Open Source IDE GAMBAS Reaches 3.0

Kevin Fishburne writes "After years of work, creator Benoît Minisini and friends are just in time for New Year's celebrations with the first stable release of GAMBAS 3. Per their web site, 'Gambas is a free development environment based on a Basic interpreter with object extensions, a bit like Visual Basic (but it is NOT a clone !).' GAMBAS is component-based, so check out the list for an idea of what you can do with it."

Comments

BASIC

[InterestingFella]

It's what I started my programming history with and I still have fond memories of it. Easy enough language that got me interested in programming and provided me instant fun. There never really was any other such comprehensive language with quick-to-see results. Drawing on screen was easy, syntax was easy and reading from input was easy. You got fun things done quickly. As much as some "I'm better than you" geeks like to take a stand about it, BASIC was important part of history.

Re:BASIC

[gadzook33]

Yes, BASIC was great (for the role it played). VB is an abomination that should be wiped from the face of the earth. I mean really people, learn something that starts with 'C'.

Re:BASIC

[Anonymous Coward]

COBOL?

Re:BASIC

[icebraining]

ColdFusion.

Re:BASIC

[Yetihehe]

AMOS for Amiga was even better, it had some multimedia extensions (playing sound, sprites, bobs (blitter objects - like a sprite, but could be the size of a whole screen), mouse zones (for fast mouse position testing for buttons) and even special instruction which waited for vertical refresh. It was about as easy as basic.

Re:BASIC

[arth1]

But it's not BASIC. It's an IDE built around BASIC with extensions. To say it's BASIC is a bit like a canoe with an outrigger is to canoeing.

Re:BASIC

[gadzook33]

See, Caml would have at least been a worthy joke.

Re:BASIC

[arth1]

Brilliant idea.
A couple of other popular choices include csh and CL (common lisp). And from my mainframe days, I'd say that CICS is almost a language of its own too.

Re:BASIC

[Anonymous Coward]

Common Lisp thank you.

Re:BASIC

[hedwards]

I suppose you also fondly remember running with scissors and poking eyes out with sticks.

Re:BASIC

[Anonymous Coward]

No, the true abomination is PowerShell.

Re:BASIC

[QuantumRiff]

But.. Powershell is like Perl had an affair with Bash, and didn't tell her husband, DOS.

So Dos helped raise and shape that little bastard, like it was its own :) Unaware of why it was so different..

Re:BASIC

[Mitchell314]

But a lot of old BASIC dialects were similar, they were variants of the same language packaged in a development environment. If a language can load arbitrary C libraries, then that opens up a *ton* of doors.

Re:BASIC

[moderators_are_w*nke]

AMOS was where I learned coding, before moving on C

Re:BASIC

[Anonymous Coward]

I've had a look at Gambas3 just now. You know what... it worked first time and was dead-fucking easy to get a gui running.

People knock Visual Basic for Applications (built into Excel/Word)... but so many quick and dirty applications get built with it because:

1. The component style of programming is extremely powerful - particularly when you can call upon very flexible and high level Office components. Especially when you have a good IDE - which VBA does.

2. It's shockingly easy to get your first application running (don't knock it... businesses run on the phrase "get it done now" - the first thing that works gets used).

3. You can give someone, for example, one MDB file and they click on it... and it runs (same with a spreadsheet). Easy to develop, easy to distribute.

Slag it off all you like. Microsoft does understand how to get people to use it. It's something that Free software developers never seem to learn. Maybe Gambas has?

Re:BASIC

[geminidomino]

You think having the real daddy's regex would have been such a giveaway. I guess sometimes you just really don't want to know!

Re:BASIC

[HiThere]

The component style of programming does have LOTS of advantages. It's the BASIC part that I can't stand. If they replaced it with Python, Ruby, ... even LISP or FORTH, then I'd seriously consider it. C might be acceptable, but I sort of doubt it. It would depend on the fit, and I don't think that C woruld fit. But Pascal might work. (N.B.: I don't consider Lazarus to be a good fit, and it would need to handle Unicode strings. So the easiest language to adapt would probably be Python3.)

Please note that I didn't mention Java. I have not seen an acceptable implementation of this kind of thing for Java, and there have been many of them. Perhaps the language needs to be basically an interpreter based language. (This doesn't mean that it can't be compiled, this means the language design is aimed at an interpreted implementation. So of the languages I mentioned I consider Python the most promising, and Ruby the second most. [Ruby is deficient in the handling of Unicode. It CAN do it, but it's not as designed for it as Python {even Python2, but especially Python3}.])

BASIC, however, is not a good base language.

Re:BASIC

[jbolden]

BASICS are good to write because

1) Syntax is well defined and thus lexiconical interpretation is easy.
2) There is an extremely low number of keywords
3) It is well understood.

LISPs are by far the easiest interpreters to write.

Python and Ruby are orders of magnitude more difficult to write interpreters for. Multiple man years instead of man days.

Re:BASIC

[martin-boundary]

That's a meaningless objection. Which BASIC are you thinking of as canonical?

The original language (and first historical version) is Dartmouth BASIC, which went through several version changes, adding new features over time. In its final version, the original BASIC was renamed True BASIC. The dialects available on micros didn't necessarily implement all the features of Dartmouth BASIC, for example matrix operations were missing from GW-BASIC or Commodore BASIC, IIRC.

Re:BASIC

[icebraining]

But why would they write their own interpreter instead of writing their IDE around libpython?

Re:BASIC

[theshowmecanuck]

Oh that's crap. And I learned C as my second language after VAX BASIC, and worked professionally as a C programmer on Unix for years. VB is a decent language/environment and has its place. Don't blame the tool because the worker holding it has no talent or other issues. Or to draw from one of my favourite sayings, it is the carpenter who cut the board three times and still found it too short. But it sure as hell wasn't the saw's fault. The worst you can really say about VB is that it allows bad programmers to create working programs in a similar way that poorly written SQL can appear to work until you realize the result set is missing data that you need (which was a similar complaint some C programmers had about Java when it first came out). But in the hands of a competent programmer there isn't really anything wrong with it. I don't use VB, but I have seen it used well. I have seen C used really freakin' badly too. So give up already.

