Vb6 , the Future, and what I have discovered

[Schmidt]

It's not essential but it's more productive. A lot of people have deadlines to meet.

That's not true - because the VB6-version would consist of *equally* efficient (if not more efficient) code
(we also have deadlines to meet, you know...).

Olaf

[Niya]

Yeah, but they best not be involved in THIS discussion.

To what are you referring to when you say "they"?

[Niya]

That's not true - because the VB6-version would consist of *equally* efficient (if not more efficient) code
(we also have deadlines to meet, you know...).

Olaf

Efficiency is not the problem. It's more about how quickly something that exists only in your mind can be manifested into reality.

[Schmidt]

None of the "fancyness" was ever needed.

Not true - VB6 clearly beats C-implementations, when "higher-level-stuff" like e.g. "GUI-frameworks" needs to be implemented.
(I've pointed that out, already in post #188)...

I don't disagree with anything you said, but you can't stop humanity for constantly trying new things.
Hopefully, it's to the purpose of making things better, but we often go through some swamps along the way, and often abandon the effort while in one.

True - and the only thing I'll add to that is, that so far no C# or VB.NET example was posted,
because of "fear, to be proven to wade through exactly such a swamp".

Nobody really want's to hear "some such stuff" about "their king" (the one who walks around naked).

Olaf

[Schmidt]

Efficiency is not the problem. It's more about how quickly something that exists only in your mind can be manifested into reality.

Yep, that's what I'm talking about the whole time (things that "only exist in your mind")... SCNR

Olf

[Shaggy Hiker]

To what are you referring to when you say "they"?

Those with deadlines shouldn't be participating in this discussion. I'm in a two day meeting.

[Shaggy Hiker]

Not true - VB6 clearly beats C-implementations, when "higher-level-stuff" like e.g. "GUI-frameworks" needs to be implemented.

Being better at it does not make it necessary, it just makes it nicer. I'm not advocating everybody going back to ASM, but if we can write something in a higher level language, then it would be possible to write it in ASM....given time. So, necessary is never the driver, only nicer.

True - and the only thing I'll add to that is, that so far no C# or VB.NET example was posted,
because of "fear, to be proven to wade through exactly such a swamp".

What are you talking about? Is this thread now supposed to be explicitly about how is .NET better than VB6, or vice versa? What are the bounds of that? What are the ground rules? Who gets to decide? How is the judging to be done? We already covered that any such attempt would devolve into partisan bickering, and nobody appeared to doubt that outcome.

[Schmidt]

Being better at it does not make it necessary, it just makes it nicer. I'm not advocating everybody going back to ASM, but if we can write something in a higher level language, then it would be possible to write it in ASM....given time. So, necessary is never the driver, only nicer.

In post #188 it was shown, that the resulting binaries were significantly smaller than the C-produced binaries
(although the same OpenSource-rendering-lib was used in the backend by both implementations, and is included in both "downloads").
And I've also pointed out, that I needed *significantly* less time to produce that result (than the highly skilled C-developers, Miguel de Icaza among them).

I'm not talking about "nicer" (and not about "Me") - I'm trying to make a point about the "achievable efficiency with the VB6-package"
(its IDE, its Object-Model, and its Compiler... as they exist today).

What are you talking about? Is this thread now supposed to be explicitly about how is .NET better than VB6, or vice versa?

No, but this thread is about the "future of VB6".

And thus also about "the language" and "the language-features" we "want or need" in a successor.

Now it gets constantly (relentlessly) pointed out (e.g. by Niya, Sitten and even tg), that "VB6 as it currently is",
is "not sufficient" (even "punch-card-like").

And that is just not true - and I argue, because VB6-community-members who follow this thread,
might think "mabe they are right, perhaps VB6 really is not suitable for challenging, modern stuff"
(which would then lead to "less support from the VB6-community" for the project, in turn).

What I know for sure is, that even when the new compiler would "only" be an 1:1 translation (feature-wise),
then we can already address "modern stuff with that VB6-compatible successor".

Though I never said, that the project should "stand still, after achieving that 1:1 compatibility".
Things which are comparably "cheap" to implement in addition (as e.g. Shift-Operators or unsigned Integer-Types, or Class-Inheritance),
will be implemented - but I'd suggest to do (and discuss) that in a second step.

Hoping to have made my stance more clear with the above text...

Olaf

[DEXWERX]

So in .NET 2.0 they changed it to the Event-Based Asynchronous pattern.

Sub ShowWebResults()
    AddHandler _httpClient.GetStringCompleted, Addressof WhenDataIsDownloaded
    _httpClient.GetStringAsync(AddressOf WhenDataIsDownloaded, url)
End Sub

Sub WhenDataIsDownloaded(ByVal sender As Object, ByVal e As GetStringEventArgs)
    RemoveHandler _httpClient.GetStringCompleted, AddressOf WhenDataIsDownloaded
    Dim data As String = e.Data
    webBrowser.LoadHtml(data)
End Sub

Is it me or is that even more complicated than a VB6 event handler?

