Page 2 of 7 FirstFirst 12345 ... LastLast
Results 41 to 80 of 247

Thread: https://www.radbasic.dev/

  1. #41
    PowerPoster wqweto's Avatar
    Join Date
    May 2011
    Location
    Sofia, Bulgaria
    Posts
    5,120

    Re: https://www.radbasic.dev/

    Quote Originally Posted by Elroy View Post
    Personally, I'd be impressed if they provided enough so that we could paste the following code into an IDE, compile it as an x64 executable, and then see it successfully execute. And I don't want to see a hocus-pocus video of them doing it.

    Code:
    
    Option Explicit
    
    Sub Main()
        MsgBox "Hello World!"
    End Sub
    
    

    If they have anything up and running, that would seem to be a relatively trivial task. No forms to deal with, no classes to instantiate, no test of the robustness of any code editor ... just compile and execute the simplest of programs.

    The fact that we've been provided with nothing but glossy web-pages and YouTube videos is highly suspect. This is all vaporware in the extreme.
    How many real compilers for any language have you seen in these forums? Something written in VB6 that can produce working standalone .exe files for instance?

    I've seen a lot of IDE, a lot of YouTube and a lot of vapor in the meantime. . .

    cheers,
    </wqw>

  2. #42
    Hyperactive Member
    Join Date
    Jul 2002
    Posts
    481

    Re: https://www.radbasic.dev/

    So just a question, how many solutions used his method:

    Use ANTLR to create a AST
    Use AST tree to compile using LLVM

    Not sure he used all references below but I think no one ELse has taken this appraoch. Doing it this way drastically reduces effort because he is using tools already out there, therby saving tons of time and effort:

    https://www.antlr.org/
    http://llvm.org/docs/LangRef.html
    https://github.com/antlr/grammars-v4...isualBasic6.g4
    https://github.com/antlr/grammars-v4/tree/master/vb6

    Who Has tried it this way?

    Anxiously waiting the reply!
    Last edited by axisdj; Jun 29th, 2019 at 03:46 PM. Reason: spelling

  3. #43
    Fanatic Member TTn's Avatar
    Join Date
    Jul 2004
    Posts
    685

    Re: https://www.radbasic.dev/

    Quote Originally Posted by axisdj View Post
    So just a question, how many solutions used his method:
    Use ANTLR to create a AST
    Use AST tree to compile using LLVM
    Who Has tried it this way?
    I haven't seen anyone go this route specifically.
    Is ANTLR easy to use? I got lost in the documentation pretty quickly. It must of taken some time to harness that.
    How old is the developer of RADB?

  4. #44
    PowerPoster wqweto's Avatar
    Join Date
    May 2011
    Location
    Sofia, Bulgaria
    Posts
    5,120

    Re: https://www.radbasic.dev/

    Quote Originally Posted by axisdj View Post
    So just a question, how many solutions used his method:

    Use ANTLR to create a AST
    Use AST tree to compile using LLVM

    Not sure he used all references below but I think no one ELse has taken this appraoch. Doing it this way drastically reduces effort because he is using tools already out there, therby saving tons of time and effort:

    https://www.antlr.org/
    http://llvm.org/docs/LangRef.html
    https://github.com/antlr/grammars-v4...isualBasic6.g4
    https://github.com/antlr/grammars-v4/tree/master/vb6

    Who Has tried it this way?

    Anxiously waiting the reply!
    QB64 is self-hosted. Rust is self-hosted. Zig is self-hosted. Self-hosting is the gold standard a language has to reach to be accepted seriously. You have to be able to compile yourself and the compiled version has to produce verbatim compilation (second generation) of yourself.

    The only way to achieve self-hosting in this case is to start w/ ANTLR and later come up with custom written parser and leave LLVM as an external dependancy. This seems like a viable route, provided that rubberduck-vba project poured tremendous efforts to weed out bugs in the VBA grammar that they started w/ from original ANTLR repo and they push fixes upstream as far as I remember. They are using C# too.

    Translating AST you get from ANTLR to LLVM codegen is a lot of effort on it's own. You have to implement the ultimate COM client that VB6 is, and come up with ways to generate COM interfaces, coclasses, IDispatches and typelibs for those ActiveX DLL/OCX projects. This is a lot of research in LLVM IR w/ a lot of follow up questions on the llvm's dev mailing list. Linking and static/dynamic run-times are also a PITA.

    To be a sole maintainer of such a project you have to be either very young and clueless or have a support team of 2-3 friends that can keep your interest when your will dwindles after long months of minimal progress.

    cheers,
    </wqw>

  5. #45
    Junior Member radBasicHost's Avatar
    Join Date
    Jun 2019
    Posts
    27

    Re: https://www.radbasic.dev/

    Quote Originally Posted by wqweto View Post
    QB64 is self-hosted. Rust is self-hosted. Zig is self-hosted. Self-hosting is the gold standard a language has to reach to be accepted seriously. You have to be able to compile yourself and the compiled version has to produce verbatim compilation (second generation) of yourself.

    The only way to achieve self-hosting in this case is to start w/ ANTLR and later come up with custom written parser and leave LLVM as an external dependancy. This seems like a viable route, provided that rubberduck-vba project poured tremendous efforts to weed out bugs in the VBA grammar that they started w/ from original ANTLR repo and they push fixes upstream as far as I remember. They are using C# too.

    Translating AST you get from ANTLR to LLVM codegen is a lot of effort on it's own. You have to implement the ultimate COM client that VB6 is, and come up with ways to generate COM interfaces, coclasses, IDispatches and typelibs for those ActiveX DLL/OCX projects. This is a lot of research in LLVM IR w/ a lot of follow up questions on the llvm's dev mailing list. Linking and static/dynamic run-times are also a PITA.

    To be a sole maintainer of such a project you have to be either very young and clueless or have a support team of 2-3 friends that can keep your interest when your will dwindles after long months of minimal progress.

    cheers,
    </wqw>
    Hi!

    I'm the lead developer of RAD Basic.

    There were a lot of comments. I try to summarize all in this answer (some of them will be in a new FAQ section to the webpage):

    1. No, it is not a fake. You could argue if it is possible, how could it be done, etc. But it is a serious project, no vaporware.

    2. It is a project with a minimal implementation. Sure, if it tries to compile a VB project of 60000 LOC, it will fail miserabily.

    3. Question: So, if it is a minimal implementation, how you could say it is (or will be) 100% compatible? Asnwer: The current implementation is not about MVP (minimum viable product). It can't do mathematical operations! The implementation is about the challenges: COM / ActiveX support, reimplement MSVBVM60.dll and COMCTL32.ocx, etc. So, if this it could be solved, the rest is like any other compiler.