Re:BASIC

[HiThere]

Why start from scratch?

The subject, as I understand it, is "What is a good component language, given that some people really think that BASIC is a bad choice." If you don't think that, then of course it's going to seem like a good approach to you. And this thread has nothing to offer. I happen to think that component style programming could be quite useful, but I don't find BASIC an acceptable controller language.

In some ways the component approach fits in well with the characteristics of BASIC, I just don't like the language. And Python and Ruby both seem like decent substitutes. They seem to share all of the advantages of BASIC, and have a minimal number of the faults. LISP is another language that would be a good fit, as would FORTH, I just don't happen to like either of them. Note that in none of these languages is there the required declaration of variables. That doesn't seem to fit well with the component style, and may be why I don't think the Delphi/Lazarus attempts to do this with Pascal were satisfactory.

Note that it's also possible to go too far in the direction of flexibility. I think that Squeak Smalltalk went too far. Also notice that in none of these languages is efficiency the major consideration. This is appropriate for a language that is going to be controlling a GUI, but I might be extremely wise if it's easy to hook in routines that were compiled in a language that *DOES* consider efficiency important. Unfortunately this is much easier said than done, especially if you want the strings of your GUI controlling language to be utf8 strings (which I do). Handling utf8 strings in C still seems to be an "occult art".

For that matter, if all you're interested in is how easy the interpreter is to write, then you should pick FORTH. It's much easier to get a minimal complete system up in FORTH. You didn't even seriously consider that, so you have some other reason for selecting BASIC. (Not saying it's not a choice that a lot of people won't like. Just saying that *I* don't like it. And that that's why I've been ignoring it despite the many attractions.)

P.S.: A small part of my feeling towards BASIC is due to the MS AccessBasic I programmed in about a decade ago (well, I guess a decade and a half now). I would write a routine and it would work perfectly...for awhile. Then I'd need to write it out to text, delete the version in the system, and read it back in from text. No changes were either needed or indicated, but until I did that it would stop working. Another time I caught it adding two 3 digit numbers and getting the wrong answer...which caused me to wonder just how many times I *hadn't* caught it. At that point I had gotten close to the point of total distrust. I suppose that it's unfair to blame it on the language, when what it was was the MS implementation, but I still can't manage to trust any results that come out of BASIC. (My suspicion was that the compilation stored the code invisibly in the same file as the source, and the compiled code was somehow becoming corrupted, but, of course, there was no way to tell whether or not this was what was actually happening, and even if I had been able to tell, it wouldn't have been useful information, because I still wouldn't have any other way to fix the problem than delete the subroutine and reload it.)

Re:BASIC

[HiThere]

Why start from scratch?

The subject, as I understand it, is "What is a good component language, given that some people really think that BASIC is a bad choice." If you don't think that, then of course it's going to seem like a good approach to you. And this thread has nothing to offer. I happen to think that component style programming could be quite useful, but I don't find BASIC an acceptable controller language.

In some ways the component approach fits in well with the characteristics of BASIC, I just don't like the language. And Python and Ruby both seem like decent substitutes. They seem to share all of the advantages of BASIC, and have a minimal number of the faults. LISP is another language that would be a good fit, as would FORTH, I just don't happen to like either of them. Note that in none of these languages is there the required declaration of variables. That doesn't seem to fit well with the component style, and may be why I don't think the Delphi/Lazarus attempts to do this with Pascal were satisfactory.

Note that it's also possible to go too far in the direction of flexibility. I think that Squeak Smalltalk went too far. Also notice that in none of these languages is efficiency the major consideration. This is appropriate for a language that is going to be controlling a GUI, but I might be extremely wise if it's easy to hook in routines that were compiled in a language that *DOES* consider efficiency important. Unfortunately this is much easier said than done, especially if you want the strings of your GUI controlling language to be utf8 strings (which I do). Handling utf8 strings in C still seems to be an "occult art".

For that matter, if all you're interested in is how easy the interpreter is to write, then you should pick FORTH. It's much easier to get a minimal complete system up in FORTH. You didn't even seriously consider that, so you have some other reason for selecting BASIC. (Not saying it's not a choice that a lot of people won't like. Just saying that *I* don't like it. And that that's why I've been ignoring it despite the many attractions.)

P.S.: A small part of my feeling towards BASIC is due to the MS AccessBasic I programmed in about a decade ago (well, I guess a decade and a half now). I would write a routine and it would work perfectly...for awhile. Then I'd need to write it out to text, delete the version in the system, and read it back in from text. No changes were either needed or indicated, but until I did that it would stop working. Another time I caught it adding two 3 digit numbers and getting the wrong answer...which caused me to wonder just how many times I *hadn't* caught it. At that point I had gotten close to the point of total distrust. I suppose that it's unfair to blame it on the language, when what it was was the MS implementation, but I still can't manage to trust any results that come out of BASIC. (My suspicion was that the compilation stored the code invisibly in the same file as the source, and the compiled code was somehow becoming corrupted, but, of course, there was no way to tell whether or not this was what was actually happening, and even if I had been able to tell, it wouldn't have been useful information, because I still wouldn't have any other way to fix the problem than delete the subroutine and reload it.)

Re:BASIC

[gambix]

Because we don't like Python ?... For challenge too... To create something new, that fit our programming way. There is tons of available reason. Because world need outsider, some people that not systematiquely do the same than other. Often they fail, and the project death, but on some few time the project match a real need and find users. Gambas is not a new project. We are working on it since 10 years now, and i can say i really have fun to work with it ... the rules for us is just to take fun on working with it... and it's really a pleasure. In fact i wan't to ask a question ... Why python have not an IDE like Gambas ??? Python have 10 year more that gb ... A great language for sûre, but .... Well i'm a men that don't like python so scuse me :)

Re:BASIC

[icebraining]

Have you read the comments I was replying to? I wasn't the one who suggested using Python, that was HiThere. I was just replying to jbolden who said writing your own Python interpreter is hard.

Re:BASIC

[jbolden]

Because remember the idea was a scripting language for a new application. If I write ABC app, ABC doesn't have any hooks into python. I have to extend python and then embedded an entire python. I have to worry about all sorts of things like python security because it is now running inside my app, and all the features of this embedded libpython and how they interact with features of my API.