Private WithEvents m_httpClient As HttpClient

Sub ShowWebResults()
    m_httpClient.GetStringAsync(url)
End Sub

Private Sub m_httpClient_GetStringCompleted(ByVal sender As Object, ByVal e As GetStringEventArgs)
    webBrowser.LoadHtml e.Data
End Sub

edit:VB6 automatically managed the event handling ConnectionPoints, and basically hides a lot of the boiler plate. No need for delegates or AddressOf.

[Shaggy Hiker]

In post #188 it was shown, that the resulting binaries were significantly smaller than the C-produced binaries
(although the same OpenSource-rendering-lib was used in the backend by both implementations, and is included in both "downloads").
And I've also pointed out, that I needed *significantly* less time to produce that result (than the highly skilled C-developers, Miguel de Icaza among them).

I'm not talking about "nicer" (and not about "Me") - I'm trying to make a point about the "achievable efficiency with the VB6-package"
(its IDE, its Object-Model, and its Compiler... as they exist today).


No, but this thread is about the "future of VB6".

And thus also about "the language" and "the language-features" we "want or need" in a successor.

Now it gets constantly (relentlessly) pointed out (e.g. by Niya, Sitten and even tg), that "VB6 as it currently is",
is "not sufficient" (even "punch-card-like").

And that is just not true - and I argue, because VB6-community-members who follow this thread,
might think "mabe they are right, perhaps VB6 really is not suitable for challenging, modern stuff"
(which would then lead to "less support from the VB6-community" for the project, in turn).

What I know for sure is, that even when the new compiler would "only" be an 1:1 translation (feature-wise),
then we can already address "modern stuff with that VB6-compatible successor".

Though I never said, that the project should "stand still, after achieving that 1:1 compatibility".
Things which are comparably "cheap" to implement in addition (as e.g. Shift-Operators or unsigned Integer-Types, or Class-Inheritance),
will be implemented - but I'd suggest to do (and discuss) that in a second step.

Hoping to have made my stance more clear with the above text...

Olaf

Ok, I'm fine with all that. I'm even fine with this thread now being about "the future of VB6". The original question pretty much died out in the first few posts. Axisdj seems okay with that, too, so it can go in whatever direction people want to take it. Somebody wanted to take it to a different forum, but it seems like this is the best place for it, though it is not now, nor ever was, a programming question.

[Sitten Spynne]

Is it me or is that even more complicated than a VB6 event handler?

Private WithEvents m_httpClient As HttpClient

Sub ShowWebResults()
    m_httpClient.GetStringAsync(url)
End Sub

Private Sub m_httpClient_GetStringCompleted(ByVal sender As Object, ByVal e As GetStringEventArgs)
    webBrowser.LoadHtml e.Data
End Sub

edit:VB6 automatically managed the event handling ConnectionPoints, and basically hides a lot of the boiler plate. No need for delegates or AddressOf.

I don't personally use WithEvents outside of designer code, but there's no real technical reason other than, "It's the C# in me showing through". The VB6 snippet you posted is almost copy-paste compatible with .NET: you'd need parenthesis around the parameter list to LoadHtml() and I'm only 75% certain it'd be a compiler error.

------Other topic:

I feel like some people are arguing "efficiency" to mean CPU cycles and other performance metrics, where other people are arguing "efficiency" to mean developer time. Make sure you're talking about the same efficiency with each other before arguing.

I really only jumped in because I wanted to point out a concern that being Windows Client focused does not seem like a very good idea in the current universe, and that there's an overal aversion to new features that can be detrimental. In the end, it's your project and when you design your own language you get to put what you want inside of it.

I got too involved and wrote too much. If you don't like the ideas I bring to the table, that's cool. I'm not the one embarking on this project and ultimately the features you include are up to you. I think you've got a good handle on how you want the project to progress.

[Eduardo-]

No one ever writes representative code when they discuss how VB is "so close to English".

Based on what you say that?
That's exactly the kind of arrogant thinking that deprecated VB6 in favour of .NET
I could suggest the opposite: that to have a language with a simple syntax (and a human friendly one) favours to write more advanced programs, and in less time.

But if you want to write very "representative code" then why don't you go ahead and do it in Assembler?

About your examples: I didn't say that VB6 was English, and I already said that it had abbreviations.
It is not English, but compared to others it is closer.

I remember that I read somewhere about a language that aimed to be as close to natural language as possible (I don't remember its name now)... but that approach can lead to something too verbose.
I don't say that VB6 is perfect or that it's the best possible, but it is in a fair balance in not being verbose and being understable (of course, with some study).

Then, you spend a whole post just to show how .NET evolved for trying to be more English like.
Doesn't that make my point then?
What's the story?

Did you ever hear that at a time it was said that VB6 developers were more productive than C++ ones?
That may be true perhaps even today, but VB6 became not so RAD lately as it used to be in the past.

[Eduardo-]

I also want to point that for the same reason (I mean, because it is closer to English) is why I prefer DAO over ADO.

FindFirst
FindNext
NoMatch
OpenRecordset
OpenDatabase

It is almost English, very intuitive.

It also has the possibility to handle the database structure in a simple and intuitive object oriented way. With TableDefs, Fields, Relations...