    4. wqweto is right: translating AST to LLVM IR is hard, very hard. So this isn't the current approach, I dismissed after some tries. The current approach is ANTLR/AST representation to C language. And Then compile the C output with LLVM C compiler (CLANG). About this, the devlopment of new features goes by:
    4.1 Write the smallest VB6 program with this feature.
    4.2 Write The smallest C program with same feature. It have to mimic VB6 at 100%.
    4.3 Extract the relevant parts of C program into compiler and/or RB Forms Library.

    5. Question: Why this C approach? Answer: First, because it was the Vb6 approach for native code (not P-CODE). Second, As said, it is easier than LLVM IR.

    6. Language/compiler don't need to be self hosted. Self hosting the language is a good thing because you demostrates to the wolrd that it's useful and powerful. But keep in mind the original, VB6 from Microsoft, it ins't self hosted. Javascript neither is. The main implementations are c/c++ bases, as Java JRE. (library and classes are in java, but not the compiler and core runtime).

    Thanks very much for your feedback. I hope we could continue share ideas.

    Cheers!

  6. #46
    PowerPoster
    Join Date
    Jun 2013
    Posts
    7,219

    Re: https://www.radbasic.dev/

    Quote Originally Posted by radBasicHost View Post
    1. No, it is not a fake ... it is a serious project, no vaporware.
    I don't doubt, that your intentions are serious...
    What I doubt is, that you are fully aware, what you're "getting into"...

    As for "fake or not fake" - there's a very simple way, to prove your technical competence:
    Release at least the executables for your compiler (or IDE/Compiler combination, in case it is integrated like VB6.exe),
    so that others can take a look where you stand...

    If you don't do that - then "more glossy ads, mockups, and videos" will not help
    (neither with your fund-raising, nor with the overall-progress of your project).

    Again, your competence is, what "potential investors out of the VB6-community" want to be sure about.
    From what you write further below, I'm (personally) not sure, that you can go this way succesfully to the end,
    because you seem to underestimate a whole lot of things... and/or approach them in the wrong way...

    Quote Originally Posted by radBasicHost View Post
    2. It is a project with a minimal implementation. Sure, if it tries to compile a VB project of 60000 LOC, it will fail miserabily.
    So, what's the open points, you've identified, which make it "fail miserably" on a midsized VB6-project?
    Can you list "things which work already" - and "things which are still to be implemented" to make it work "100%"?

    Quote Originally Posted by radBasicHost View Post
    3. Question: So, if it is a minimal implementation, how you could say it is (or will be) 100% compatible?
    Asnwer: The current implementation is not about MVP (minimum viable product). It can't do mathematical operations! The implementation is about the challenges: COM / ActiveX support, reimplement MSVBVM60.dll and COMCTL32.ocx, etc. So, if this it could be solved, the rest is like any other compiler.
    Ok, so you're talking about COM basically (as the first hurdle to solve - or judging from your video, apparently "already solved to some extent").
    1. what is your plan (or current implementation), regarding (VB6-compatible) COM-Classes? How do you build them?
    2. what is your plan (or current implementation), regarding (VB6-compatible) UserControls? How do you build them?
    3. what is your plan (or current implementation), regarding (VB6-compatible) COM-Events (for both of the above: *.cls and *.ctl modules)?

    As for your: "the rest is like any other compiler":
    4. do you have experience already, writing such a "simpler one" (if yes, is there a code-repo somewhere)?
    5. what is your plan about VB6-UDTs - and their "auto-serializing/deserializing-feature" to and from Files?
    6. what is your plan about (V)BStrings and VB-SafeArrays (Redim, Redim Preserve)?

    I'd suggest, that you try to answer all of the above questions in a way, that (even if you have not implemented anything from 1. to 6.)
    will ensure us, that you at least "know what you're talking about".

    Quote Originally Posted by radBasicHost View Post
    4. wqweto is right: translating AST to LLVM IR is hard, very hard. So this isn't the current approach, I dismissed after some tries. The current approach is ANTLR/AST representation to C language. And Then compile the C output with LLVM C compiler (CLANG). About this, the devlopment of new features goes by:
    4.1 Write the smallest VB6 program with this feature.
    4.2 Write The smallest C program with same feature. It have to mimic VB6 at 100%.
    4.3 Extract the relevant parts of C program into compiler and/or RB Forms Library.
    I have to say, that the block you wrote above - is the one, which raises the most "alarm-bells" on my end...

    Because it seems like you approach the whole problem in a "non-methodical fashion" (seeing "what sticks", when you throw it at a wall - then repeating - "until everything remains sticking").
    If that is the case, then your project will fail after some time.

    What's needed for 100% compatibility with VB6, is basically my points 1-6 above.
    You *need* to produce 100% compatible COM-Class-instances (COM-Dlls) - including their Events - and you need to produce 100% compatible UserControls (in case you planning to compile to OCXes).
    If that is given (100% COM-compatibility to UserControls and most importantly Classes), you don't need to "fiddle around" with the CommonControls-API yourself -
    just recompile (with your new compiler) the existing VB6-sources which implement this functionality already (as e.g. Krools CommonControl-Replament-Project does).

    Quote Originally Posted by radBasicHost View Post
    6. Language/compiler don't need to be self hosted. Self hosting the language is a good thing because you demostrates to the wolrd that it's useful and powerful. But keep in mind the original, VB6 from Microsoft, it ins't self hosted. Javascript neither is. The main implementations are c/c++ bases, as Java JRE. (library and classes are in java, but not the compiler and core runtime).
    They don't need to be, but compilers (Compiler-IDEs) which can compile themselves have advantages.
    In that regard, I also don't understand, why you were choosing C# as the language you're implementing the Compiler and IDE with.

    That does not make any sense to me, for the following reasons:
    - VB6 is still more "raddish" than C# (you'd save time with it, writing all the new stuff)
    - When VB6 would have been choosen - and your new compiler ends up being "100% compatible" - the selfhosting of new IDE and compiler would have been accomplished automatically

    You asked for comments, and I hope I gave you enough of them.
    Please keep in mind, that these are all technical ones - there's absolutely nothing personal about them all...

    HTH

    Olaf

  7. #47
    PowerPoster wqweto's Avatar
    Join Date
    May 2011
    Location
    Sofia, Bulgaria
    Posts
    5,120

    Re: https://www.radbasic.dev/

    Quote Originally Posted by radBasicHost View Post
    . . . translating AST to LLVM IR is hard, very hard. So this isn't the current approach, I dismissed after some tries. The current approach is ANTLR/AST representation to C language. And Then compile the C output with LLVM C compiler (CLANG).
    I'm not sure translating to C will be easier, Olaf is proponent of this approach while I prefer to keep codegen within LLVM API w/o strings floating around so much (if possible).

    Wish you best of luck and do keep us posted!

    cheers,
    </wqw>

  8. #48
    Addicted Member
    Join Date
    Aug 2013
    Posts
    236

    Re: https://www.radbasic.dev/

    Good luck RB, looking forward to hearing of positive developments!