And all that to avoid whipping off a custom BASIC or LISP?

Re:BASIC

[jbolden]

There are python IDEs, but in general there aren't good IDE for any of the dynamic languages. The reason is that IDE technology makes heavy use of static language features. It is an unsolved problem how to create a powerful dynamic IDE.

Re:BASIC

[jbolden]

I was saying that a BASIC is cheap to implement. In terms of the problems with Access and math, yes I've had serious problems too. Those BTW SQL Server as well. They weren't problems in BASIC they were problems in the underlying old versions of the C/C++ math libraries that had been compiled into those apps that the controller languages called. That is absolutely inexcusable on Microsoft's part but it is not a problem of BASIC.

____

So in terms of why BASIC, that was my #3 "3) It is well understood." People may not like BASICs but they know them. FORTH is an easy interpreter to write, RPL which is a FORTH derivative is one of the best controller languages ever. But... FORTH is much more poorly understood. It really is a huge hurdle, like losing 2/3rds of your possible users to have to do "2 3 +" over "2 + 3". And that same thing bites you with a LISP. I really thought Apple was on the right track with Dylon where they were creating a LISP with Algol syntax, an idea I think Perl, Python and Ruby proved correct for general programming.

If I had to pick something other than a BASIC, I'd use Javascript / XML because again it is well known.

Now if the language is a bigger deal, that is I expect people to write large applications in it... then the quality of the language matters much more than the (3) issues I listed above. I'd probably use a 3 tier system:

a) A macro controller language, something like a BASIC
b) A full featured dynamic language capable of interacting with the APIs of the language. There a Ruby would be perfect.
c) A static language capable of linking in engine components and writing alternate engines. There I'd want something like a F#, Haskell...

Re:BASIC

[icebraining]

No, all that if you want to use Python instead of BASIC or LISP.

I wasn't saying they should use Python, I'm saying that if they did want to use it - like HiThere suggested - there would be no need to write a brand new interpreter, since there's libpython.

Re:BASIC

[jbolden]

I don't think you are getting this.

I have an app, ABC. The purpose of the control language is to manipulate the internal data structures of ABC app. The purpose is not to offer a generic python. Python out of the box doesn't allow me to manipulate ABC's data structures, libpython doesn't let me do that. I need to rebuild the python where I'm changing the internal primitives. I can just link in some library and go. Most likely I want to change what are or are not primitive objects in the language. So for example in Excel control language I'd want sheetname!Column(Letters)Row(numbers) to be a special type of syntax with different parsing rules, a special type of variable. I can't use the generic parser.

What you might be thinking of is, is having ABC offer an API and creating a python library to manipulate that API. In which case the API is essentially the control language. but it can be minimally featured because the feature set comes from the library not the API. And that is totally possible and you could use libpython for that. But... that's an entirely different approach from the ground up.

Re:BASIC

[jbolden]

I can just link in some library and go

That should read I can't just link in some library and go

Re:BASIC

[ogdenk]

It's what I started my programming history with and I still have fond memories of it. Easy enough language that got me interested in programming and provided me instant fun.

BASIC was a great first language and is still useful in teaching simple programming constructs. It was a better fit for early micros than today's machines however. You could get down and dirty with PEEK&POKE statements and talk to hardware directly. Not quite the same anymore.

There never really was any other such comprehensive language with quick-to-see results.

Ummm.... I have to call BS on that one. Both TCL and LISP come to mind.

Drawing on screen was easy, syntax was easy and reading from input was easy.

Depends which version of BASIC we're talking about. Different dialects had WIDELY varying levels of hardware support. Atari BASIC had easy to use graphics statements, BASIC on the C64 did not. QBASIC and QuickBASIC for DOS had lots of graphics commands and supported old-school line numbering as well as more modern labels.

Compatibility was another issue. A program written in Microsoft BASIC wouldn't run without modification on Atari BASIC for example.

Speed was another issue... different BASIC environments were faster than others. Atari BASIC was slow, TurboBASIC XL was 5 times faster on the same hardware and was compatible.

As far as helping me to understand how a computer really works, assembler was more useful. Not saying force assembler on a 10-year-old or expect to write a masterpiece 3D engine. Doing some simple math, throwing some text on the screen, allocating/writing to/reading from RAM, etc isn't so bad.

 

You got fun things done quickly. As much as some "I'm better than you" geeks like to take a stand about it, BASIC was important part of history.

BASIC just isn't a real good fit for modern environments IMHO. HyperCard w/ HyperTalk on the Mac was interesting and fun to play with but it would stick out like a sore thumb today as well.

A simple high-performance BASIC with an easy to use integrated 3D library that's as easy to use as sprites were might get people interested. The problem with getting kids interested is that most of them want to write games. Writing 3D games is tough.... creating 3D models and environments is tougher. There's no QUICK gratification like you had with simple low-rez sprites back in the day. When kids have dual-screen stereoscopic 3D handheld game systems, spending a week learning to move a sprite around seems dull and irrelevant.

Re:BASIC

[icebraining]

Well, no, if you want to change the syntax or semantics of the language itself you can't use libpython, but then you're not using Python, you're using your own derivative language.

Re:BASIC

[obsess5]

Implementing a minimal FORTH interpreter (actually a compiler to intermediate representation) might be simple, but a full-on ANSI FORTH interpreter is not so easy. I used to read comp.lang.forth and there were frequent postings from people who were trying to write their own interpreters who got stuck. FORTH is also not the easiest language to learn, the addition example you mention being the least of the hurdles. FORTH is not fully postfix and when you get into creating words, things get hairy. I enjoyed using the language, but I think LISP or SCHEME would be easier to learn. (And, in my experience, SCHEME was easier to learn.)

Re:BASIC

[jbolden]

OK good, that's exactly the point. You want your application's data objects to be the primitives in your new derivative language. For example in a shell you want files to be primitive objects, while in most other languages you wouldn't. And in a spreadsheet you are going to want cells to be primitives. That's a good way to say it, you want to use a BASIC because you are going to need create a derivative language which means you can't just pull in a library from a preexisting language.

