Borland Releases Kylix 3.0 for Delphi and C++
An anonymous reader writes "Looks like Borland is giving us Kylix for C++ after all. Kylix 3.0 is available in Open, Professional, and Enterprise versions. Time to start banging out those CLX apps! The Register also has a story about this."
The writer seems to think that Kylix 3 will be a C++ replacement for Kylix 2. Kylix 3 will support C++ in addition to the existing feature set.
I wonder how seemless the Delphi and C++ will be. I also wonder if you can use the GNU C++ compiler for the C++ part, or only the Borland C++ compiler.
I don't know about Delphi, but most of our code is developed in Delhi.
Any fool can talk, but it takes a wise man to listen.
Just Linux.NET? I mean honostly, it's just trying to do what .NET is already doing. Of course, even with the trial .NET port, this is a better solution as .NET was never designed for linux.
Personally I don't see why this item is such news. IIRC Borland always said they would support C++ at some later date, and now they are.
Now we just have to wait for really poor UI#s like all of the early Deplhi ones...
This is great news for developers, as long as they only want there apps to run on x86 boxen.
Since when is having multiple xterms "not easy to use"?
[1] http://www.heise.de/newsticker/data/hos-23.07.02-0 00/
[2] http://www.borland.com/kylix/open/index.html
I can think of several great products released by Borland over the years - Turbo Pascal, Turbo C, and Turbo Basic to name a few of them. There was a time when Borland's products were the "compiler of choice" for most developers, especially in the CP/M and early MS-Dos days. For the longest time, Borlands Debugger (TDebug) was shipped with Microsoft's C compilers.
I've actually used Kylix 2.0 Open Edition to cross-compile a shareware game I've been working on in Delphi 6. It's very convienent to have one set of source code, and simply re-compile with Kylix for Linux distribution! (Yes, avoiding Windows API calls and such helps... plus I'm using JEDI-SDL (Simple DirectMedia Layer) for graphic/sound cross-platform stuff). Perhaps now instead of people saying, you can't develop cross-platform games with Delphi you should use C++... I can simple agree (instead of arguing and pointing out that Kylix is cross-platform) and say, sure, I could do that and use the same compiler I've been using all along... leaving certain code in Delphi, but re-writting parts in C++ (just for fun) but no, I don't have to use C++ it's just a language - I prefer Kylix 3.0 for my development environment!
This would be a major reason for schools on tight budgets to turn to Linux for computer programming classes! Why pay for the Microsoft tools (and tax) when this is available? Seems the LTSP folks should look into this...
Code, Hardware, stuff like that.
two powerful object-oriented languages in on development solution
Netcraft has confirmed: Taco-snotting is dying.
Yet another crippling bombshell hit the beleaguered Taco-snotting community when recently IDC confirmed that Taco-snotting accounts for less than a fraction of 1 percent of all homosexual acts. Coming on the heels of the latest Netcraft survey which plainly states that Taco-snotting has lost more fag practitioners, this news serves to reinforce what weve known all along. Taco-snotting faggots are collapsing in complete disarray, as further exemplified by failing dead last in the recent Faggot World comprehensive snotting test.
You dont need to be a Katz to predict Taco-snottings future. The handwriting is on the wall: Taco-snotting faces a bleak future. In fact there wont be any future at all for Taco-snotting because Taco-snotting is dying. Things are looking very bad for Taco-snotting. As many of us are already aware, Taco-snotting continues to lose faggotshare. White ink flows like a river of bubbly, thick jizz. The circle-snot is the most endangered of them all, having lost 93% of its core snotters.
Lets keep to the facts and look at the numbers.
Circle-snotting leader Jeff Homos Masterbates states that there are 7000 snotters of the circle-snot. How many users of anal snot are there? Lets see. The number of circle-snotting versus anal snot posts on Usenet is roughly in ratio of 5 to 1. Therefore there are about 7000/5 = 1400 anal snot users. SnotOS posts on Usenet are about half of the volume of anal snot posts. Therefore there are about 700 users of SnotOS. A recent article put the circle-snot at about 80 percent of the Taco-snotting market. Therefore there are (7000+1400+700)*4 = 36400 circle-snot users. This is consistent with the number of circle-snot Usenet posts.