[Niya]

It seems this topic is destined to cause controversy every time it comes up and it's becoming clear to me that programming languages are religions in their own right. Anyone with a semblance of life experience knows that no amount of logic or reasoning could convince anybody that their religion is fake. A Christian or a Muslim would perform all kinds of mental gymnastics to justify the existence of his God. Nobody could convince him otherwise. Just like no one can ever convince me that MS didn't do the right thing with .Net and no one can convince a VB6 programmer that they did.

I recently decided to get a bit of perspective and I lurked elsewhere for threads like this. I found a couple in a C forum, one of which I had actually posted here before. Here it is for anyone curious. That thread, very much like this one consisted of an old vs new type argument between one person and the rest of the forum but instead of VB6 vs .Net it was ASM/DOS vs Windows/C. The guy wrote some database engine in ASM and he took issue with some of the differences between Windows and DOS, the most significant being that DOS was a real mode OS. He didn't like the fact that Windows didn't allow direct control over a PC's hardware like DOS did. He went to war over it and like many in this thread, constantly derided MS for the decision to abandon DOS for Windows. I nearly read that whole thread when I first found it and it was eerily similar to this one despite being about something completely different.

I found a couple other similar threads, but none so passionately dividing. Nonetheless, the conclusion that I have arrived at is that these kinds of topics are heavily tainted with subjectivity, therefore, no kind of consensus could ever be reached between those of us that like what MS did and those of you that didn't. As such, I'm bowing out of this discussion. And if I ever get involved in another shitstorm like this, don't be afraid to remind me of this post. Debating this really is a waste of time, whether it's VB6 vs VB.Net, DOS vs Windows or whatever. People will always use their subjective experience to justify their views which results in never ending debates. I need to try and remember this from now on and stay away from these talks.

One final thing. With regards to my pending response to Olaf about COM. While I haven't produced any working example or demonstration yet, I have been heavily researching COM and experimenting with it in VB6 and VB.Net and I have come to the conclusion that it could indeed comfortably support a programming language with some modern features like inheritance. One potential problem I could see as far as using COM as an object based system for a programming language would be it's data types. However, clever use of VARIANTS could overcome it's shortcomings so there's that. I'll keep working on it, though not for the purposes of this discussion. Only because I personally find it interesting.

And with that, I'm out of this discussion. Have a good day everyone.

[Jonney]

Many years ago, it was because so many people like you mistakenly said that VB6 was not suitable for developing Web-App, many people gave up the trying. Now is the time to correct the mistakes, and I'm learning to develop Web-App with VB6. For now, it is very good.

talking only, a sample says please.

[Sitten Spynne]

Nonetheless, the conclusion that I have arrived at is that these kinds of topics are heavily tainted with subjectivity, therefore, no kind of consensus could ever be reached [...]

This is more or less how it sits.

I can argue about this or that syntax trick until I'm blue in the face, but all said and done I feel like software engineering's taught me there is rarely an objective "best" in these kinds of discussions.

VB6 is a very good tool that needs a bit of a facelift to be useful, that's what this thread's about. VB .NET is an attempt at that facelift. I think it's "better" than VB6, but we're still talking about "a very good tool" vs. "a very good tool". The gap between the two is not very big. If you update VB6 in the ways you've said, it'll be even smaller.

[Niya]

This is more or less how it sits.

I can argue about this or that syntax trick until I'm blue in the face, but all said and done I feel like software engineering's taught me there is rarely an objective "best" in these kinds of discussions.

Yea, you can't get away from subjectivity in these kinds of topics. When I was lurking around the C forums and another C++ forum, I came across a discussion about C vs C++. While it didn't devolve into mud slinging like this one did, I did see a lot of the same arguments I saw here. C programmers were talking about how overly complicated and bloated C++ was.

While reading that thread I recalled an incident back back in the ZDoom forums, the one I mentioned that was about modding DOOM. Zdoom(a port of DOOM) has a scripting language called ACS that can be used to control how things behave in the game. It's syntax is identical to C. The lead programmer behind Zdoom wanted to replace it with another language called Lua and this single discussion had the potential to completely destroy the community, with many members including another developer who was integral to the project, threatening to leave. They derided it as being too much like BASIC which was wordy and verbose. Visual Basic even came up in that discussion if I recalled correctly.

The point I'm making is that when it comes to languages and development platforms, no matter where you go, you can find people who passionately love a particular language/platform just as easily as you can find people who passionately hate it. And I realize now that I underestimated just how passionate people can be about this kind of stuff.

While I myself can be quite passionate about a language, I will drop it almost without complaint if it becomes more of a burden than some other alternative. As much as I loved VB6, when I learned VB.Net, VB6 felt burdensome. And when the day comes that VB.Net has to bite the dust, I will leave it behind too. I'm already thinking about alternatives to target mobile devices because of how popular those platforms have become. Java seems to be the best bet for mobile. I'm not quite there yet though but it will happen eventually. Even on the Windows platform, I'm thinking about using C# as my go-to language because of how much it has exploded in popularity, even outside of the Windows platform. It's future is certainly brighter than that of VB.

Point is, I cannot wrap my head around why the VB6 supporters are so persistent in fighting this up-hill against the momentum of change. It makes no sense to me which is the underlying reason why I have been vocal in this thread. I realize that now and that's it's probably best to just accept it. There really is no point in trying to convince anyone of anything. What really matters is that we're all happy with our decisions.