  9. #49
    Hyperactive Member
    Join Date
    Jul 2002
    Posts
    481

    Re: https://www.radbasic.dev/

    Hello Everyone,

    It appears there is enough talent on this forum to get this done. I think each person giving suggestions adds to the big picture.

    Here is the issue, each of the GURUs here stand firm on their approach for obvious reasons, but can you imagine if all of you worked together, each making some compromise, I think this is the best case scenario which would lead to a vb6 replacement.

    Each GURU has in depth knowledge of how to solve this issue, I wonder if a virtual round table meeting via audio or video conferencing could achieve a good result where all parties involved could work together.

    I see it where small tests could be agreed on for each step of the way to determine the best path forward for each segment of this project.

    If all/anyone is up for it, lets set up a virtual meeting, just to see what happens.

    My 2 cents.

  10. #50
    Junior Member radBasicHost's Avatar
    Join Date
    Jun 2019
    Posts
    27

    Re: https://www.radbasic.dev/

    Hi Olaf,

    A lot of ideas here. Let me share my point of view:

    I don't doubt, that your intentions are serious...
    What I doubt is, that you are fully aware, what you're "getting into"...

    As for "fake or not fake" - there's a very simple way, to prove your technical competence:
    Release at least the executables for your compiler (or IDE/Compiler combination, in case it is integrated like VB6.exe),
    so that others can take a look where you stand...

    If you don't do that - then "more glossy ads, mockups, and videos" will not help
    (neither with your fund-raising, nor with the overall-progress of your project).
    Well, it's your point of view. A lot of people pre-order videogames which only share videos, and some only cinematics, not playable levels.
    But, sure release executable will be fine. But it's dangerous. It has to be tested very well. What about it didn't run because you don't have .Net version required? What about it fails compile because a bug? In fact, yesterday, while I was preparing a new video it failed to compile a VB project with a form containing two textbox, two labels and a button. That was because a bug that I fixed quickly. But people could think: "Uuh! It fails to compile such a simple project. It's vaporware, I've already said it!"

    Developers have to be careful releasing "Work In Progress" versions. It could have a negative impact. I have to calibrate all scenarios, as you know. Maybe, we could go with a demo in real time in youtube, for skeptics (just ideas).

    Again, your competence is, what "potential investors out of the VB6-community" want to be sure about.
    From what you write further below, I'm (personally) not sure, that you can go this way succesfully to the end,
    because you seem to underestimate a whole lot of things... and/or approach them in the wrong way...


    So, what's the open points, you've identified, which make it "fail miserably" on a midsized VB6-project?
    Can you list "things which work already" - and "things which are still to be implemented" to make it work "100%"?
    Sure, it's a prototype, not a final product. Not a MVP product. The VB grammar is simple, a lot of statements are not implemented, not strings function and so on. I'm implemented the challenging parts, so I can trust I can go for it and it will be successfully.

    Ok, so you're talking about COM basically (as the first hurdle to solve - or judging from your video, apparently "already solved to some extent").
    1. what is your plan (or current implementation), regarding (VB6-compatible) COM-Classes? How do you build them?
    2. what is your plan (or current implementation), regarding (VB6-compatible) UserControls? How do you build them?
    3. what is your plan (or current implementation), regarding (VB6-compatible) COM-Events (for both of the above: *.cls and *.ctl modules)?
    About this three: COM-Classes are written in C. In fact, a lot of OCX used in VB6 were written in C. I'm translate the code to C. I'm working in C. So, no problem about it. UserControls, CLS, CTL, just write C code. The idea is: all that you do in VB6, could be done in C, so write automatically in C and compile it with CLANG (LLVM).

    As for your: "the rest is like any other compiler":
    4. do you have experience already, writing such a "simpler one" (if yes, is there a code-repo somewhere)?
    Yes, I have. But I have no repo. Sorry. I don't like share personal information. But hey! For the trust of the community: At university my final project degree was about a compiler (translator) of pseudo-code to java. It's in catalan (similar to spanish). You could check out:http://dugi.udg.edu/item/http:@@@@hd...@@2072@@201003

    A compiler could be split to:

    1. frontend (compiler language to IR/IL)

    2. optimizer (working in IR/IL domain)

    3. backend (write machine code from IR/IL)

    This project is about constructing a frontend. IR/IL will be the C code. The optimizer and backend goes to LLVM. And the generated frontend will use support libraries for Win32 controls, strings management, etc.

    5. what is your plan about VB6-UDTs - and their "auto-serializing/deserializing-feature" to and from Files?
    6. what is your plan about (V)BStrings and VB-SafeArrays (Redim, Redim Preserve)?
    Sincerely, I didn't think about it yet. If the funding goes well and the project go further, this will be thought and sorted out. For example, redim and redim preserve could be implemented with some functions in plain C with dynamic memory, copy of arrays, etc.

    I'd suggest, that you try to answer all of the above questions in a way, that (even if you have not implemented anything from 1. to 6.)
    will ensure us, that you at least "know what you're talking about".
    Sure, I answered honestly all of this.

    I have to say, that the block you wrote above - is the one, which raises the most "alarm-bells" on my end...

    Because it seems like you approach the whole problem in a "non-methodical fashion" (seeing "what sticks", when you throw it at a wall - then repeating - "until everything remains sticking").
    If that is the case, then your project will fail after some time.

    What's needed for 100% compatibility with VB6, is basically my points 1-6 above.
    You *need* to produce 100% compatible COM-Class-instances (COM-Dlls) - including their Events - and you need to produce 100% compatible UserControls (in case you planning to compile to OCXes).
    If that is given (100% COM-compatibility to UserControls and most importantly Classes), you don't need to "fiddle around" with the CommonControls-API yourself -
    just recompile (with your new compiler) the existing VB6-sources which implement this functionality already (as e.g. Krools CommonControl-Replament-Project does).
    Hey, I think I explained bad. Sure, there is a methodical fashion. There is a compiler and a grammar. But the idea is to mimic 100% VB6. So, I just test separately the part, and then integrate it into the compiler in a very methodical way. It's not just throw C code over and over. Oh my GOD!

    They don't need to be, but compilers (Compiler-IDEs) which can compile themselves have advantages.
    In that regard, I also don't understand, why you were choosing C# as the language you're implementing the Compiler and IDE with.

    That does not make any sense to me, for the following reasons:
    - VB6 is still more "raddish" than C# (you'd save time with it, writing all the new stuff)
    - When VB6 would have been choosen - and your new compiler ends up being "100% compatible" - the selfhosting of new IDE and compiler would have been accomplished automatically
    VB6 only could save me time in the IDE. But the IDE is the small part. The really important part is the compiler. And I chose ANTLR for lexical and syntactic parser. It could output to some languages, but not VB6. So, C# for the compiler (it could be java too), and the IDE goes to C# to remains to same language. But in this project the IDE is the small part. Maybe the VB6 will be more powerful than RAD Basic 1.0. In fact, you could develop in VB6 IDE and compile with RB Compiler. The effort is in compiler and libraries, not the IDE.