Due to the troubles of CowboiKneels walnuts, abysmal sales and so on, the circle-snot went out of business and was taken over by SNOTi who sell another troubled Taco-snot. Now SNOTi is also dead, its corpse turned over to yet another gay whorehouse.
All major surveys show that Taco-snotting has steadily declined in faggotshare. Taco-snotting is very sick and its long term survival prospects are very dim. If Taco-snotting is to survive at all it will be among heterosexual hobbyist dabblers. Taco-snotting continues to decay. Nothing short of a jizz-soaked miracle could save it at this point in time. For all practical purposes, Taco-snotting is dead.
Fact: Taco-snotting is dead.
Any links to screenshots of the IDE?
I know they haven't even released a client yet, but Bioware told us that they couldn't port the NWN toolset because there is no borland c++ builder for Linux.
Now there is. Anyone think they'll port it after the client?
I looked all over the Kylix website, and failed to locate any screenshots for their product. What gives?
:-/
Yes, I know I can download a Trial, if I go through the registration process... but I just wanted to look at it.
I've never heard of a product page without product images. Wierd.
Anyone have a link?
They're *selling* it. They're only giving away an "Open" version. The open version can't be used to create commercial apps, and doesn't contain all of the components that make delphi development so easy and efficient. I use delphi at work (I've got it open now, actually), and I've tried the Open version of Kylix before at home... it's way watered down compared to the pro/enterprise versions of delphi/kylix. Still a really useful tool for simple, quick, and fast apps.
It's not a bad business model, considering most people doing non-comercial development won't pay a hefty price for a professional development environment anyways.
Kylix is a commercial app and must use Qt :-)
under the terms of the commercial license.
But the apps created with Kylix (Open) are not commercial so they can/must use Qt under the terms of the GPL. Probably
C++ Builder is still my compiler of choice for Windows development.
;)
Thankfully they're not listening to you.
Is there standard command line compiler in Kylix? Is it possible to compile existing programs - for example Linux kernel - with it?
It will be nice if gcc wouldn't be only choice.
Most definitely. I used Turbo Pascal 7.0 and Turbo C++ 3.0 for a long, long time, well after Windows 95 had come out.
How do written C++ GUI programs function with Kylix? Does one write C++ and does use thereby headers for the CLX library, which is implemented in Pascal (must do thus without multiple inheritance etc.)? And then by Pascal over the detour of non-object-oriented code one accesses the Qt-library written in correct C++? If it's like that, then one remains nevertheless rather directly during a genuine C++ Entwicklungsumgebung.
So you'd use a proprietary product, just not if it uses another proprietary product? (that is distributed under a very similar, and even less restrictive, license!)
How is Borland any less evil than Troll Tech?
When does it give after Delphi.NET a Kylix.NET based on mono?
While it seems that Delphi, Kylix and JBuilder are going strong, Borland has all but abandoned the (seemingly) less profitable Borland Database Engine. Too bad, too since a lot of applications make use of the BDE.
I'm using such an application right now, and due to some size limits on the BDE's config file, i'm unable to use it as efficiently/effectively as possible. Borlands tech support doesn't cover the BDE, and the only newsgroup on the BDE is not monitored by Borland folks frequently enough (if at all).
While i can understand scaling back certain aspects of one of your product lines, to shift focus/userbase to newer/better products, we're talking the BDE here for Pete's sake!
It's just frustrating.
There are 01 types of people in this world. Those that understand binary, and me.
To download a trial version you have to register. If you choose a username that is already in use, it prompts you to choose another one with a form where all the ``spam me'' boxes you unchecked in your first submission have been re-checked . Bastards.
I feel tricked I went to the Borland site to try to download, foudn out I had to register. Only to find out after registering that only version two is downloadable..
ok borland have been shipping their CPP compiler for a while on linux (nothing new)
what this release does I would think is link aganst the CLX to do the GUI stuff
and add their CPP IDE (which is actually visual and drag and drop unlike microsofts which is just a text editor but they call it visual...)
now if they where useing STANDARD CPP why should using GCC be impossible (I suspect they have a few broken things in terms of standard support just like gcc has a few broken things) and the fact that CPP changes every meeting does not help but it would be nice if they said what gcc would require to do this
(I am not talking about opening up CLX just linking your code with it useing gnu tools)
borland selling in effect libs and a IDE would be a good thing IMHO
regards
john jones
2. Fabulous Thunderbirds - Tuff Enuff
4. Our Lady Peace - Superman's Dead
5. Soundgarden - Spoonman
Borland's only a couple years late.