[dilettante]

Well, you could just stop trolling discussions in the VB forum and stick to the .Net forums. If people want to use .Net they can do so. Forget the proselytizing.

How many people go into .Net forums and do this sort of trolling? It has to be very close to zero.

[Shaggy Hiker]

This thread wasn't about this topic, either. This thread was about supporting an alternative VB flavor to take off. Not a VB6 clone, nor an update. A different direction.

What it turned into is a different matter, and a bit indescribable. There have been a few different discussions weaving through this. A discussion of the compiler that Olaf is working on, a discussion of other basics, a discussion of .Net vs VB6, and a bit of pretty mild mud slinging. There was even a suggestion that the thread be moved elsewhere, which it reasonably could, but if it were moved to General Developer, it might as well be merged into the (now moribund) discussion along the same lines that happened over there, or the one in Chit-Chat that was a passive-aggressive anti-.NET masterpiece.

Ultimately, this is something people want to talk about. The threads on the topic started in other parts of the forum appeared to starve out, possibly due to the lopsided nature of participation. This may be the one environment where this conversation could continue. This thread would likely have starved out, as well, except that it has covered so many topics. Had it stuck with the original question and not strayed, it would have ended in less than a page. Had it stuck with Olaf's compiler it likely would have ended after two pages. However, it topped seven pages in just a couple days, largely because people always found a new nit to pick.

If the general consensus is that the discussion should end here, let me know and I will close the thread. At this point, people are being quite polite, and I thank everybody for that. Disagreeing without being disagreeable, so I haven't done anything with this despite it have long since left the original topic and despite it not being a specific programming question, but more philosophical in nature.

[jpbro]

Point is, I cannot wrap my head around why the VB6 supporters are so persistent in fighting this up-hill against the momentum of change.

I'm pretty much done with this thread too, but I thought I'd address this from my perspective.

I'm don't feel like I'm fighting change at all. What I have is a couple of decades investment in a product I've developed that is paying the bills. There have been 3 major versions of the product developed over that time, and there are many hundreds of thousands of lines of code.

If memory serves, the first version I worked on was in 16-bit VB4, when I came in to an existing project to add some features and fix some bugs. Shortly thereafter, it was clear the product would have to move to 32-Bit. There was a fairly straight-forward migration path from VB4 16-bit to VB4 32-bit (again as I recall). We skipped VB5 and then VB6 came out with all its bells and whistles. Again, there was a fairly straight forward migration path from VB4 32-bit to VB6. The major version changes were not full rewrites, much of the same code could be used with minimal or no modification (which had the benefit that user data files could easily be migrated between versions too).

Then VB.Net came out and I decided to give it a try since it was supposed to be the future for VB6 developers. It was just a bad experience. No straight-forward migration, felt like a completely different language, and the IDE was painfully slow and clunky IMO. Granted this was early days, but I had a reasonably powerful machine -certainly better than most of the users of my software, so I couldn't imagine them installing the gargantuan (at the time) .Net framework and feeling like they were getting a downgrade in performance for what would have to be a completely new version of the software.

So I had a choice - rewrite the whole thing, or find a way to modernize VB6 to extend my software. If I was going to rewrite completely, it would not be in VB.net based on my experiences at the time. I've since heard some people claim that things are much better now, but then I hear from others in this very thread that maybe the IDE is getting clunky again, so that's not very inspiring.

I dabbled in a few different languages, and even considered doing a web only version. The web stuff was pretty lame back then in the pre-HTML5 days - many of my users still on old IE6 at the time I believe. Everything seemed like a step backward in terms of performance, functionality, features, etc...

It was around this time I stumbled on Olaf's work (at the time I think it was dhRichClient3, but maybe even any earlier version). The RPC classes & SQLite functionality inspired me to build a client-server version of my software since I could use my existing VB6 classes and access them over a network easily. It was work, but it felt like incremental work - I could do a bit here and there, working on individual areas and after some time have everything migrated with the bulk of the code still in place and just some new client-side wrapper code to glue it all together. Around this time some third-party paid OCXs also helped modernize the appearance of my software (this was pre RC5 Cairo). I updated the front-end without changing the backend code, so while it was a lot of work it still felt "incremental" and familiar (since I was still working with the language, IDE, and other tools that I knew very well).

So at this point I had a nice modern looking, well-performing, client-server desktop app...and then HTML5 really starts to take off :O So I can either dump it all and rewrite everything just for the browser, or I can try to keep as much of my old code as possible and glue a web front-end on to it. Well that extension approach worked pretty well last time, so let's try it again Fortunately, a lot of the "hard" work was previously done moving logic out in to separate classes in order to accommodate the RPC client-server migration, but I could not find a good way to interface my code with the browser (including much needing things like SSL). I happened upon Nginx which included FastCGI protocol support, but I couldn't find any VB6 FastCGI implementations. So I rolled my own, and in a short time I was able to put together a proof of concept web-front end for my VB6 server backend. This combined with other emerging niceties like LetsEncrypt, Linux VPS Hosting, WINE, etc... meant it was surprisingly easy & inexpensive (in terms of time and money) to get stuff from my VB6 backend to the browser over an encrypted connection.