Re:BASIC

[jbolden]

Remember the point is an embedded language so it won't be an ANSI FORTH, it must different because it is going to be using the application objects as primitives. I think FORTHs are amazing at new primitives.

Anyway, yes I agree an LISP is easier to learn. Especially since you can use XML syntax.

Re:BASIC

[theshowmecanuck]

I see a moderator doesn't agree with me but doesn't understand that just because they don't agree doesn't mean I should be modded out of the discussion.

Re:BASIC

[mcn]

Pascal? I certainly think that is a good programming language for learning purposes... at least back in the days when Java wasn't born yet... It's readable, non-cryptic...

Re:BASIC

[HiThere]

FORTH actually *is* fully post-fix, but the mingling of compile-time vs. execution time can make that difficult to see. And the last time I tried one just couldn't embed a " into a string. (Admittedly, that was over a decade ago, and my "standard FORTH" was FIG-FORTH. I don't have any real knowledge of ANSI-FORTH.)

I actually found FORTH easier to wrap my head around than LISP. I've never given Scheme a fair trial. But after Neon (an object oriented FORTH for the Mac) died during it's transition to MSWind95, I lost interest in Forth. A lot of good things died trying to make the move from Mac to MSWind95. Another language that died trying to do that was Prograf, a dataflow language. (To me it seemed that it's real problem was not it's platform, but that it didn't have any non-graphical representation accessible to end users. This made even simple programs insanely large when printed out.) If it had just lived it would have been an ideal language to use for programming multiple-cpu computers. (Outside of the representation problem.)

Excellent resource for beginners

[oakgrove]

Gambas is a great basic-esque ide for beginners and on any debian derivative is just an apt-get away. Of course use your respective repo tools elsewhere.

Re:Excellent resource for beginners

[Anonymous Coward]

it has the most disgusting website ever. Who thought that intentionally blurry text is cool? It just makes it hard to read.

Re:Excellent resource for beginners

[tomhudson]

it has the most disgusting website ever. Who thought that intentionally blurry text is cool? It just makes it hard to read.

CTL+A is your friend. Or just remove the text-shadow:0 0 6px black; using firebug. Because it really IS that ugly!

Re:Excellent resource for beginners

[arth1]

It's not blurry text. It's text-shadow.
Depending on your font, font size and display DPI, that might make it hard to read, yes. The lower DPI and smaller font you have, the worse it's going to look.

The error here is the typical web design 101 blunder of combining scalable and fixed size elements.
I am disappointed that HTML and CSS after all this time still lets you do that, cause the result becomes "looks good for me, looks shit for you" pages.
Where's HTML/CSS with strong typing? It shouldn't be possible to affix a shadow in px to a font in pt - IMO, the browser should then discard the text-shadow.

Re:Excellent resource for beginners

[Anonymous Coward]

Whats WORSE, is they use flash !!!
For a darn IDE image??
In heavens name WHY ??
Web page designed by 13 y/o ??

GET RID of the FLASH !!!

Re:Excellent resource for beginners

[InterestingFella]

Great example of how "greatly" open source devs get UI and design, tho

Re:Excellent resource for beginners

[Anonymous Coward]

Yeah because as we know, all proprietary ui's are spectacular. We also know you are a jackass, cmdrpony.

Re:Excellent resource for beginners

[tomhudson]

I wouldn't limit it to open source devs - Microsoft has produced its' share of turkeys. So has Apple (hockey puck mouse, anyone? How about the "toaster"?)

Devs can produce ok UIs provided that they follow the decades-old rule that form should follow function. But how many times have you been presented with a "finished" design that is non-functional by some "design team"?

Re:Excellent resource for beginners

[tomhudson]

It's nothing to do with scaling. I have my dpi set at 120 (instead of the standard 96) AND I have my browser set to zoom in an additional 40%, and it's STILL blurry. I have enough problems reading text w/o that sort of silliness.

Re:Excellent resource for beginners

[geminidomino]

Firefox 4
Firefox 5
Firefox 6 ...

Re:Excellent resource for beginners

[Anonymous Coward]

Right, because the design of the Gambas website is the result of the unanimous agreement of all open source developers in the world.

Re:Excellent resource for beginners

[Anonymous Coward]

All of which are functional and fine to use, with UI improvements rather than worsening.

Re:Excellent resource for beginners

[Anonymous Coward]

Holy shit, I thought you were exaggerating. There needs to be a Bulwer-Lytton style contest where sites like that can compete to see who can cause the most lost lunches.

That is just... astonishing.

Re:Excellent resource for beginners

[DrXym]

Firefox is an example of an open source app that gets usability. Compare Firefox to Mozilla that preceded it. It's a vast improvement in terms of usability, jettisoning menus, buttons and obscure advanced features and concentrating on delivering a premium browser experience. People who want extra features can install extensions but the default is minimalist. Problem with Mozilla (and now Seamonkey) is that it merged the functionality of a browser, HTML editor and mail app into a single executable and was subsequently very cluttered.

And subsequent iterations of Firefox have simply improved the experience.

need a good C++ IDE, though

[Anonymous Coward]

Nothing against BASIC, it's what I learned on in the 70's, but most really serious projects are done with C or C++ based languages. Kernels, device drivers, video games, serious applications, all C++. We miss a really good open source C++ IDE. There are some so-so ones. I've tried several, like KDevelop, but they always fall short of (I hate to say it) MS Visual Studio. For as much as I dislike microsoft, MSVC is the gold standard of IDEs for C++.

Re:need a good C++ IDE, though

[liquidweaver]

QT Designer is quite good.

Re:need a good C++ IDE, though

[Hodapp]

I have found Qt Creator much easier to work with than Visual Studio. It requires some more explicit management in its project files (as they are just QMake projects), but I've found that I prefer this to the endless point-and-click mazes that VS subjects me to.

It's handy enough on its own for C/C++ projects, independent of whether you are using Qt or not, but it's extremely helpful for developing applications that use Qt as it integrates very well with Qt Designer.

