I'm not sure. I mean I was a fly-on-the wall and read the comments as they unfolded, but I can see it from both sides.
Paul probably meant it as a joke, and he points out that he considers it a "lame" version because it contains only a subset of the improvements he's made. I can see how that could be particularly annoying, to have put that much work in, only having to back-track and essentially make a version where all that is left out. Might make him feel like why did I bother?
But I can also see it from the point of view of others, myself included here - who were interested in reducing down the perceived bloat of Livecode. Does it really have to be as large in size as it is? Turns out most of the 'bloat' is the browser CEF engine (which is non-functional in Linux anyway for most distros, and could perhaps be replaced with something else.** more on this in a moment)
I still wonder if standalones can be honed so they generate smaller apps, but this isn't a reflection on Paul or his work - it's just that in the interests of efficiency, I think people may have wondered if more bloat was being added.
Turns out, I really don't think there has been any added. The features I've seen and incorporated into the OXT Lite version (visual guides, dark mode, script colourisation themes), I'd consider things that any good IDE should really have.
Size breakdown: (at least on Linux). The CEF engine is a 200+MB, but the real things that add to the size are the runtime engines).
![size-breakdown.png](./download/file.php?id=1492&sid=30d7fca4acf87f2338e2b8f17ce31340)
- size-breakdown.png (59.44 KiB) Viewed 6533 times
If we wanted to go really minimal, we could use OpenXION-1.4, which is essentially command-line xTalk. No IDE. The site is mostly down, and what I've got - I grabbed from thewaybackmachine and archive.org snapshots.
The same way that Runrev have done, they've essentially put an interpretation layer on top of xTalk to make what is now Livecode. We could do the same with OpenXion with a Page TKinter python IDE on top of it. So it walks the xTalk walk, looks like we want it look, and we have absolute control over how we script
everything. No dependencies on engines which contain nasty surprises or code errors from Livecode.
I'd like to go this approach with a minimal version, but it does mean having to re-invent the wheel so to speak. A lot of work would be involved in getting a useable GUI on top of OpenXion, but the result would be a lean efficient IDE.
I don't want to offend Paul, because it's very evident he's put a ton of work in, for what is essentially a thankless task. We are all doing this for free because I think we share a common goal of getting this thing (whatever you call your particular fork of it), to be as good as it can be.
From that point of view, I think we are all on the same page.