    If I went to VB6, I have to write the tokenizer part and the state graph for the grammar (syntactic parser) by hand. Sure, it could be done, but I disagree with you. VB6 for the compiler part was the unproductive way. It's a shame, but it's the reality. And Microsoft wrote the VB6 compiler in C, not in VB6.

    You asked for comments, and I hope I gave you enough of them.
    Please keep in mind, that these are all technical ones - there's absolutely nothing personal about them all...
    Sure, there are technical ones, but some personal too, no (about my knowledge)? No problem about it.

    Cheers!

    Carles Royan

    PD: I'm not native English. Please excuse linguistic errors in a such large text

  11. #51
    PowerPoster
    Join Date
    Jun 2013
    Posts
    7,219

    Re: https://www.radbasic.dev/

    Quote Originally Posted by radBasicHost View Post
    Well, it's your point of view. A lot of people pre-order videogames which only share videos, and some only cinematics, not playable levels.
    I guess in those cases (of vid-games) - the authors were already known in the gaming-community
    (less of a need to "prove your competence", when you can point out "prior art, attached to your name").

    In case of a new programming-language + large needed runtime-libs + a lot of GUI-related-stuff,
    people will (rightly) expect that you prove your competence in (all) these areas beforehand (IMO).

    And since we talk about VB6 here - one more requirement is necessary: -> "large experience with COM and existing COM-interfaces",
    it's especially that latter part, I have my doubts about (judging from, how you answered my earlier questions 1..to..3).

    Quote Originally Posted by radBasicHost View Post
    But, sure release executable will be fine. But it's dangerous. It has to be tested very well.
    What about it didn't run because you don't have .Net version required? What about it fails compile because a bug?
    Just point out "the right .NET-version" - and I'm sure many of the more experienced devs here will have no problem running your demo.
    The goal of the whole exercise not being "bug-hunting" - but instead: "Will it do, exactly as shown in your videos?" -
    because that would already "rule out mockups, which faked the compilation-process".

    Quote Originally Posted by radBasicHost View Post
    In fact, yesterday, while I was preparing a new video it failed to compile a VB project with a form containing two textbox, two labels and a button.
    That was because a bug that I fixed quickly...
    Now, that you've fixed the bug - and your compiler is producing C-Code - can you zip-up and post this generated C-Code here and now?
    Should be easy enough for you (we'd also not have to bother with executables or .NET-runtimes)...
    Just let the devs here take a look at your generated C-Code - can you do that at least?

    Quote Originally Posted by radBasicHost View Post
    About this three: COM-Classes are written in C. In fact, a lot of OCX used in VB6 were written in C.
    I'm translate the code to C. I'm working in C. So, no problem about it.
    Great, then there should be no problem, posting the C-Code you've produced for your above:
    "two textbox, two labels and a button" example...

    Quote Originally Posted by radBasicHost View Post
    ... I have no repo. Sorry. I don't like share personal information.
    As said, just zipping up the produced C-Code from your last example above, would be enough (at least for me).


    Quote Originally Posted by radBasicHost View Post
    Ok, so that proves - that you at least know how to use ANTLR in conjunction with a high-level objectoriented language (Java), translating your own "toy-language".
    Well, it's harder when you're trying to translate from a "NonToy-HighLevel-ObjectOriented language" (VB6) to a procedurale one (C).

    What you wrote in the remaining paragraphs does not convince me at all, that you're fully aware, what you're getting yourself into.
    You posting the zipped up C-Code (for the little Demo you've mentioned above) - would go a long way, to clear up doubts and concerns about that...

    Olaf

  12. #52
    PowerPoster
    Join Date
    Jun 2013
    Posts
    7,219

    Re: https://www.radbasic.dev/

    Quote Originally Posted by axisdj View Post
    It appears there is enough talent on this forum to get this done.
    As I see it, there's only 3 people in this forum, who have the skills to approach the problem.
    And only two of them have (more or less) declared, that they are actively working on a new compiler/IDE.
    That'd be Vlad (wqweto) and me...
    (..."the trick" is the 3rd one I'd consider capable enough, but he never mentioned explicitely what his plans are - regarding compiler-development).

    Quote Originally Posted by axisdj View Post
    I wonder if a virtual round table meeting via audio or video conferencing could achieve a good result where all parties involved could work together.
    Not sure, how a "virtual round table" could help... IMO this forum here should be entirely sufficient, to clear things up fast...

    My position is this (only mentioning the most important "base-points" - I'm sure there's many more detail-points at lower levels, where more disagreement would ensue):
    1. C-emitting has several advantages over LLVM-usage (which I've nevertheless experimented with, but ruled out about 5 years ago)
    2. already the first release has to work in a platform-independent manner
    3. and that involves incorporating a platform-indepently designed, stable and well-tested "new Class-Runtime" already in the development-phase of new IDE and compiler
    4. point 2. also requires a COM-implementation which will work not only on Windows, but on other platforms as well
    ... (basically "a built-in regfree mode" has to be ensured from the get go, as well as a platform-independent way for "reflection", to parse out Dll-contained "COM-types").
    5. Development of new IDE and compiler should be in VB6

    In my opinion, Point 5 of the above list - is the only one where agreement exists between us (I've included it only for completeness).

    All other points (1..to..4) above are "fixed ones", which I will not "back off from" (because they are fundamental).

    Knowing how "the minds of developers work" (especially those of the skilled ones), I'm quite sure that Vlad will also not back-off from "his own plans" -
    IMO he is long aware of my above points - and never indicated anything with regards to "switching over" to "my way of approaching the problem".

    I guess, it was "unspoken, but clear" between the two of us implicitely (never having exchanged any Emails), just by reading each other forum-posts -
    so a "round table" was never really needed.

    HTH

    Olaf

  13. #53
    Fanatic Member TTn's Avatar
    Join Date
    Jul 2004
    Posts
    685

    Re: https://www.radbasic.dev/

    Quote Originally Posted by Schmidt View Post
    As I see it, there's only 3 people in this forum, who have the skills to approach the problem.
    And only two of them have (more or less) declared, that they are actively working on a new compiler/IDE.
    That'd be Vlad (wqweto) and me...
    (..."the trick" is the 3rd one I'd consider capable enough, but he never mentioned explicitely what his plans are - regarding compiler-development).
    Olaf
    I'd be careful of listening to Olaf and his estimates on how things should work.

    When the new compiler (and new IDE-env.) is in a shape, that it can produce Binaries from at least
    VB6 *.cls and *.bas Modules, then the vbRichClient-Dll itself will be recompilable, and also the new
    IDE-Environment with the new Compiler will be recompilable with this first (VB6-generated) Compiler.
    The timeframe until "we" (I hope the community will not leave me alone with that) will achieve this
    final self-hosting state, will be (realistically seen) another 1.5-2 years or so.