So here we are today.There are other things that would be nice to have like 64-bit support (especially useful for the backend server software), cross-platform desktop development (I'm seeing more Macs lately, though I suspect that might revert now that Apple seems a bit lost on the professional computer side), etc...

Should I go for a full rewrite now? Or hope we can get an extended VB6 going before the bell tolls?

I'm pretty sure of this though - it won't be VB.net I choose if I have to move to a whole new platform & rewrite.

[Shaggy Hiker]

The IDE isn't getting clunky, it's getting overly ambitious. That makes it less performant, I just wouldn't call it clunky. It tries to do a dozen things on every keystroke, and a bunch of those things are graphics. So, VS is beginning to have the performance curve of a graphics-heavy game engine: If you run it on a powerful machine, it's great. If you run it on a less capable machine, it can lag. Some of those features are really nice. At least one (trivial) feature in VS2017 is something that would have been nice from the very beginning, but they all put a burden on the hardware. Games deal with this by allowing the user to opt out of certain levels of detail or graphical features if their hardware can't handle it smoothly. VS may need to go the same way. The graphical features are things I don't need, even if I do find them convenient.

I guess I ought to be a bit more clear. I didn't name the feature I found really convenient because I don't know what the thing is called. It's the bit that draws a faint line connecting things like If to End If, of For to Next. If you have a few levels of nesting, those visual cues make the code considerably more clear, though I don't know what they are called. I could live without them, and do quite often, but it IS a useful feature.

[Elroy]

I'm don't feel like I'm fighting change at all. What I have is a couple of decades investment in a product I've developed that is paying the bills. There have been 3 major versions of the product developed over that time, and there are many hundreds of thousands of lines of code.

If memory serves, the first version I worked on was in 16-bit VB4, when I came in to an existing project to add some features and fix some bugs. Shortly thereafter, it was clear the product would have to move to 32-Bit. There was a fairly straight-forward migration path from VB4 16-bit to VB4 32-bit (again as I recall). We skipped VB5 and then VB6 came out with all its bells and whistles. Again, there was a fairly straight forward migration path from VB4 32-bit to VB6. The major version changes were not full rewrites, much of the same code could be used with minimal or no modification (which had the benefit that user data files could easily be migrated between versions too).

Then VB.Net came out and I decided to give it a try since it was supposed to be the future for VB6 developers. It was just a bad experience. No straight-forward migration, felt like a completely different language, and the IDE was painfully slow and clunky IMO. Granted this was early days, but I had a reasonably powerful machine -certainly better than most of the users of my software, so I couldn't imagine them installing the gargantuan (at the time) .Net framework and feeling like they were getting a downgrade in performance for what would have to be a completely new version of the software.

...

I'm pretty sure of this thought - it won't be VB.net I choose if I have to move to a whole new platform & rewrite.

Well said, jpbro. Regarding what I quoted, I am in precisely the same boat.

It's as if Microsoft just truly can't see the harm they've done (and continue to do by continuing to drive out support for VB6, and adamantly refuse to update it or open-source it), and the netters just really don't "get" what we're talking about.

Best Regards,
Elroy

[jpbro]

The IDE isn't getting clunky, it's getting overly ambitious. That makes it less performant,

Sorry, clunky was the wrong word - slower/more bloated/less performant was what I was referring to.

[dilettante]

But the VBA Editor in MS Office products hasn't changed either, at least not very noticeably aside from minor but nice UI gingerbread. And that's still considered fully supported.

I think it is simply that most, nearly 100%, of Microsoft has moved on and too many years have gone by. I'm just happy we get the minimal level of support for VB6 that we do. It could easily have stopped being usable at all in Windows Vista and beyond.

[Eduardo-]

It seems this topic is destined to cause controversy every time it comes up and it's becoming clear to me that programming languages are religions in their own right. Anyone with a semblance of life experience knows that no amount of logic or reasoning could convince anybody that their religion is fake. A Christian or a Muslim would perform all kinds of mental gymnastics to justify the existence of his God. Nobody could convince him otherwise. Just like no one can ever convince me that MS didn't do the right thing with .Net and no one can convince a VB6 programmer that they did.

I recently decided to get a bit of perspective and I lurked elsewhere for threads like this. I found a couple in a C forum, one of which I had actually posted here before. Here it is for anyone curious. That thread, very much like this one consisted of an old vs new type argument between one person and the rest of the forum but instead of VB6 vs .Net it was ASM/DOS vs Windows/C. The guy wrote some database engine in ASM and he took issue with some of the differences between Windows and DOS, the most significant being that DOS was a real mode OS. He didn't like the fact that Windows didn't allow direct control over a PC's hardware like DOS did. He went to war over it and like many in this thread, constantly derided MS for the decision to abandon DOS for Windows. I nearly read that whole thread when I first found it and it was eerily similar to this one despite being about something completely different.

I found a couple other similar threads, but none so passionately dividing. Nonetheless, the conclusion that I have arrived at is that these kinds of topics are heavily tainted with subjectivity, therefore, no kind of consensus could ever be reached between those of us that like what MS did and those of you that didn't. As such, I'm bowing out of this discussion. And if I ever get involved in another shitstorm like this, don't be afraid to remind me of this post. Debating this really is a waste of time, whether it's VB6 vs VB.Net, DOS vs Windows or whatever. People will always use their subjective experience to justify their views which results in never ending debates. I need to try and remember this from now on and stay away from these talks.