Re:need a good C++ IDE, though

[Jorl17]

I fully agree. Visual Studio is amazing, even for someone as "anti-Microsoft" as myself to admit it. Qt Creator is the closest thing I get to pure satisfaction when I code C++ without Visual Studio. It indents it right, it has great themes, great shortcuts, great everything!

Ever heard of ECLIPSE ????

[Anonymous Coward]

Visual Studio may be great for debugging, but is not that great as an actual programming editor. Too many dumb azz steps to get simple jobs done.

Re:Ever heard of ECLIPSE ????

[InterestingFella]

Like what? Visual Studio is by far the best IDE there is. It is almost outstandingly powerful and still lightweight with the newest versions. It's nice to work with. Apart from VisualAssist, which is great addon, there's not much there to improve over others. In fact, the assist features of VS is on par with others, VA just takes it step further. Anyone who would use Eclipse over VS just doesn't know what he is doing.

Re:Ever heard of ECLIPSE ????

[dokebi]

Anyone who would use Eclipse over VS just doesn't know what he is doing.

Or maybe I use linux for development? You know, like a proper geek?

Re:Ever heard of ECLIPSE ????

[HisMother]

Not to go too far off topic, but my few experiences with VS have been abysmal: the editor sucks (the selection behavior alone is enough to make me want to throw the keyboard through the monitor), search is hellishly clumsy, there's no completion of any kind, it rarely sees symbol definitions it should, etc. It's so bad, that I suspect it has to be my fault: it's unintuitive for me, and I don't know how to use it. Can you recommend a reference for an accomplished programmer who just doesn't understand Windows tools?

Re:Ever heard of ECLIPSE ????

[Anonymous Coward]

I use botg vs2010 and Eclipse daily. I prefer Eclipse. Please tell us lost souls what else we are doing wrong. I'm waiting with baited breath for your pontification on emacs and vi. Also, eat shit you arrogant prick. Many developers that would code infinite circles aeound you know what they are doing, use eclipse and would laugh in your face over your fanboy partisanship. A shit dev blames tge tools.

Re:Ever heard of ECLIPSE ????

[laffer1]

I don't understand why people think Eclipse is so awesome. I use it at work currently and when I get time, I'm going to switch away from it.

1. Eclipse is too modular. It can't do anything on it's own and the hundreds of plugins aren't tested with each other. This causes bugs, incompatibilities and UI integration problems.
2. Eclipse has gotten better, but it still suffers from the refresh problem. It doesn't poll for changes in source files frequently enough and if you dare add a file outside the IDE, you have to manually refresh the view to see it.
3. Eclipse wasn't written with swing and requires SWT which means that you can only run it on platforms that SWT has been ported on.
4. Eclipse is not intuitive. Things like wizards don't behave properly. When writing Java code, one would assume that Apache Axis 2 projects would be supported with the latest web project type. They're not. You can't switch without recreating your whole project. You also can't generate a client only from an axis 2 project.
5. Eclipse is ugly. It still looks like an IBM product. Intellij, Netbeans, hell even visual studio are more appealing.
6. Every "killer" feature I've seen an Eclipse developer mention is also available in Netbeans. The few things I can't do in Netbeans are third party add-ons that haven't been tested well and don't integrate. It turns into multiple eclipse installs.. one for Java, one for C++, one for PHP, etc. This is wrong. Netbeans got this right.
7. Source formatting in eclipse is terrible. It breaks things up into little tiny lines and wraps things way too much. Java is verbose.. i need more than 80 characters unfortunately to be legible. It's also a hassle to configure this compared to other IDEs I've seen. It's so bad, some people have made plugins just to do that.
8. Eclipse warnings are useless. People get so used to having yellow lines on the side, they don't take any warnings seriously. It causes one to do bad things.

I realize that this is going to start a flamewar, but before anyone tries to say I'm wrong please try some recent versions of other IDEs. Most complaints I hear about Netbeans, Visual Studio, etc. are for very old versions. If eclipse is the gold standard, our standards are too low.

Re:Ever heard of ECLIPSE ????

[Anonymous Coward]

Thanks. I thought I was the only one.

Add the strange tripping-over-itself VS does when building the project and there's a mistake in the code and so it will report a compile error (which is good) and from this point on I can't compile again until I closed and restarted visual studio because some kind of database is locked (at least it says so). No idea how people can get any useful work done in it without despairing.

The resource editor keeps adding #define s to resource.h and never removes them again - so you end up having hundreds of IDs trying to refer to something that doesn't exist.

The strange compiler quirks like non-working template speciaizations, "someone initialized the runtime in the wrong way" errors and abysmal memory management (ever tried allocating memory in a DLL and then freeing it in the main program?)

Never had any problem in Eclipse. Not one in all those years.

Re:Ever heard of ECLIPSE ????

[Mitchell314]

Many developers that would code infinite circles aeound you know what they are doing.

Actually, in code, infinite loops tend to be a bad thing . . .

Re:Ever heard of ECLIPSE ????

[Anonymous Coward]

The "hell even visual studio" remark is unwarranted. I'm no Microsoft fanboy, but let's be honest: Visual Studio is far and away better than any of the IDEs discussed.

That being said, in the Java world, I still prefer Eclipse. I don't think it's awesome, if anything it's deeply flawed. But for my needs, it's still the best thing out there. I do really like what IntelliJ is doing, it's so much lighter weight. But it just doesn't have the plugin ecosystem that eclipse has.

Those useless warnings that you complain about are the warnings that the Java compiler spits out, it has nothing to do with Eclipse. I can't speak for everyone, but I strive to write warning-free code. Eclipse makes this easier.

So in summary: yeah, eclipse sucks, but for many people it's still the best option available. Personal preference plays a big role here also.

Re:Ever heard of ECLIPSE ????

[Anonymous Coward]

visual studio is lightweight

Sputter...cough...lightweight?? What? Please stop with the fanboy bs, soppsa. Who the hell is modding this joke up?

Re:Ever heard of ECLIPSE ????

[tomhudson]

Anyone who would use Eclipse over VS just doesn't know what he is doing.

Or maybe I use linux for development? You know, like a proper geek?

What? You use eclipse when you have the awesomeness of vim available to you? (okay, I admit it - I switched to jEdit recently, but still, for a lot of code, eclipse is massive overkill).