    Olaf Schmidt, April 2013
    In my opinion, Olaf is not qualified to make an accurate time line. As far as we know, Olaf's project plan has been abandoned due to a failure to understand what it takes to build a new IDE/compiler.

  14. #54
    PowerPoster
    Join Date
    Jun 2013
    Posts
    7,219

    Re: https://www.radbasic.dev/

    Quote Originally Posted by TTn View Post
    I'd be careful of listening to Olaf and his estimates on how things should work.
    Simple question to you - why do you think "a not met timeline" has anything to do with "the how" (of approaching things)?

    Quote Originally Posted by TTn View Post
    In my opinion, Olaf is not qualified to make an accurate time line.
    That might be true - but then - this is true for every developer who plans for "ambitous, complex implementations"
    (not holding timelines is the norm - not the exception - especially with huge projects like this).

    Quote Originally Posted by TTn View Post
    As far as we know, Olaf's project plan has been abandoned due to a failure to understand what it takes to build a new IDE/compiler.
    I'd think this is a "low blow" - and that even wqweto (despite our different views on things) would disagree with you on this part.

    Besides, if you look at what you've quoted from me - especially this part:
    "I hope the community will not leave me alone with that"...

    Then you already have your explanation for the "timeline-issue"
    (because that's exactly what happened and what continues to happen,
    most members of the community not understanding, why my points 1-5 are necessary).

    The other reason is (I've explained that already a few times), that I'm not self-employed anymore,
    but work "a normal job" since 2014 (which leaves *significantly* less time for the project).

    Olaf

  15. #55
    Fanatic Member TTn's Avatar
    Join Date
    Jul 2004
    Posts
    685

    Re: https://www.radbasic.dev/

    Quote Originally Posted by Schmidt View Post
    Simple question to you - why do you think "a not met timeline" has anything to do with "the how" (of approaching things)?


    That might be true - but then - this is true for every developer who plans for "ambitous, complex implementations"
    (not holding timelines is the norm - not the exception - especially with huge projects like this).


    I'd think this is a "low blow" - and that even wqweto (despite our different views on things) would disagree with you on this part.

    Besides, if you look at what you've quoted from me - especially this part:
    "I hope the community will not leave me alone with that"...

    Then you already have your explanation for the "timeline-issue"
    (because that's exactly what happened and what continues to happen,
    most members of the community not understanding, why my points 1-5 are necessary).

    The other reason is (I've explained that already a few times), that I'm not self-employed anymore,
    but work "a normal job" since 2014 (which leaves *significantly* less time for the project).

    Olaf
    An "unrealistic" time line is not the way to approach the problem. You failed to deliver, and IMO will never complete your project. You didn't just miss the time line, you've tripled it at this point, including time before the job. You can't blame anyone else for this, so stop trying. It's good that you admit underestimating the time, and hopefully you can teach others from your mistakes. If you are going to dish it out, you must be able to take the same critique.

  16. #56
    PowerPoster
    Join Date
    Jun 2013
    Posts
    7,219

    Re: https://www.radbasic.dev/

    Quote Originally Posted by TTn View Post
    An "unrealistic" time line is not the way to approach the problem.
    Then you're even less experienced than I thought you'd be...
    (because: if you're not taking an "optimistic stance" with the timeline, then you'll never start larger projects in the first place).

    See, the *realistic* expected Timeline would be the time the original team
    (of roughly 10 skilled people at Microsoft) was needing until VB6 was born:
    - about 10 people times 3-4 years working full-time on it (including finishing the versions which came before - mainly VB4 and VB5)
    - so that makes the realistic time-span 30-40 man-years

    I knew that above "realistic timespan" beforehand (when I started with this stuff, roughly 10 years ago) -
    but I was counting on "RAD-effects" from developing it in VB6 - and was (had to be) optimistic.

    Looking back - so far, I've invested about 3-4 man-years into the project within the last 10 years
    (working only "On/Off" - and not entirely "full-time" on it).
    And yes, I'm aware that it will take at least "another man-year or three" until things are finished,
    which will make it "4-7 man-years total" when it's done...

    So, in the end - using VB6 to develop it, will have "payed off" - because the needed man-years were less than the original team had to invest.

    As for "dishing out" - I was (so far) only pointing out "technical problems only"
    (asking the "new kid on the block", to "show stuff" which convinces me and others, that he's aware of the problems).

    Your line of "argumentation" is not based on any "technical points" at all so far (regarding compiler and IDE-development)
    It's purely personal, without any back-up behind your insults.

    Olaf

  17. #57
    PowerPoster wqweto's Avatar
    Join Date
    May 2011
    Location
    Sofia, Bulgaria
    Posts
    5,120

    Re: https://www.radbasic.dev/

    Let me quote a post of mine from 1 year ago on the same topic:

    Quote Originally Posted by me in 2018
    My point is that vNext is going to be a one-man show, two max. Just lowering expections of anyone reading such threads that community here is miraculously going to form a team of experts/hobbyists that are going to first plan, then implement. . . anything.

    vNext might come up next month w/ a post on vbforums in Linus style: "Hey guys, here is something I've been workng lately. Enjoy!" or equaly possible this is not goint to happen in the next 10 years.

    Here is my predictions how it's possible to happen: someone will compile LLVM w/ stdcall conventions, make VB6 wrappers and port kaleidoscope toy-language compiler in VB6. Then he or someone else will write a basic VB6-toy-language compiler. Then he or someone else will tweak it to use LLVM's interpreter for edit&continue. Then he or someone else will implement language services on top of symbol-tables and hook VSCode IDE. Then he or someone else will undertake VB6 Forms subsystem as a completely separate (dependent) project. Then Olaf will try to port RC5 (for Windows only) as a separate project. Then he or someone else will implement Win/COM compatibility layer for Linux as a separate project, etc.
    The actual progress so far is probably closer to the 10 years estimate (LOL) but nevertheless I'm working on the VBScript cloning as a PoC for "VB6-toy-language compiler" and COM is hard. It's precisely because I couldn't use ANTLR for parser generator (no VB6 target) that I had to implement a PEG parser generator for VB6 which I'm currently trying to learn how to use :-))

    Olaf is very adamant about incorporating RC5 in vNext and I find his aim orthogonal to the actual compiler project. IDE is also optional, provided that VSCode can be integrated if there are good/fast compiler services that expose symbols and code structure (probably w/ reparsing only changes as an optimization).

    For any community effort the community has to walk the walk and come up with canonical VB6 for doing common things. There is no generally accepted coding style, no generally used code linters or formatters and I find a lot of code snippets here in these forums that are next to illegible. Probably this is somewhat like egg vs chicken problem and a big project like a self-hosting compiler will set up the canonical style for future efforts (but who knows).