One final thing. With regards to my pending response to Olaf about COM. While I haven't produced any working example or demonstration yet, I have been heavily researching COM and experimenting with it in VB6 and VB.Net and I have come to the conclusion that it could indeed comfortably support a programming language with some modern features like inheritance. One potential problem I could see as far as using COM as an object based system for a programming language would be it's data types. However, clever use of VARIANTS could overcome it's shortcomings so there's that. I'll keep working on it, though not for the purposes of this discussion. Only because I personally find it interesting.

And with that, I'm out of this discussion. Have a good day everyone.

I think that you are very wrong.
A lot of reasons have been given, so it seems to me that if you decide to ignore them the religious one is you.

Not all the people are the same, some people like to know a litle of many things and no to go in deep in anything, others like to know well something and invest a lot of time on that something.

Some people can feel confortable with some tool, other dislike it.
So I think you need to be more open, because how you see the things may be not be the only way to see them.

[Eduardo-]

The gap between the two is not very big.

It is as big as a full rewrite of hundreds of thousands of line of code (just in my case).

[Eduardo-]

While I myself can be quite passionate about a language

If you are not passionate, I only can think that you don't have big programs written in a language that you need to mantain working.
Or you don't care about your work or your life.

Perhaps you do programming as a hobby. Perhaps you are rich and don't need to work...
Or you are a "consultant", that know a bit of many things and advise others but don't do any actual programming entrepreneurship.

Well, you could just stop trolling discussions in the VB forum and stick to the .Net forums. If people want to use .Net they can do so. Forget the proselytizing.

How many people go into .Net forums and do this sort of trolling? It has to be very close to zero.

Exactly, I was thinking the same thing.

[Shaggy Hiker]

Sorry, clunky was the wrong word - slower/more bloated/less performant was what I was referring to.

It's got ALL of that!

[Eduardo-]

There was even a suggestion that the thread be moved elsewhere, which it reasonably could, but if it were moved to General Developer

When you move something from here to some other sub-forum, for many of us (and I can guess that the most of us), it is the same as to delete it.

[Shaggy Hiker]

Not most of you. Most of the people who have contributed to this thread also participate frequently in other forums. Furthermore, the request was made by somebody who participates primarily in this forum. I didn't move the thread simply because it seems like this is a pretty good place for it.

[Eduardo-]

Yes, it is a moment of change. A moment to go back to VB6. And to make a successor.
I don't understand why some people fight this that much.
They only come here to say that "they did well" (leaving) and we did wrong (staying).
They are haters.

[Eduardo-]

Not most of you. Most of the people who have contributed to this thread also participate frequently in other forums. Furthermore, the request was made by somebody who participates primarily in this forum. I didn't move the thread simply because it seems like this is a pretty good place for it.

In my case, I only load this forum, and from time to time go to the codebank to take a look.
I don't go almost to any other place in the forum.

[Shaggy Hiker]

That's pretty reductionist. People have taken considerable time laying out arguments for this and that. To sum it all up with: They are haters.....heck you should run for the US Congress.

[labmany]

I'm pretty much done with this thread too, but I thought I'd address this from my perspective.

I'm don't feel like I'm fighting change at all. What I have is a couple of decades investment in a product I've developed that is paying the bills. There have been 3 major versions of the product developed over that time, and there are many hundreds of thousands of lines of code.

If memory serves, the first version I worked on was in 16-bit VB4, when I came in to an existing project to add some features and fix some bugs. Shortly thereafter, it was clear the product would have to move to 32-Bit. There was a fairly straight-forward migration path from VB4 16-bit to VB4 32-bit (again as I recall). We skipped VB5 and then VB6 came out with all its bells and whistles. Again, there was a fairly straight forward migration path from VB4 32-bit to VB6. The major version changes were not full rewrites, much of the same code could be used with minimal or no modification (which had the benefit that user data files could easily be migrated between versions too).

Then VB.Net came out and I decided to give it a try since it was supposed to be the future for VB6 developers. It was just a bad experience. No straight-forward migration, felt like a completely different language, and the IDE was painfully slow and clunky IMO. Granted this was early days, but I had a reasonably powerful machine -certainly better than most of the users of my software, so I couldn't imagine them installing the gargantuan (at the time) .Net framework and feeling like they were getting a downgrade in performance for what would have to be a completely new version of the software.

So I had a choice - rewrite the whole thing, or find a way to modernize VB6 to extend my software. If I was going to rewrite completely, it would not be in VB.net based on my experiences at the time. I've since heard some people claim that things are much better now, but then I hear from others in this very thread that maybe the IDE is getting clunky again, so that's not very inspiring.

I dabbled in a few different languages, and even considered doing a web only version. The web stuff was pretty lame back then in the pre-HTML5 days - many of my users still on old IE6 at the time I believe. Everything seemed like a step backward in terms of performance, functionality, features, etc...