Re:Ever heard of ECLIPSE ????

[tomhudson]

If eclipse is the gold standard, our standards are too low.

"Gold" is our 3rd-lowest level. Sort of like "Professional" - it's really for amateurs. You want AT LEAST Platinum. Or Onxy. Or even the craptastic Diamond level. That will give you the absolute minimum. And for our Enterprise-y customers, there's Prozilla, to Manhattan, and all the way up to Rushmore.

What do they mean? Who knows - it SOUNDS like it should cost more, and that's all that counts!

Re:Ever heard of ECLIPSE ????

[Anonymous Coward]

1. Eclipse is too modular.

Since when has that been a complaint you'd find on Slashdot?

Eclipse isn't a "product". It's more akin to Seamonkey than Firefox.

Re:Ever heard of ECLIPSE ????

[Anonymous Coward]

Welcome to nuSlashdot. Where speaking the truth about Microsoft is a guaranteed -1. Sad. In the name of Malda, $DEITY save us!

Re:Ever heard of ECLIPSE ????

[KermodeBear]

I, unfortunately, have had the same issues with Eclipse. I find it to be incredibly slow, bulky, and difficult to use for even standard tasks.

I do a lot of PHP work. I used to use the old, pre-Eclipse version of Zend Studio which was pretty good. When they moved to the Eclipse platform, it ended being a terrible product. I thought that, perhaps, the Zend people just didn't do a great job building on top of Eclipse and I decided to go to Eclipse directly. What a mistake - even the "easy installers" with all of the pre-installed, pre-configured PHP-specific plugins were a nightmare and rarely worked.

These days, I currently use a simple editor with syntax highlighting. The big, bulky IDEs are just too much, are a pain in the rear to work with, and force you to work within a very specific paradigm. It is their way or the highway. And, many of them don't support simple things like SFTP.

Every year or so I poke around and see if anything new has come along, and I might try a new product or two, but I always end up going back to "simple syntax highlighting text editor".

Re:Ever heard of ECLIPSE ????

[HisMother]

Seriously. Well, bye Slashdot, it used to be fun.

Re:Ever heard of ECLIPSE ????

[datavirtue]

Does that include those who can't bring themselves to install VS? How is it lightwieght by the way? Last time I installed it took two hours.

Re:Ever heard of ECLIPSE ????

[sproketboy]

Wrong. IntelliJ IDEA is.

Re:Ever heard of ECLIPSE ????

[mpol]

I heard good things about PHPStorm. I haven't used it myself, but you could give it a try.

Re:Ever heard of ECLIPSE ????

[Anonymous Coward]

1. yes. solution to this problem is to just write code as it is meant to be written. use it as an editor and nothing else (at first). why are you complaining about plugins you were not using anyway (BE, before eclipse)? its also easy as hell to write your own builders and launchers, and thats what most of the non editing plugins are focused on.
2. wrong. eclipse has auto refresh on the project, it can even run builds on external file touches. (this is not always a good thing, and it is optional behavior)
3. yes. no easy solution.
4. wrong. if your complaining about wizards, dont call yourself a programmer. if your not calling yourself a programmer, dont complain about IDE's.
5. wrong. this is preference, i rank them intellij, eclipse, netbeans, and lastly visual studio.
6. yes. its easier to have multiple specialized installs than to integrate everything into one.
7. wrong. code formatting is a preference and eclipse uses the style advocated by sun (probably the most widely accepted). guess thats oracle now. my style is different from that, but i dont use a code formatter to do anything but delete trailing white spaces. code formatting is also customizeable, look at those preference pages if you use the IDE often. it will save you grief. you will be amazed at the options.
8. yes. annoying as shit, turn them off. there is even a wizard that takes you to the preference page where you can turn that individual warning off. i recommend getting there manually.

FLAMEWAR!!!!

Useless

[Anonymous Coward]

So, it is BASIC - a language people here look down on and claim "nobody uses" and it is for Linux - which, while popular here, basically nobody uses (very low percent of the installed base of computers). So, maybe three people will use this language? It is only the confluence of BASIC users and Linux users - which has to be very small.

Thumbs up for Gambas

[Anonymous Coward]

I'm a big fan of Gambas.Though I haven't used it in a few years it is a great way to quickly put together graphical applications. It has pretty good help files, simple (BASIC) syntax and is a great way for beginners to learn and for more experienced programmers to throw together simple apps in a hurry. I'm very happy to see this project is continuing to be developed.

Spaghetti code with gotos AND inheritance!

[John Hasler]

Wow! The possiblilties are endless! Do we get overloading as well?

First use on CSI?

[tomhudson]

I'm a big fan of Gambas.Though I haven't used it in a few years it is a great way to quickly put together graphical applications.

Quick - someone tell the writers of CSI so they can now ""whip up a GOOEY in Gambas so they can track the killers' IP address" ... or something equally zombified.

NOT what I wanted

[Anonymous Coward]

I want a BASIC ENVIRONMENT just like
the old days. This is where I can type
"run" and "save" on the same command line
as "10 let A = B + C".

I do not want objects or events.

Why ? Because I can do VERY VERY FAST
quick and dirty little programs and solve problems
quickly - IF the BASIC was put into my linux terminal,
iphone, smartphone, tablet, etc.. It was proven
to work well and occupy under a paltry 32KB
(both ROM and RAM) 30 years ago.

I realize an ENVIRONMENT means you need line numbers.
So what? It makes BASIC editing MUCH faster without
saving and retrieving darn files in a separate editor.

And because teenagers can have an easy
Calculator Language to learn to solve everyday
problems.

Re:NOT what I wanted

[nurb432]

So you use something like freebasic and let others who want those features use Gambas

Re:NOT what I wanted

[Mitchell314]

Please, the only major feature Gambas has that FB doesn't have is inheritance (which is in development (and has been for quite some time now, starting to wonder if it'll ever come)).

/StartObscureLanguageFlameWar :P
Of course, if really want a good flamewar, bring up QB 64.

Re:NOT what I wanted

[innocent_white_lamb]

Bywater Basic
Bas
blassic

Any of those should be what you're looking for (and they all work quite well).