    It's probably worth watching closely projects like rust and zig as these are LLVM based languages, esp. the second one is a fringe language, a one-man show similar to my vNext efforts that I'm "borrowing" ideas (and code) from.

    cheers,
    </wqw>

  18. #58
    Fanatic Member TTn's Avatar
    Join Date
    Jul 2004
    Posts
    685

    Re: https://www.radbasic.dev/

    Quote Originally Posted by Schmidt View Post
    Then you're even less experienced than I thought you'd be...
    (because: if you're not taking an "optimistic stance" with the timeline, then you'll never start larger projects in the first place).

    See, the *realistic* expected Timeline would be the time the original team
    (of roughly 10 skilled people at Microsoft) was needing until VB6 was born:
    - about 10 people times 3-4 years working full-time on it (including finishing the versions which came before - mainly VB4 and VB5)
    - so that makes the realistic time-span 30-40 man-years

    I knew that above "realistic timespan" beforehand (when I started with this stuff, roughly 10 years ago) -
    but I was counting on "RAD-effects" from developing it in VB6 - and was (had to be) optimistic.

    Looking back - so far, I've invested about 3-4 man-years into the project within the last 10 years
    (working only "On/Off" - and not entirely "full-time" on it).
    And yes, I'm aware that it will take at least "another man-year or three" until things are finished,
    which will make it "4-7 man-years total" when it's done...

    So, in the end - using VB6 to develop it, will have "payed off" - because the needed man-years were less than the original team had to invest.

    As for "dishing out" - I was (so far) only pointing out "technical problems only"
    (asking the "new kid on the block", to "show stuff" which convinces me and others, that he's aware of the problems).

    Your line of "argumentation" is not based on any "technical points" at all so far (regarding compiler and IDE-development)
    It's purely personal, without any back-up behind your insults.

    Olaf
    This is hilarious. So now, in order to suite your argument, you are condoning the OP's over optimism...well no,... but yes it's ok for me to do that. Backpedal. Let's remember that Olaf said realistic, not optimistic. Its funny how he actually gets real personal and often is passively aggressive/condescending to other members in this forum, but then objects when forced to confront the hypocrisy. To be fair, I don't think you notice that you do this, but it still doesn't make it ok. So why is your delay? Because RAD will pay off.... Not making any sense, forget it. I've made my point to the rational members of this forum.

  19. #59
    PowerPoster
    Join Date
    Jun 2013
    Posts
    7,219

    Re: https://www.radbasic.dev/

    Quote Originally Posted by TTn View Post
    So now, in order to suite your argument, you are condoning the OP's over optimism... well no,... but yes it's ok for me to do that.
    Optimism is necessary in either case - period.
    But deep knowledge about the topic (and especially COM) is a necessity as well...
    I cannot see, why it is so difficult for some people to comprehend that.

    And in case of radBasicHost, I was trying to find out how deep that knowledge of him goes... (politely, AFAIK).
    I stand by what I wrote earlier, that (so far, and in my personal opinion) only 3 guys here in the forum have the needed skills,
    maybe radBasicHost will be the 4th - but unless he's posting something else than videos, that question remains open (for me).

    You are apparently not interested in his answers to the questions I brought up - fine by me -
    but I'm not going to invest into a project, which has only videos to offer - and nothing else.

    Olaf

  20. #60
    PowerPoster wqweto's Avatar
    Join Date
    May 2011
    Location
    Sofia, Bulgaria
    Posts
    5,120

    Re: https://www.radbasic.dev/

    @radBasicHost: Did I mention that zig creator is on patreon: https://www.patreon.com/andrewrk/overview

    I far as I can see he is collecting in donations $1,970 per month as of now which is the *maximum* you can expect from your crowd-funding efforts as Andrew is very active both code-wise on github and speaking at conferences about the ideas behind his language. Also the zig as a language has some strong followers/contributors too, although probably VB6 (legacy) code-base is orders of magnitudes larger in total.

    Now it's up to you to convince me to direct my $10 support from zig to your VB6-clone effort :-))

    cheers,
    </wqw>

  21. #61
    Junior Member radBasicHost's Avatar
    Join Date
    Jun 2019
    Posts
    27

    Re: https://www.radbasic.dev/

    Quote Originally Posted by Schmidt View Post
    As I see it, there's only 3 people in this forum, who have the skills to approach the problem.
    And only two of them have (more or less) declared, that they are actively working on a new compiler/IDE.
    That'd be Vlad (wqweto) and me...
    (..."the trick" is the 3rd one I'd consider capable enough, but he never mentioned explicitely what his plans are - regarding compiler-development).



    Not sure, how a "virtual round table" could help... IMO this forum here should be entirely sufficient, to clear things up fast...

    My position is this (only mentioning the most important "base-points" - I'm sure there's many more detail-points at lower levels, where more disagreement would ensue):
    1. C-emitting has several advantages over LLVM-usage (which I've nevertheless experimented with, but ruled out about 5 years ago)
    2. already the first release has to work in a platform-independent manner
    3. and that involves incorporating a platform-indepently designed, stable and well-tested "new Class-Runtime" already in the development-phase of new IDE and compiler
    4. point 2. also requires a COM-implementation which will work not only on Windows, but on other platforms as well
    ... (basically "a built-in regfree mode" has to be ensured from the get go, as well as a platform-independent way for "reflection", to parse out Dll-contained "COM-types").
    5. Development of new IDE and compiler should be in VB6

    In my opinion, Point 5 of the above list - is the only one where agreement exists between us (I've included it only for completeness).

    All other points (1..to..4) above are "fixed ones", which I will not "back off from" (because they are fundamental).

    Knowing how "the minds of developers work" (especially those of the skilled ones), I'm quite sure that Vlad will also not back-off from "his own plans" -
    IMO he is long aware of my above points - and never indicated anything with regards to "switching over" to "my way of approaching the problem".

    I guess, it was "unspoken, but clear" between the two of us implicitely (never having exchanged any Emails), just by reading each other forum-posts -
    so a "round table" was never really needed.

    HTH

    Olaf
    Hi Olaf,

    I'm sorry, but I completely disagree with you.

    I'm not enter in personal context, as you does, qualifying people (not only me) and their knowledge. How do yo know there are 3 people who could do a project like this? Do you know everyone of this forum?

    I see there are in this forum a lot of ego, because there are responses and criticism side of the main purpose of this thread.

    About your answer:

    I saw your work about a new implementation of common controls wrapping GTK+. You are a skilled person in software development. But about what you said, I really doubt about your knowledge in compilers area.

    About your points:

    1. Yes, I think C-emitting is the way to go.
    2. No. It could be. VB6 was no platform independent. RAD Basic want to be VB7, so it can't be platform independent. It sounds weird, but I'm a linux heavy user. I really like a VB solution cross platform. But it is not the way to go. I'm developed a lot of cross platform solution. True cross platform app is not only about language/compiled object, is a lot of your code. You can develop with a platform independent language and be tiered to a platform because you use platform dependent calls, winapi, path like C:\WINDOWS, etc.