Good try, anyway.
RHIDE had me pretending to be using Turbo C++ for the longest time.
DataCAD is written in Delphi. I wonder how hard a port would be?
I bet you have a NY Times registration. Is VA software getting kickbacks for clickthroughs?
I think so.
I love the smell of choice in the morning :-)
Never Winter Nights gameeditor for linux hasnt been released because of this. It was build with c++ builder but they couldnt release it for linux because they didnt have Kylix yet to allow. Yew! Wonder who fast isle can provide the editor ;))
yush
They dual license the libraries so they can be used either GPL or proprietary... you really should know what you're talking about before you flame...
So you'd use a proprietary product, just not if it uses another proprietary product? (that is distributed under a very similar, and even less restrictive, license!)
How is Borland any less evil than Troll Tech?
Huh? How is Troll Tech evil? People wanted QT under the GPL, and lo and behold, they released it under the GPL. Seems like a nice bunch of folks to me.
Please stop trolling.
hosted on Sourceforge http://freeclx.sourceforge.net/
Grrrrrr.....
anyhows
C++ looks good, C Builder has always had far better debuging tools than Delphi, I hope Kylix C++ has decient debugging.
The professional version now has a postgres driver, there was a serious lack of DB drivers in Kylix 1 professional.
Kylix 2 had an odd dependency on Wine I hope thats fixed now.
Looks like it's time to upgrade that Kylix 1 Pro box set I've got sat on the shelf...
thank God the internet isn't a human right.
Perhaps more important, on the C++ side, is the general fact that code compiled with one C++ compiler is not likely to work with code generated by a different compiler. C++ specifies the use of name mangling, but doesn't specify how that is to be implemented. So all your class names and so on are represented in some compiler-specific way in the shared libraries, making them inaccessible to programs compiled with a different C++ compiler, unless the makers of the two compilers have gotten together and somehow agreed on exactly how to do the name mangling (which has not happened to date, that I've heard of).
I can't speak for C support, but the failure of Borland C++ to support the GNU compiler is inevitable, and not anyone's fault, except maybe Bjarne Stroustrup's.
one hundred twenty
is just enough characters
to write a haiku
exporting the project to GNU makefile....
:)
several shortcuts and completion style things
debugging spawn processes, connecting to running processes. Almost anything debugging related it appears
dataaware compents such as labels (!), edit boxes, listboxes...
and a lot of other nice features.
No complaints, they are trying to sell it
-- Who is the bigger fool? The fool or the fool who follows him? --
That's why it's an odd dependency, Kylix 1 didn't need wine for the GUI (unless it was static?)
It wouldn't be so bad, if you were complaining that you refused to use a Free product due to its reliance on proprietary libraries. But you're complaining that you refuse to use a proprietary product, due to its reliance on proprietary libraries. (Why was this modded up?!? Hey.. have I just been trolled?)
It reminds me of a joke (by W.C. Fields, I think?):
A guy asks a lady, "Madam, would you sleep with me for a million dollars?" She thinks about it, and answers that she would. Then the guy asks, "Would you sleep with me for ten dollars?" Insulted and indignant, she asks, "What sort of woman do you think I am?!" His reply: "We've already established that. We're merely haggling over the price."
As copyright owner of this comment, I authorize everyone to defeat any technological measure which limits access to it.
While I'd aggree that Microsoft makes it 'cheep' to by Visuial Studio, they don't make it that cheep to buy the platform licences.
Also Visual Studio has the crapest development environment I've ever used and the help is shite. It's easier to write code in notepad than VC++, even the windows 9x version of notepad with the 64k file size limit.
thank God the internet isn't a human right.