Re:NOT what I wanted

[Anonymous Coward]

Try SpecBAS...

https://sites.google.com/site/pauldunn/https://sites.google.com/site/pauldunn/

Re:NOT what I wanted

[Anonymous Coward]

Sorry, the URL has changed...
It's now www.specbas.co.uk :D

Re:NOT what I wanted

[ZXDunny]

Or indeed, http://www.specbas.co.uk/ Disclaimer: I'm the author of that interpreter. It's based on the old 8bit BASIC style, where you get a "command line" where you enter your code and execute statements. It's line-based, and has sprites and graphics and stuff like mod music and such. It's a little pet project which a small number of people enjoy messing around in. It's not a serious programming language.

Re:NOT what I wanted

[obsess5]

And prior to 30 years ago, there were the Tiny BASICs espoused by Dr. Dobb's Journal of Tiny BASIC Calisthenics & Orthodontia - Running Light without Overbyte, which ran in 2KB.

Windows?

[Osgeld]

Its the main reason I have never bothered with this, it looks like everything I would want except windows support ... which though I run linux on the desktop most of the world does not

Its really not in my interest to sit down and learn a whole new system if I have to toss it out of the window the second I'm on a MS OS

Re:Windows?

[gambix]

Well, a little help would be welcome. Work for a windows version is in progress, but the lack of means in terms of knowledge and time are that it is unfortunately not a priority. It is on this would be a hell of a competitor in the world of windows.

Re:Windows?

[Osgeld]

trust me, I am Ctarded, you dont want my help

Re:Windows?

[gambix]

haha :-)

Cute characters

[evil_aaronm]

Is it required that open source projects have cute characters? Mozilla's fox and other characters, Linux's Tux, BSD's devil, Freemind's butterfly, etc.

I understand "guerrilla marketing" but to whom are we marketing: prepubescent teenie boppers?

Re:Cute characters

[Robert Zenz]

Yeah, because a window and an half-eaten apple is way better.

Re:Cute characters

[Jah-Wren Ryel]

I understand "guerrilla marketing" but to whom are we marketing: prepubescent teenie boppers?

Anime and manga fans.

Re:Cute characters

[marcosdumay]

I understand "guerrilla marketing" but to whom are we marketing: prepubescent teenie boppers?

Everybody likes a cute character. With maybe the exception of male prebubescents and adolescents.

Re:Cute characters

[Hatta]

It's BASIC, yes it's aimed at teenagers.

Re:Cute characters

[Mitchell314]

I think it's as much tradition as it is advertising. Heh, Dragonfly BSD has not one, but two pages dedicated to dragonflies. :D

Re:Cute characters

[gambix]

Yes and i really like the relation to Toys... we want a simple and complete language for doing simply things ... that's all :-)... that's why a so cute shrimp. Happy new year

Re:Cute characters

[Anonymous Coward]

I hate that fucking linux penguin.

Re:Cute characters

[icebraining]

Don't forget Plan9's Glenda, which pretty much out-cutes all of those.

Python too

[nurb432]

Would like to have a true "GUI IDE" Rad tool for python too. ( something like what wavemaker is to java, or ironspeed is to .net tho the latter isnt free ) While it may sound lazy, drag and drop interface creation and good remote debugging speeds development time, even for professionals.

Re:Python too

[BorelHendrake]

You might want to give Eric a try. Eric is an IDE for python which can call Qt designer. This allows you to create the gui using Qt and with PyQt write your actual program in Python.

Re:Python too

[nurb432]

Ya i have and its the best we have so far, but the GUI builder 'extension' is a bit kludgey and still has an 'external' feel ( since it is :) )

I have seen a few other smaller projects but none seem ready for prime time. I just hope one is out there and i missing it.

Actual use

[nurb432]

Not to slight the work he is doing, but has anyone used gambas in any 'real' projects? I have seen lots of toy/pet projects but noting major.

Re:Actual use

[Osgeld]

it wont work on MS systems, otherwise it could be a good VB alternative, otherwise like I stated in a previous post why bother learning it if your never going to be able to use it on other systems

I already know a half dozen useless programming systems, not wasting my time on yet another

Re:Actual use

[Kevin Fishburne]

I'm writing an MMO with it: http://www.youtube.com/user/VasCorpBetMani The users mailing list would probably be the best place to ask what the serious folks are doing with it.

Re:Actual use

[gambix]

some big applications in Gambas : http://gambasdoc.org/help/app/guygle?v3 http://gambasdoc.org/help/app/laurus?v3 http://gambasdoc.org/help/app/domotiga?v3 http://gambasdoc.org/help/app/kedicad?v3 http://gambasdoc.org/help/app/txapucnc?v3 There is many others professionnal applications... but i've not urls for them There is http://gambasforge.org/ and http://gambas.doc.org/ both writted in gambas.

Re:Actual use

[Anonymous Coward]

There's a data modeling tool for Postgres which is written in Gambas:

http://sourceforge.net/projects/pgdesigner/

Re:Actual use

[Anonymous Coward]

Yes, GAMBAS has been used to write large apps of which I am the author of.
One large app I have written in GAMBAS has been running 24 hours a day, 7 days a week since 2006.
Solid as a rock!
-Fernando

Re:Spaghetti code with gotos AND inheritance!

[Mitchell314]

C and C++ have GOTO as well, should we shun them for that too?

nice web site

[Anonymous Coward]

i thought my eyes were going for a minute... then i realized the website intentionally has text that looks blurry...

Re:Spaghetti code with gotos AND inheritance!

[moderators_are_w*nke]

Just because you can it doesn't mean you should

Re:Spaghetti code with gotos AND inheritance!

[tomhudson]

Look at all the case: statements in a switch - they're all GOTOs (the case: is a label). So are your virtual method tables. The break; statement also does a goto to the next instruction after the enclosing set of statements (switch, for, whatever).

So don't knock goto - the software you're using depends on it.

made my year

[A Pressbutton]

someone called fishburne posting about gambas what are the chances of that? could be a variable from the poission distribution

Re:made my year

[Kevin Fishburne]

Haha, never thought about that. Fish. Shrimp. If only CmdrTaco had posted my submission we could have had a real meal on our hands.