    I think the first step is doing a true replacement VB6 tiered to Windows. Next version could be platform independent. But you have to begin your VB project in a platform independent way, or migrate it. Not imports of winapi for example. For this, you can rely in a library for GUI (RB Forms), with your code decoupled and it could be changed easy to another library cross platform based in, for example, GTK+.

    3. As (2). Sure, we need a new class replacement (MSVBVM, COMCTL, etc.). But no need at this moment cross platform. It's good to design with this goal in mind, but I think it is no need at this moment.

    4. I think it is the wrong approach. If you want a true VB replacement, it is virtually impossible. How do you run a COM OCX/DLL of a third party in Linux? Will you wrap with WINE? Will you add an extra layer translating Win32 call to system C calls? This way will fail for sure.

    5. It's my opinion, but i think this is a really wrong approach. If you restrict your tools you will fail a lot of power and productive. You say you will use tools of 20 years ago for escaping the hole of no new tools since 20 years ago.

    I see you have strong thinkings about it. But I disagree with your approach.

    But, it's good confront different approach to the problem, and no problem if you don't support RAD Basic project.

    Cheers!

  22. #62
    Junior Member radBasicHost's Avatar
    Join Date
    Jun 2019
    Posts
    27

    Re: https://www.radbasic.dev/

    Quote Originally Posted by wqweto View Post
    @radBasicHost: Did I mention that zig creator is on patreon: https://www.patreon.com/andrewrk/overview

    I far as I can see he is collecting in donations $1,970 per month as of now which is the *maximum* you can expect from your crowd-funding efforts as Andrew is very active both code-wise on github and speaking at conferences about the ideas behind his language. Also the zig as a language has some strong followers/contributors too, although probably VB6 (legacy) code-base is orders of magnitudes larger in total.

    Now it's up to you to convince me to direct my $10 support from zig to your VB6-clone effort :-))

    cheers,
    </wqw>
    Hi!

    I'll try it!

    Very interesting zig creator. As I've said, this is a different approach. It is not a new environment, it would be a direct clone.

    I think this kind of project, if you want to be successful, it has to be done in a professional way. And it is not related about knowledge of developers. Professional way is about not a side project. It's your job. You are working on it the whole working hours. It's the only way to complete it. If not, it would be a project that in some moment it would remains abandoned and unfinished.

    That the reason of future Kickstarter campaign.

    Cheers!

  23. #63
    PowerPoster
    Join Date
    Jun 2013
    Posts
    7,219

    Re: https://www.radbasic.dev/

    Quote Originally Posted by radBasicHost View Post
    I'm sorry, but I completely disagree with you.
    No problem...

    Quote Originally Posted by radBasicHost View Post
    I'm not enter in personal context, as you does, qualifying people (not only me) and their knowledge.
    How do yo know there are 3 people who could do a project like this? Do you know everyone of this forum?
    Yes, I read every posting here in the VB6-Forum, so I know (roughly) the skill-level of everyone who posts here.
    If you doubt that something like that is possible, then please consider, that I'm reading about 50
    developer-postings every day on average (and that for the last 25 years, across several forums and newsgroups).
    So, please consider that "neuronal network" as "trained enough"...

    Quote Originally Posted by radBasicHost View Post
    I really doubt about your knowledge in compilers area.
    I have no problem with that either...

    But let's focus again on the thing you want to distract from (for the sake of the other readers here):
    - So far we've only seen videos from you...
    - neither a compiled executable was offered for download...
    - nor did you post any of the C-files, you allegedly generate with your approach

    I cannot come up with any good reason, why you'd avoid posting these things,
    when your efforts are indeed serious ones (and not fake).

    Olaf

  24. #64
    Hyperactive Member
    Join Date
    Jul 2002
    Posts
    481

    Re: https://www.radbasic.dev/

    Post Deleted because was not directly related to RadBasic
    Last edited by axisdj; Jul 2nd, 2019 at 09:27 AM.

  25. #65
    The Idiot
    Join Date
    Dec 2014
    Posts
    2,721

    Re: https://www.radbasic.dev/

    please keep the topic within radbasic, since the thread is about that and not about people, if they can or can not, if theres just 3 or less that can make a compiler or not.

    what the thread should be about is a showcase of radbasic, what it can do right now, the plans, the obstacles, q&a.
    to pick a fight to the dev of radbasic is just ego talking, maybe jealousy? i dont give a crap about all that and ultimately its useless flaming since we are programmers and we want to know if radbasic is possible or not.

    can the community help? sure fine. but only if its possible and only if you are mature enough to be able to do that, to argue and ego-compete with each other will only stress it and ruin the good mood of making something nice.

    radBasicHost, when you ready, I think it would be nice with some examples, this to show your progress and calm some members here that seems to take this personal.

  26. #66
    Super Moderator Shaggy Hiker's Avatar
    Join Date
    Aug 2002
    Location
    Idaho
    Posts
    38,988

    Re: https://www.radbasic.dev/

    Quote Originally Posted by Schmidt View Post
    I guess in those cases (of vid-games) - the authors were already known in the gaming-community
    (less of a need to "prove your competence", when you can point out "prior art, attached to your name").

    Olaf
    Actually, I think that in the case of games nobody really cares, and that's a key distinction. With a video game, one might reasonably take a $50 flyer on something that looks interesting. If it works out, that's fun and games for some number of hours or days. If it doesn't....then move on to the next thing.

    You don't do that with a programming environment so readily. Sure, you can abandon something just as easily as you would abandon a game that simply didn't work, but if you do not quickly abandon a development environment, then your investment in the environment is sure to be FAR higher, which means that the cost of moving away is going to be far higher. Not an economic cost, of course, but a time and intellectual cost, which is all the more dear.

    VB6 isn't a game that doesn't work on your current hardware. The reason for this thread is that the investment people have in VB6 is vastly greater than people have in some game that no longer runs on their current hardware.

    So, what people will do with video games doesn't compare to what people will do with development environments. If you end up liking a video game, your investment is shallow, so like it or not, the cost isn't high. Not so with development environments.
    My usual boring signature: Nothing

  27. #67
    Super Moderator Shaggy Hiker's Avatar
    Join Date
    Aug 2002
    Location
    Idaho
    Posts
    38,988

    Re: https://www.radbasic.dev/

    Quote Originally Posted by baka View Post
    please keep the topic within radbasic, since the thread is about that and not about people, if they can or can not, if theres just 3 or less that can make a compiler or not.

    what the thread should be about is a showcase of radbasic, what it can do right now, the plans, the obstacles, q&a.
    to pick a fight to the dev of radbasic is just ego talking, maybe jealousy? i dont give a crap about all that and ultimately its useless flaming since we are programmers and we want to know if radbasic is possible or not.

    can the community help? sure fine. but only if its possible and only if you are mature enough to be able to do that, to argue and ego-compete with each other will only stress it and ruin the good mood of making something nice.

    radBasicHost, when you ready, I think it would be nice with some examples, this to show your progress and calm some members here that seems to take this personal.
    I second this.