Borland doesn't use standard C++, well that's not quite true, there C++ compiler used to be and probably still is the most ANSI compliant but Borland have there own C++ extensions.
Closures
and
Properties.
a closure works like this
struct closure{
* thispointer;
* functionpointer
}
calling closure(1,2,3) effectivly does thispointer->function(1,2,3) or
functionpointer(thispointer,1,2,3);
Properties are user friendly, natural getters and setters.
I don't think that GCC will works with them...
thank God the internet isn't a human right.
Shut up you idiot.
Using the borland framework has NO license issues with Qt under any situation. Borland worked that out with trolltech. If you bypass borland's library and talk directly to Qt, only then are you bound by the Qt license. Otherwise, quit worrying it's all ok.
A search of google did reveal however that there is a shocking number of companies who seem to believe that there is something called "GNU Perl" including apparently IBM. I'm not holding my breath for RMS to spend any of his time correcting this widespread inaccurate credit of Perl to the GNU project.
No they don't...
Maybe the core debuggers the same but CBuilder has a more fully featured debugger e.g. intergrated buffer over run checking and all sorts of other lovely stuff...
So ya don't want to register to download the free Open Edition (GPL'd) or the trial version of Kylix 3.0, but you want to know more? Here ya go:
Screenshots of Kylix IDE...(an alpha version)
http://www.drbob42.com/kylix/hotshot.htm
Features of Kylix 3.0...
http://borland.com/kylix/pdf/kyl3_feamatrix.pdf
Or, go here and take the Product TOUR (if you have time to view their Flash 5 presentation and get all the sales-speak you can swallow i.e. good to show to boss).
Terms and conditions
:).
As a condition to download or receive an activation key for Kylix 2 Open Edition, we are asking for your consent to be contacted from time to time by us and/or one of our affiliates, representatives or partners, by either email, postal mail, telephone and/or facsimile, to ask you about your experience with Kylix 2 Open Edition and/or inform you about other products and/or services offered by Borland or its affiliates or partners that we think may be of interest to you. If at any time you wish to revoke your consent, you may do so by updating your account profile.
I hereby consent.
I do not consent.
Read another way "We'll give you this free, but we'll bombard you with spam, snail mail and phone calls." I don't mind the first two as much (filters work well for spam), but my phone number is my precious. Then again it is a $10,000 fine to make an unsolicited call to a cell phone
Not quite. People really wanted it under the LGPL or BSD licenses, just like GTK+, FLTK, FOX, wxWindows, etc.
One of the problems (unless you follow Stallman's manifesto) is that although the Free version is free for open-source, their commercial licenses are structured so that if at any point in time your software project is touched by a free (free, non-commercial, acedemic, etc) version of Qt, you may never at any later time buy a commercial license and release your software commercially.
Anybody know where to download the open version from? It's not on borlands site!!!
Have they finally improved stability and speed? Kylix 1 and 2 both felt very slow (because of all the Wine stuff), and somehow Kylix 2 crashes when I try to save a project.
To be honest, I'm disappointed at Kylix. It does not provide the quality I expect from Borland.
Somebody please tell me they've improved stability and speed!
And what's new in Kylix 3? Is there something new in Kylix 3? I don't even know what has changed between Kylix 1 and Kylix 2 (certainly not stability).
C++ ABI in Linux has had many issues. C++ binaries produced in EGCS, GCC 2.95.x, 2.96.x, 3.x, and possibly the upcoming 3.2 (I read something about breaking compatibility to fix a bug in the mailing list) are all incompatible.
How does Borland address this issue? Or does it only produce binaries for one specific configuration?
Borland's planning on insuring all future programmers? Man, how do I get in on that gig? You can never have too much insurance, I tells ya!
Now your talking shite.
Visual studio is
nested MDI with no optout option, this is great if you suffer from claustrophobia and hate those evil shitty docking windows that just won't go where you put them.
Hard to find you way around.
Piss poor HTML help
Not visual.
Has hidden options all over the place.
Certainly KDevelop, JBuilder don't follow that design, i haven't used Eclipse, ProjectBuilder so I can't comment.
Use somthing like CBuilder and you'll find out what an IDE should be like, and don't go crying to mummy when billy boys cock doesn't taste so good
Nag screens are a relatively new invention to shareware, it doesn't define the genre.
