OpenXTalk Forums

Wayback, Wayback . . .

https://web.archive.org/web/20110809000 ... ecode.com/

Reading this page from 2011 (when the Open Source version of LiveCode was released) is both informative and thought-provoking:
- -
Areas that seem interesting insofar as they did not come to fruition:

Themes

In order to facilitate the open language project, the IDE requires the notion of a scoped project

A palette architecture will allow developers to extend the IDE without having to work on the IDE itself

An object architecture allows developers to extend LiveCode with new objects

(this might be what is involved in the new LiveScript Widget thing)

The object architecture will be used to wrap the existing LiveCode objects

We need to ensure that all existing LiveCode script will continue to work as it currently does

(This was broken with the introduction of more robust Unicode handling, as a lot of the functions handled by numToChar and charToNum were shifted to numToCodePoint and codePointToNum . . . but not all of them: stacks using these functions had to be rewritten)

We are replacing all engine string usage with a modern form that will allow us to provide transparent unicode support

(This did happen, but see the 'price' above)
- -

I don't think a Physics engine was ever forthcoming.

Neither was the box2d support, and a lot of the things on this list may have completely had their priorities changed.

It would have been great to see a lot of them come to fruition.

The unicode support seems to have been done, but I wonder why they were contemplating internalising the documentation (wrapping it inside the engine source code).

I'm currently thinking of doing the exact opposite - making it as externalised as possible, even to the point of translating it all to a regular IDE stack - but a dynamic stack that can load documentation for user plugins as required (not just to keep Paul happy - that's a joke, not a dig at Paul - just because I know we talked about the importance of this loading dynamically for plugin authors in the past). That in itself is a lot of work - just getting it out of the database, although I've written a script to do that and chug through it.

(edit: split into a new topic, which continues here)