    The subject of VB6 and whether there is a new version, and what it will be, or when, gets pretty contentious and has a tendency to go off the rails. Strong opinions abound. Best to stick with the topic at hand.
    My usual boring signature: Nothing

  28. #68
    Hyperactive Member
    Join Date
    Mar 2018
    Posts
    460

    Re: https://www.radbasic.dev/

    How much is radbasic going to cost?

  29. #69
    Lively Member
    Join Date
    Aug 2017
    Posts
    75

    Re: https://www.radbasic.dev/

    Last edited by germano.barbosa; Jul 3rd, 2019 at 06:14 PM.

  30. #70
    Addicted Member
    Join Date
    Aug 2013
    Posts
    236

    Re: https://www.radbasic.dev/

    Quote Originally Posted by germano.barbosa View Post
    https://sourceforge.net/projects/visual-basic/

    What do you think of this compiler?
    Thanks . One of the videos for this on youtube has this link in the comments

    http://www.kidev.net/index.htm

  31. #71

    Thread Starter
    Hyperactive Member
    Join Date
    Mar 2019
    Posts
    416

    Re: https://www.radbasic.dev/

    For those of us who make a living from our software there are commercial realities at play here. Yes VB6 is old. Yes it lacks some features but apart from the "coolness" of replacing it with something new what is there to gain? Why not just migrate to a newer mainstream platform (apart from the obvious)

    I for one would rather be in the semi mainstream with VB6 than out on a limb with something developed and supported by just a few people. I can imagine myself chasing down hard to find bugs and inconsistencies in compiled code or spending time trying to get the new IDE to work properly.

    We have reasonable committed support for the runtime from MS at the moment with their "it just works" statement.

    With the support of this and other forums I have not found too many things I cannot get done today.

    I absolutely love the idea of a code compatible replacement & 64 bit completion but as I make my living from this stuff for me personally it would need to come from a substantial organisation with a track record of being around so that when I am trouble there is a company to go to & other users to converse with to try and find a solution.

    The technical difficulties of doing this will be immense. One of VB's great strengths is that it reasonably reliably hides the complexity of what is going on under the hood. Duplicating this on the myriad of platforms I have to support from W2003 SP1 to today's OS's is going to be a mission.

    I agree with Olaf that there are just a few people in this and other forums with the skills, knowledge and sheer grit to get this done.

    It will be interesting to see how this turns out.

  32. #72
    PowerPoster yereverluvinuncleber's Avatar
    Join Date
    Feb 2014
    Location
    Norfolk UK (inbred)
    Posts
    2,235

    Re: https://www.radbasic.dev/

    Well, this forum is not created nor run by a large organisation and it works sufficiently well as you stated. The point is that there does not have to be a company behind it for it to succeed.

    I would prefer a FOSS solution as a community can then get behind it and it can become self-seeding as it were. However, much smaller or even one-man organisations have come up with successful products that people use daily.

    The only commercial products I use now are Photoshop, Windows and VB6 (and I can't buy a copy of VB6 from MS for love nor money), everything else is FOSS, from LibreOffice to open hardware monitor, all one man efforts or community driven products.

    When a product is a one-man show and it is closed source then that worries me a lot more as we all know how difficult it is for one man (or woman!) to maintain a software project right to the end, however, it CAN be done - see RJTextEd as an example

  33. #73
    PowerPoster yereverluvinuncleber's Avatar
    Join Date
    Feb 2014
    Location
    Norfolk UK (inbred)
    Posts
    2,235

    Re: https://www.radbasic.dev/

    This latter URL seems to be dead or taken over by someone else.

    Quote Originally Posted by germano.barbosa View Post
    What do you think of this compiler?
    Well it has always been a lovely piece of work that has never seen much progress in all the time it has been available. However, the recent update has given it a visual overhaul, the code is there for someone to build upon and it could act as jumping off point for someone to start building an IDE that looks similar to VB6.

    It seems that it could act as a useful starting point for a compiler too but that is not for me to say.

    Visia never had a forms generator and that is why I never used it myself for anything other than a trial.

    I do not remember the source code being made available for it in the past but now it is on sourceforge anyone can build the Visia IDE and compiler so it does seem like a nice bit of work.
    Last edited by yereverluvinuncleber; Jul 4th, 2019 at 05:01 AM.

  34. #74

  35. #75
    PowerPoster ChrisE's Avatar
    Join Date
    Jun 2017
    Location
    Frankfurt
    Posts
    3,042

    Re: https://www.radbasic.dev/

    well I would have hope if Microsoft made VB6 open source, and one or many people could build a VB7 - 8 or whatever you want to call it

    https://www.i-programmer.info/news/1...ource-vb6.html
    to hunt a species to extinction is not logical !
    since 2010 the number of Tigers are rising again in 2016 - 3900 were counted. with Baby Callas it's 3901, my wife and I had 2-3 months the privilege of raising a Baby Tiger.

  36. #76
    The Idiot
    Join Date
    Dec 2014
    Posts
    2,721

    Re: https://www.radbasic.dev/

    well, microsoft dont want, for some unknown reason they dont want to share the source.
    just open source c2.exe and link.exe could make a difference as well.
    wonder if anyone has asked that, to give the source of c2/link alone?

  37. #77
    Lively Member Grant Swinger's Avatar
    Join Date
    Jul 2015
    Posts
    71

    Re: https://www.radbasic.dev/

    I doubt we'll ever see VB6 open-sourced unless VBA is as well. And as long as VBA is part of the Office cash cow that probably won't happen.

  38. #78
    Hyperactive Member
    Join Date
    Jul 2002
    Posts
    481

    Re: https://www.radbasic.dev/

    New video posted showing RadBasic 64bit compiling..

    https://www.facebook.com/radbasic/vi...7638598297678/

  39. #79
    The Idiot
    Join Date
    Dec 2014
    Posts
    2,721

    Re: https://www.radbasic.dev/

    looking very promising axisdj.

  40. #80
    PowerPoster yereverluvinuncleber's Avatar
    Join Date
    Feb 2014
    Location
    Norfolk UK (inbred)
    Posts
    2,235

    Re: https://www.radbasic.dev/

    Quote Originally Posted by Grant Swinger View Post
    I doubt we'll ever see VB6 open-sourced unless VBA is as well. And as long as VBA is part of the Office cash cow that probably won't happen.
    In some ways this is a good thing as it means it will stimulate others to come up with an alternative solution, exactly as it is proving...

Page 2 of 7 FirstFirst 12345 ... LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  



Click Here to Expand Forum to Full Width