It was around this time I stumbled on Olaf's work (at the time I think it was dhRichClient3, but maybe even any earlier version). The RPC classes & SQLite functionality inspired me to build a client-server version of my software since I could use my existing VB6 classes and access them over a network easily. It was work, but it felt like incremental work - I could do a bit here and there, working on individual areas and after some time have everything migrated with the bulk of the code still in place and just some new client-side wrapper code to glue it all together. Around this time some third-party paid OCXs also helped modernize the appearance of my software (this was pre RC5 Cairo). I updated the front-end without changing the backend code, so while it was a lot of work it still felt "incremental" and familiar (since I was still working with the language, IDE, and other tools that I knew very well).

So at this point I had a nice modern looking, well-performing, client-server desktop app...and then HTML5 really starts to take off :O So I can either dump it all and rewrite everything just for the browser, or I can try to keep as much of my old code as possible and glue a web front-end on to it. Well that extension approach worked pretty well last time, so let's try it again Fortunately, a lot of the "hard" work was previously done moving logic out in to separate classes in order to accommodate the RPC client-server migration, but I could not find a good way to interface my code with the browser (including much needing things like SSL). I happened upon Nginx which included FastCGI protocol support, but I couldn't find any VB6 FastCGI implementations. So I rolled my own, and in a short time I was able to put together a proof of concept web-front end for my VB6 server backend. This combined with other emerging niceties like LetsEncrypt, Linux VPS Hosting, WINE, etc... meant it was surprisingly easy & inexpensive (in terms of time and money) to get stuff from my VB6 backend to the browser over an encrypted connection.

So here we are today.There are other things that would be nice to have like 64-bit support (especially useful for the backend server software), cross-platform desktop development (I'm seeing more Macs lately, though I suspect that might revert now that Apple seems a bit lost on the professional computer side), etc...

Should I go for a full rewrite now? Or hope we can get an extended VB6 going before the bell tolls?

I'm pretty sure of this though - it won't be VB.net I choose if I have to move to a whole new platform & rewrite.

Story of my life!!

[Eduardo-]

That's pretty reductionist. People have taken considerable time laying out arguments for this and that. To sum it all up with: They are haters.....heck you should run for the US Congress.

I can see that. They hate people that din't do the same thing that they did.
Perhaps it is because they don't feel sure that they did the right thing.

On the other hand, I don't hate them because they left.
I'm sure about my decission because I know what my reasons are to stay in VB6.
They for sure have valid resons (for them) to chose another language, their circunstances may be different.

If someone doesn't care about this language, why does he come here trolling?

[labmany]

I can see that. They hate people that din't do the same thing that they did.
Perhaps it is because they don't feel sure that they did the right thing.

On the other hand, I don't hate them because they left.
I'm sure about my decission because I know what my reasons are to stay in VB6.
They for sure have valid resons (for them) to chose another language, their circunstances may be different.

If someone doesn't care about this language, why does he come here trolling?

Couldn't agree more.

I never visited the .NET forums and probably will never do, I do not care to visit or troll although I 'hate' .NET since its first release!

[Niya]

I think that you are very wrong.
A lot of reasons have been given, so it seems to me that if you decide to ignore them the religious one is you.

Those reasons don't make sense to me. Sorry. I wish it did, but it just didn't.

and the netters just really don't "get" what we're talking about.

If you are not passionate, I only can think that you don't have big programs written in a language that you need to mantain working.

I already rewrote all the ones that mean anything. A couple of them, quite large.

The biggest application we have was written in VB6 over the last 20 or so years, with quite a few contributions from me. It is currently being maintained and updated in VB6 just fine. Porting that one is inevitable though as VB6 is beginning to show it's age. Porting it would be nothing new because it is itself, a rewrite of a very large and much older version that was written in QuickBasic. Only one client still runs that one today. I've been around long enough to know that this kind of thing can be expected every 15-20 years. Having to do re-writes and ports every decade or so isn't all that bad to me. I'm kinda used to it.

Perhaps you do programming as a hobby. Perhaps you are rich and don't need to work...
Or you are a "consultant", that know a bit of many things and advise others but don't do any actual programming entrepreneurship.

Let's not start getting personal here. For the most part, I stayed within the bounds of what could be considered a technical/philosophical discussion. I never made any speculation as to what any of you do in your personal lives or made any passive-aggressive attempt to pass judgement on anyone here as a person. Check that disrespectful tone when you talk to me please. I might disagree strongly with some of you here but I don't think unkindly of any of you.

[ChrisE]

he guy's just let this Thread die..

it's like a lot of Meeting's, 10 People go in, and nothing really come's out
except the coffee is empty , the peanuts were old....


VB vs .Net in VideoCut
https://www.youtube.com/watch?v=fqtxZklEZuI

[FunkyDexter]

Ha! I've barely been around the forum for about a year but I see this discussion hasn't lost any of it's venom

Personally I think there's a bunch of viable options for what to do next if you're currently bound to VB6.

1. You can continue to use VB6. You might get the occasional pain point when MS do some OS release that breaks some feature but, on the whole, they've done a pretty fair job of keeping it working for you even if they haven't been giving you new features. It's an option with a shelf life but it feels to me like you've still got a decent amount of time as long as you're focussed on desktop.