Before nag screens were time limited apps.
Before time limited apps there was missing functionality (what the parent was refering to).
Before missing functionality, there was time limited apps.
etc, etc, etc
Soemthing like Paint Shop Pro has run the gambit on how to enforce its shareware over all the versons I have seen.
> But you're complaining that you refuse to use a proprietary product, due to its reliance on proprietary libraries.
When will people learn to think???
The difference between a proprietary application, and a proprietary common library, is HUGE.
If, in the future, Borland starts to misbehave (say they get bought by Microsoft), then you lose one application, namely, Kylix.
But if Trolltech becomes a problem, and you have to give up Qt (proprietary version), then you lose access to all the applications that are dependent on it, including Kylix, Opera, Hancom Office, applications from The Kompany, and so on.
Look at it this way. Which produces the greater lock-in on Windows, Borland's Delphi, or the Microsoft Foundation Classes? The answer is obvious.
When I deal with a company like Borland, that is the company I want to deal with. I don't want a third party controlling shared libraries and interface standards on my PC.
Does anyone know if mysql support still requires a particular version?
> People wanted QT under the GPL...
No, we wanted it under the LGPL, like the rest of the libraries.
Trolltech chose the GPL instead, because it let them pretend to be open, when in fact it does the opposite, by forcing commercial developers use a proprietary license for Qt.
I find it interesting that the original post, that raised a valid concern, was modded down to (Score:-1, Flamebait), while the above post, that states an obvious falsehood, gets modded up to (Score:4, Interesting). Does Trolltech employ astroturfers?
This post should be moderated as 1:Informative.
thanks -mike
Which raises the question: If Borland can so easily create graphical Linux applications that don't require Qt (using their own product), then why did Borland tie themselves to an expensive development tool that is controlled by another company?
Don't you think that Borland would have learned their lesson on Windows?
Does anyone know if it is possible to use some sort of toolchain with Kylix to cross-compile software for architectures other than x86?
> They dual license the libraries so they can be used either GPL or proprietary... you really should know what you're talking about before you flame...
But there is nothing in writing, not even in the agreement with the KDE Foundation, that says that the GPL'd version and the proprietary version must remain the same.
Thus, in terms of safety (i.e. freedom from lock-in), the proprietary version must be treated as if it were a separate product. Proprietary Qt can't be redistributed, it can't be forked, and companies that use it for closed-source development can't legally switch to the GPL'd version.
You really should think before you flame...
I don't know about the other apps you mentioned, but if you purchase Kylix, regardless what happens to Trolltech, you have the right to create and distribute apps you create with it however you like. Any licensing changes made by Trolltech can't apply retroactively, so the stuff you get with Kylix will still be fairly licensed to you to use in whatever manner you want. Okay, you may not be able to link with their latest libs, but you'll have the right to distribute the libs you got with Kylix with applications that use them anyways!
This won't be a big problem because you won't be able to link with those libs with g++ or gcc -- only C++ Builder for Linux. As for Kylix 3 Open, the issue of linking with QT is already covered by Borland's licensing restriction on it which requires that applications built with it be GPL'd, which goes back in synch with TrollTech's license policy anyways.
File under 'M' for 'Manic ranting'
All I can find is the 2.0 download which is just Delphi.
Your going to have to run the applications like this:
myapp -ns
splash screens suck - brokenware
We're close with this. In some glorious future, there will be a C++Builder that doesn't care if it's running under Linux, 'Doze, or OS X for that matter.
While we're at it, the groovy IDE is nice, but making all of the plumbing interchangeable is a Good Thing.
Don't mind paying for well thought-out product, just don't want the blood-on-goatskin experience of dealing with Redmond.
I guess CodeWarrior specializes in that sort of platform gymnastics, but their pricing for the Palm version didn't excite a purchase out of me...
Anyway, I had gaffed off the upgrade from C++Builder 5 in anticipation of this...
Get thee glass eyes, and, like a scurvy politician, seem to see things thou dost not.--King Lear
.
Delhi is in India, but Delphi is in different country.
It is in BorLand.
Please read my post again. I didn't say one word about applications built with Kylix.
