Business Application Framework

[Monte Goulding]

> It seems that it's very much in the game.  Peter's post was especially helpful for two reasons:
> 
> - It confirms the inherent difficulty in creating a general-purpose
>  VCS tool that covers all edge cases.

Yes there were a few of curly issues I had to get my head around.
> 
> - It clarifies that LiveCode's solution is even more limited in terms
>  of applicable uses cases, specialized for use in this very specific
>  business framework system.

Indeed
> 
> Given this, I think it's well past time for us all to give lcVCS a second look as the community's solution for version control.
> 
> IIRC the edge cases it may not handle are relatively uncommon anyway, and the value it provides has been strong enough for Trevor to write some very glowing things about.
> 
> Let's embrace it, enhance it, and make it a centerpiece for all of us who use Github.

It would be great to build some interest. One of the problems I found is there’s relatively few people in this community that have any version control experience and so most of them think they don’t need it because they work on their own etc. I think this caused by the circular problem of developers that need version control don’t look at LC because we can’t do it and therefore there’s little understanding of version control in the community. People I have worked on projects with using lcVCS like Trevor and Martin seem to love being about to review their change history etc. Martin didn’t have any version control experience and now works largely on his own but continues to find it helpful. Trevor simply wasn’t interested in working with anyone else unless he had version control.

The project from my perspective has two parts. lcVCS is the engine that manages the file format and is GPL. Then I have an IDE plugin and command line interface that I was intending to sell. The plugin provided some cool git integration into the IDE and the command line interface provided something for git hooks to rebuild your stacks when you merge or checkout and to export them asynchronously after an IDE save of the stack so you don’t interrupt your workflow with stackFile exports. The market for such a thing is quite small compared to the work that goes in so the deal to sell it to LiveCode where it would become a regular part of the IDE was appealing but it didn’t come off. At this stage if I were to get stuck into it again I’d like to merge both projects and release under GPL but I’d need some financial backing to afford the time...
> 
> 
> >> 2. Why hasn't it been submitted to the only resource-sharing tool we >> have built into the IDE?
> >
> > I'm not sure I've ever looked in there... Is it worthwhile?
> 
> "Sample Stacks" is a bit of a turn-off, and the older "RevOnline" name wasn't much better. But the role is very very worthwhile: it's where all of us can share stack files easily.

I’ll upload today.

> > Before you get excited lcVCS doesn't support LC 8 yet because I'd
> > need to depend on IDE code to do it at the moment and that seems
> > risky. That and there appeared to be no point working on it until
> > today...
> 
> I wouldn't worry about v8 just yet.  It's a very exciting set of new technologies, but it's going to be some months before it's as robust and performant as the current version, and there are still a few design decisions to be completed.

I’d like to support 8 if I can so hopefully Peter et al have some ideas on the widget metadata (which widgets are loaded and what their properties are etc).