English source code?

[Nimey]

It looks like most of the example applications were written by native Spanish speakers, have comments in Spanish, etc. Any good example source in English?

The New 2012 Model T ?

[assertation]

Isn't this like Ford releasing a Model T for 2012?
Who cares about Basic in 2012?

No disrespect to anyone, I'm serious. Honest thought, not meant to offend.

Re:The New 2012 Model T ?

[gambix]

Well ... What i can answer to a so stupid thing ... Ford T is died since many years ... but Ford is alway here ! ... Ford T have Wheel ... and modern Cars too So Gambas Have some little things of basic ... but all the modern tools existing too ! It's in the name ! Gambas Almost Mean BASic ! Yes we are in 2012, and Gambas have been written to make modern apps, not to remake nostalgic one.

Re:The New 2012 Model T ?

[Kevin Fishburne]

I think about 90% of programming is the skill of the programmer and 10% the language used. When you're good you can do pretty much anything with any language (within reason obviously), when you're bad you're out of luck regardless of which language you choose. Here's an example of tool versus fool:

Part 1 (start at 3:00 mark): http://www.youtube.com/watch?v=BQJKQjXpGQA
Part 2: http://www.youtube.com/watch?v=5KiC03_wVjc

GAMBAS is a hell of a lot nicer than a cargo van, I'm just addressing people's constant claims about language superiority/inferiority. The results are the only thing that really matters.

Re:The New 2012 Model T ?

[assertation]

I think there is some truth in what you say, but I've TRIED to use good practices in languages not built for it. The structures of the language have more than a minimal influence.

Languages also come with communities that have cultures that reinforce certain practices. That is where individual programmers pick up and are encouraged toward good/bad practices.

Many procedural languages have some sort of modularization, at the least prodcedures/functions/subroutines.

Any code I have seen in those languages, however, have been giant monolithic multi-page walls of sphagetti code despite the option for breaking things into modules being there.

Having worked with some of those programmers, they do know better, but the culture of those languages along with the structure helped ingrain a career of BAD HABITS.

Re:The New 2012 Model T ?

[gambix]

Just ... Gambas is not procedural :-/.

Re:The New 2012 Model T ?

[Anonymous Coward]

What a silly comment!
GAMBAS does everything any high-level OO lanag does but is easier - way easier!
-Fernando

Re:Spaghetti code with gotos AND inheritance!

[gambix]

Take a look to the Gambas IDE witch is written in Gambas... count the goto uses :-)

What sort of language do we actually need?

[Anonymous Coward]

Looking forward, I don't see much use for a language used to write desktop apps for only Linux, or maybe even Mac, or MS.

I think there will be even more web, server, and mobile development.

What might be great is something to replace javascript. Not that javascript is horrible, but it could be better. Also, something to create apps for multiple mobile platforms, like phonegap, but better.

It would really be great if we could the same, or similar language for desktop, server, web, and mobile; and any platform.

Re:Spaghetti code with gotos AND inheritance!

[Anonymous Coward]

Not to mention that assembly language/machine code is rife with jump statements in multitudinous varieties.

Did Qt Creator get folders?

[master_p]

The last time I tried it, you could not organize your code into folders. Everything had to be saved in one directory, that of the project files.

Re:Did Qt Creator get folders?

[Jorl17]

I've had projects with folders (and symlinks). They've worked perfectly.

Re:Did Qt Creator get folders?

[master_p]

Did you create the folders externally from Qt Creator? because it does not let you create folders.

Re:Did Qt Creator get folders?

[Jorl17]

Yes, AFAIR, I added them to the project after that.

Re:Spaghetti code with gotos AND inheritance!

[jgrahn]

Look at all the case: statements in a switch - they're all GOTOs (the case: is a label). So are your virtual method tables. The break; statement also does a goto to the next instruction after the enclosing set of statements (switch, for, whatever).

Uh, those are *not* gotos. GOTO is a statement
GOTO label
and nothing else[1]. switch, break, continue and run-time polymorphism may or may not have some problems in common with goto but there is no logic which says that if you accept one of them you must not criticise goto.

[1] Modulo variants WRT where the label is allowed to be: in the same function, anywhere in the same source file, and so on.

Re:Spaghetti code with gotos AND inheritance!

[tomhudson]

The switch statement is implemented by many compilers as a goto to those labels in cases where the target elements are not "equally spaced". When the targets (the labeled statements) are all one exactly after the other, (1, 2, 3, 4 ...) it's a simple matter of implement a switch by just doing some pointer arithmetic in a branch table. However, many switch statements are NOT contiguous (for example: 0:, 1: ,2:, 42:, 666:, -1: (with a fall-through to default:), so they have to be implemented with a series of CMP and JE ADDR instructions instead - and guess what the JMP ADDR target is ... it's your goto label:

There is absolutely nothing wrong with GOTO, and it doesn't necessarily lead to spaghetti code, just like not using GOTO doesn't guarantee that your code won't be an ungodly mess. It's all up to the programmer.

In cases of languages such as javascript that don't have a goto, you can get most of the same flexibility by implementing a global STACK array, a global THIS, and a separate global FUNCS array, and pushing and popping objects (esp. your local equivalent of THIS) off the STACK and functions off FUNCS, and making the default operand always be the global THIS array. So, while you may not be able to GOTO, you can achieve the same convenience (and make your code a lot more generic in the process).

Don't talk about what you don't know

[gambix]

I think here many people talk about this language but don't know it ... and many have not seen Basics since many years now .
Gambas is a try to resolve all the bad things done to the basic language ... i don't know if the result is what the most wait for
It's a full Object Language, with classes, inheritance, and all the tools that a trus OO language need to have (sorry for my bad English)
I know than many people will jnot take the time to try it ... But i hope to see beginner to use that kind of tools because it is done in a first time to give the desire to them to begin to create things, with a good object way.
Then it's a full language with all function, it load his class dynamiquely and is certainly one of the more fast on loading big UI... and when i say big UI some of them have undred forms (medical tools) and load in 1-2 sec.

Well, just really try it ... and then talk about it.