I was talking about whether I could continue to use Kylix itself, if it happened that Trolltech took proprietary Qt in a direction that I didn't like (remember: _proprietary_ Qt can't be forked).
Say I allow myself to become dependent on a number of applications that, in turn, depend on proprietary Qt (e.g. Kylix, Opera, etc.). In that case, I now have two choices:
1. Take my Linux system in whatever direction Trolltech wants -- pay a fee, accept their APIs, use Passport, whatever. Or...
2. Lose the use of future versions of all my Qt-based closed-source applications.
I may decide that the second choice is too costly (as some have decided it is too costly to switch from MS Office). If I do, then Trolltech owns me, and I am back in the same situation that I was in on Windows, where Microsoft owned me.
I am willing to use proprietary applications, where the only thing I risk is the loss of that one application.
But proprietary shared libraries are out.
That's exactly what I was saying. If you've paid for Kylix, you'll have Trolltech's commercial license already. So you _can_ take option 1, because Trolltech will have already received compensation from you indirectly through Borland (assuming, of course, that you did get the commercial version).
File under 'M' for 'Manic ranting'
I became quite excited to try it. While being completely satisfied with the GNU environment, I'd *really* like to use a fast, straightforward development environment, such as Turbo C 2.0 (do you guys remember *that*?) I know several years passed by since then, but thought Kylix is getting closer and closer to something that makes sense just like their DOS based tools did.
:-)) and reliability of an app that's apparently been "ported" from Windows. For developing C++ applications on/for Linux.
I usually stop the installation of a "commercial" application when it fails to get RPM packages housed on my Debian (sid) system. I didn't stop now, backed perhaps by my respect of their good products (Delphi excluded).
Kylix's rpm based install tool failed, too (surprise!). I didn't stop tough, not even after hacking together a wrapper script around rpm (in order to get along the "failing" package dependencies (needless to say I have all neccessary libs/packages installed). I even tried to alienalize & install the provided rpm's -- no luck either (postinstall scripts still failing).
I still didn't give up at this point, had a look inside the package contents for another way to get around. After I've realized they contain wine libraries, all my interest & excitement suddenly disappeared. In fact, turned into massive disappointment.
The GNU toolbox works perfectly, there are excellent, graphical, native debugging tools too (DDD, for instance)... no fancy IDE, F9 shortcut, sure, but anyway... why should I bother?
Which brings us to the same old mantra about RedHat vs other distros. Not to mention the quality, resource footprint (yes, I've upgraded my box since the TC2 times
Not for me, thanks. Am I the only one?
- Can't trace preprocessor code during debugging.
- Can't inspect the value of a #define during debugging. I prefer const variables.
- You undermine the code completion because it doesn't know about the existence of the property. When you press '.' or '->' you will not see your property in the popup window.
- Can very easily introduce obscure bugs that can be hard to find.
Take a look at MFC message maps to understand why using the preprocessor as a language is a mistake. I'm sorry but I disagree with you. Altering the language can be a good thing. Adding a new feature is not the only criterion for altering the language. If the alteration helps productivity then I am for the alteration being done. C++, especially, could really use it.It's slick to be able to port our apps easily in Delphi and C++ from Windows to Linux, but for those who are used to developing there, they still have to pay the big bucks for a C++ Builder or Delphi license under Windows -- Kylix Open or not.
It should be noted that Borland also releases free as in beer "Personal" versions of C++ Builder and Delphi.
... applications". Basically they include all the standard gui VCL components included in the non-free (Professional and Enterprise) versions of C++ Builder and Delphi (just with a different license) minus the data access / web components. Having used Delphi 5 Enterprise extensively at work, I can vouch that for real-world gui and database intensive application development, these features are invaluable timesavers. However, in an educational environment as suggested by the parent post, where the focus would almost certainly be beginning or at most intermediate programming (eg. general language concepts), the Personal versions of either of these would more than suffice for any classroom needs.
Both are similar in features and license to Kylix Open Edition (the Delphi personal description specifically mentions its intended use is to create "non-commercial Windows
In other words, while I wouldn't expect these to provide much of a shot in the arm for schools migrating to Linux, it speaks volumes for Borland and their commitment to Linux (Kylix) and its ideals (the open source-ish license of the Windows-based Personal versions).