2. You can take Olaf's approach and write your own language. I've got to admit that seems like an awful lot of effort for minimal reward to me but it would mean you would have exactly feature set and underlying architecture you want. And you wouldn't have to worry about it being abandoned because it's entirely in your hands. Of course, whatever you base it on could face that threat but unless you're going to start coding in binary you're standing on someone's shoulders.

3. You could move to .Net. That's what I did and it worked out well for me. I get that some of you have had less happy experiences, mainly down to the lack of an upgrade path (which, yeah, was a BIG dropped ball by MS) and I can understand why you wouldn't want to bite that bullet again. But, based on my own experience, I'd urge anyone to give it a try.

4. You could take AxisDJ's approach and start scouting out alternative Basic style languages. Whether that's FreeBasic, RichClient or some other choice, they're all worth a punt. My guess would be that most won't offer you a happier upgrade path than .Net did but some might (I get the impression RichClient does) and it's worth investigating. Come to think of it, If you can put any latent distrust of MS aside, moving to .Net belongs in this category really. Be aware that if it's not open source then you face exactly the same "it could be abandoned" dilemma as many of you already experienced with VB6. It's up to you to look at the producer and make the best objective judgement you can about how likely that is to happen. Of course, picking an open source alternative tends to be fraught with it's own problems, many of which Olaf has identified as the reasons he doesn't want to take his own offering down that path. Open Source stuff is notoriously messy unless someone is keeping a tight grip on it's direction and half the time you end up writing half the ruddy thing yourself, at which point see option 2.

5. You could go for a completely different flavour of language and take your development in a different direction. You surely won't have a happy upgrade path but, if that's not an issue, you're likely to garner (wrongly) much more respect from your peers and a higher wage packet. Like it or lump it, most of the rest of the developer world sneers at both VB6 and VB.Net - VB6 more so. It's not justified but there you have it.

6. You could leave development altogether and start a taco truck (I wonder how that worked out - well, I hope)

All of the above are viable options. Which is right for you depends on your particular circumstances and the chances are none of them are quite right so you're going to have to do something that involves a bit of pain (or possible salsa). But if you're looking at any of the above and saying "that's just stupid" then you probably need to check your attitude because you've lost your objectivity. "Wrong for you" is not the same as "stupid".

I personally don't buy the "MS screwed us over" argument. MS simply took their flagship language in a direction they thought would best future proof it. They may not have handled it particularly well (apart from the lack of upgrade path I think they did a pretty fair, if imperfect, job but that's a subjective opinion you are welcome to disagree with) but the implication that it was some sort deliberate, vindictive attack on you or your software is a bit silly. It was arguably a careless act, but not a vindictive one.

And however you cut it, you're still using it over a decade after they made that decision. If it was an attack it must have been the most monumentally worthless one in history. And while I get that there may have been some bumps along the way and those bumps have cost you sweat, pain and money, the fact that it's still working and is going to continue to do so for some time yet speaks volumes about backward compatibility.

One other thing I'd say, we all have a vested interest in where VB6 goes next because we're all developers and where it goes may or may not be of use to us. I haven't touched VB6 in well over a decade but, believe me, if some iteration comes about that closes a gap I want closed I'll be jumping all over it. A non-VB6 developer coming into this thread and expressing an opinion you disagree with is not trolling, it's someone expressing an opinion about an issue that could affect them. If you can't tell the difference between disagreement and trolling it's a pretty sad state of affairs.

Thankfully I've seen very little trolling in this thread. It's got a little heated at times but, on the whole, people on every side of the various debates that have wended their way through it have expressed their views pretty well without resorting to childish name calling, baiting or any of the other behaviours we would consider trolling. I genuinely applaud you all for that, we've seen enough of these threads go South and it's better if this one doesn't follow suit.



edit>"VB vs .Net in VideoCut"
I lolled! I lolled like drain!

[Eduardo-]

Those reasons don't make sense to me. Sorry. I wish it did, but it just didn't.

It is not that they don't make sense, they don't "make sense" to you because you are not interested.
You are only interested in defending your own decisions.

I already rewrote all the ones that mean anything. A couple of them, quite large.

How many lines? (don't lie)
How many years did it take to you to rewrite them?

Let's not start getting personal here. For the most part, I stayed within the bounds of what could be considered a technical/philosophical discussion.

Another lie. You started accusing of being religious to the ones that defend our points of view with arguments.
But your argument are "things change", "move on", "is time to leave".
Are those "arguments", really?

All you show with your messages is that you don't care about the future of VB6. And if you don't care, why are you writing here?
In my world that's called trolling.

I never made any speculation as to what any of you do in your personal lives or made any passive-aggressive attempt to pass judgement on anyone here as a person. Check that disrespectful tone when you talk to me please. I might disagree strongly with some of you here but I don't think unkindly of any of you.

You are not respecting us.
You are the one that comes to a VB6 forum to accuse of "religious" to people that exposed reasons.

Do you want me to repeat all the reasons that were exposed?
If you come here attacking to what is our livelihood, don't expect a smile.

We take this issue very seriously, and I see you don't.
Of course: you don't need it.
So don't come here saying nonsense.
That's trolling.