probably would have if Anders was still there.
I have to disagree with you on VS being " ... way better". When I was using VS2005, anything we wrote forced our clients to update to winXP SP2 which had been out for almost 2 years. However, our client base was several huge mutual fund companies and their IT departments balked at forcing the SP2 update of all their desktop machines, just so we could install an update of our application. Typical M$ crap, and we never released while I was on the project. Besides, the form designer in VS is clumsy and the IDE is un-intuitive, but that could just be my Delphi background.
All the time I used Delphi, the only thing that I encountered like that was updating the common controls DLL, which by the way was also M$'s fault, because they introduced incompatibilities.
As for java, I guess you use what your job requires, regardless if you're a fan or not.
Cheers,
TK
I have to disagree with you on VS being " ... way better". When I was using VS2005, anything we wrote forced our clients to update to winXP SP2 which had been out for almost 2 years. However, our client base was several huge mutual fund companies and their IT departments balked at forcing the SP2 update of all their desktop machines, just so we could install an update of our application. Typical M$ crap, and we never released while I was on the project. Besides, the form designer in VS is clumsy and the IDE is un-intuitive, but that could just be my Delphi background.
All the time I used Delphi, the only thing that I encountered like that was updating the common controls DLL, which by the way was also M$'s fault, because they introduced incompatibilities.
As for java, I guess you use what your job requires, regardless if you're a fan or not.
Cheers,
TK
of robustness, features etc.
I'm still doing Delphi and still writing for multiple windows versions and service pack and patch levels are still highly critical.
VS 2008 when you build to .NET 2.0 actually building to 2.0 SP1 has been a PIA I must admit.
I've been doing Delphi since 96, VS previous to 2005 sucked big time, but I interchnage quite happily now. I have more problems swapping languages in terms of syntax than IDEs.
Deployment on windows has always been arse. DLL hell, the COMCtl f***up, don't even mention MDAC....
I've deployed VB6, C# and Delphi on winders and I've had problems with all of them, naff all to do with the IDE in my opinion.
Possibly I do have an edge in that I've often worked in more than one coding environment at the same.
Delphi 2005 sucks though, when they rewrote the IDE in C++ they made a complete arse of it.
I'm still doing Delphi and still writing for multiple windows versions and service pack and patch levels are still highly critical.
VS 2008 when you build to .NET 2.0 actually building to 2.0 SP1 has been a PIA I must admit.
I've been doing Delphi since 96, VS previous to 2005 sucked big time, but I interchnage quite happily now. I have more problems swapping languages in terms of syntax than IDEs.
Deployment on windows has always been arse. DLL hell, the COMCtl f***up, don't even mention MDAC....
I've deployed VB6, C# and Delphi on winders and I've had problems with all of them, naff all to do with the IDE in my opinion.
Possibly I do have an edge in that I've often worked in more than one coding environment at the same.
Delphi 2005 sucks though, when they rewrote the IDE in C++ they made a complete arse of it.
It seems like everything I find exciting from Borland's products keep getting killed. I loved Turbo Assembler, the original Turbo Pascal and Turbo Prolog, but they were killed off during the transition to Windows. I followed along, paying good money for Turbo C/C++, only to see entire frameworks being replaced with newer ones. Delphi was a good idea, but the preference for Pascal seemed to make it difficult to follow.
I would really like to see CodeGears move into the mobile device arena. If they can get mobile device development cheap, visual and easy again I would consider it. As long as it keeps steering toward wherever Microsoft is leading, (.Net!), then I have to give it a pass.
I would really like to see CodeGears move into the mobile device arena. If they can get mobile device development cheap, visual and easy again I would consider it. As long as it keeps steering toward wherever Microsoft is leading, (.Net!), then I have to give it a pass.
Actually, TASM is still alive and kicking, but not as a standalone product (as far as I can tell, it is not a standalone product). Michael mentioned to me that they still ship Assembler tools in Delphi and/or C++ Builder, since both products support the embedding of Assembler into the Object Pascal or C++ (respectively).
J.Ja
J.Ja
Evidently the new, improved Borland isn't big on web site maintenance. This link comes in 3rd on a Google search for TASM:
http://info.borland.com/borlandcpp/cppcomp/tasmfact.html
or
http://tinyurl.com/2npwbv
http://info.borland.com/borlandcpp/cppcomp/tasmfact.html
or
http://tinyurl.com/2npwbv
I'll mention that to them Friday if I'm part of their drinks party. You want to see
http://www.codegear.com/products
for anything up to date. The stuff isn't borland anymore.
http://www.codegear.com/products
for anything up to date. The stuff isn't borland anymore.
5 minutes and their web page still isn't displaying.
must be that .net stuff you menioned.
naw, it was the new tab from the link. I stopped the loading and restarted it and the pahe came up.
must be that .net stuff you menioned.
naw, it was the new tab from the link. I stopped the loading and restarted it and the pahe came up.
I had the same experience on their page when I was trying to look something up... my internal "superego" self is telling me, "getting a link from Justin James is like what getting Slashdotted used to be, my blog posts have crushed their server!" And then reality kicks in.
J.Ja
J.Ja
get a faster response from them to one question for me...
where is the c/c++ development tools for linux?
since currently there is only code::blocks for c/c++ development for commercial apps for linux.
code::blocks under the lgpl and using WxWidgets, so it allows for commercial development.
and I did send them an email asking about it.
not a delphi, or .net, or ruby development tool, but a c/c++ tool.
where is the c/c++ development tools for linux?
since currently there is only code::blocks for c/c++ development for commercial apps for linux.
code::blocks under the lgpl and using WxWidgets, so it allows for commercial development.
and I did send them an email asking about it.
not a delphi, or .net, or ruby development tool, but a c/c++ tool.
- Keyboard Shortcuts:
- Prev
- Next
- Toggle