This is not offtopic. I don't want to imagine the nightmare that is the half-assed code you sling.
LURN TOO SPELL!!!And stop using Mozlla
Does it still have the clause in their open-edition licence that they have permission to come in & search your premesis (to check for commercial usage) if you install their software?
===> An eye for an eye makes everyone blind - MG
Well, if you'd like to stick with Borland, there's always 'J'Builder. It runs on the platforms you mention.
:)
Oh, C++, my mistake.
But when do the training wheels come off? Either you learn to adapt to new environments easily, or you use an environment that will always be there and will always provide what you need it to do.
.NET and C#. Yet Unix people are still using the same tools they've always used, emacs or vi, and just drop in a new compiler and extend the syntax rules in the editor for the new programming language. On proprietary systems, developers purchase a new system.
While you can rely on a language, you shouldn't need to rely on the tool. Especially proprietary tools.
This is probably one reason a lot of really good developers prefer the unix tools, mostly emacs or vi rather than the latest fad, because a fad is exactly what it is. One year its C++ and Visual Studio, the next year its Java, now its
So I think you're right when saying that developing in a "visual" environment is easier and you can catch on to programming quicker. But I believe that investing some more time into a "unix" tool is more valuable for your programming career. Actually, this isn't my idea--something I read on USENET.
And that's why I am spending time learning emacs. Because I don't think I'm wasting my time.
How is this unfair? Is leeching a basic human right?
Yep, you want something that is simple enough to not overload your brain with details while you are trying to grasp the general principles.
Examples:
Python's Tkinter module (from within IDLE or just a Python shell)
FLUID, the almost unbelievably easy C++ IDE for libfltk).
There are richer IDEs with more features, but in their respective computer languages, these would be hard to beat for simplicity. I think there's a pretty cool Scheme one too, but I forget what it's called.
-- What do you need?
-- Gnus. Lots of Gnus.
Kylix 3 is NOT yet available for downloaded.
If you go to this page
http://borland.com/products/downloads/download_ky
you will see that the latest version of Kylix (open version) is version 2.
So, who has the url for downloading Kylix 3 ?
Care to share ?
Muchas Gracias, Señor Edward Snowden !
The product hasn't been released yet! Not even the non-free versions. The web site doesn't make this clear, I admit.
Troll ? that was no troll. It was an opinion, you know freedom of expression and all that rot.
You seem to be defending Trolltech, yet your lack of understanding of Free Software makes you sound like a Microsoft astroturfer. Strange...
Free Software is not about money.
Free Software is about freedom.
In practical terms, it's about ensuring that you won't get locked in and controlled, the way Microsoft uses their software and protocols to control their users.
Nobody cares about the fact that Trolltech is charging for their product. If I was bothered about people charging for products, then why would I be defending the freedom of commercial closed-source developers?
What I care about is that Trolltech is following the same lock-in scheme as Microsoft, by getting commercial developers hooked on a proprietary version of Qt. Plus, Trolltech is using the GPL'd version of Qt as a smokescreen, to try to hide their lock-in scheme.
A proprietary closed-source applications is not that big a problem, because you know up front what you are getting into, and how how much you are getting locked in. I have willingly paid for such applications, for example, I bought Applixware, which I considered a safe bet because I knew that the file-formats were documented.
But proprietary shared libraries are very bad, because it gets hooks into multiple applications from multiple companies. With proprietary shared libraries, you can become totally locked in, and never see it coming. That's how Microsoft gained their control over Windows users.
As for this MS astroturfer bollocks, you might find out when you grow up that the world is not some big "Microsoft vs, You" contest. How my comments could be interpreted as you have interpreted them is beyond reason.
If Bjarn Stroustrup has anything to say about it, it won't happen. I wrote him an email about this a couple years back and he replied that he believes (or believed anyhow) that properties don't belong in C++. Without his support it's unlikely it will become a standard any time soon. Unfortunate :-( I think it's just a case of providing us with as many tools as possible and laying the responsibility of proper use on the developer, which *is* what C++ is all about, right?