Capability-dependent content/format

[Klaim - Joël Lamotte]

After having watched this presentation by Scott Meyer: http://blip.tv/oreilly-tools-of-change-for-publishing-conference/toc-09-authoring-challenges-in-a-multiplatform-world-scott-meyers-1986559

I figured that AOSL isn't fixing the problem of having one "manuscript" (original source of content) for all targets.
At best, it allows to write one manuscript by target.
It would be more helpful to allow the author to specify things that are capability-dependent, inside the original source document.

I'm looking for ideas as to add features allowing platform capability-dependent content and formatting, for the future versions of AOSL.

This impact also text resources so I believe it's a really complex subject.

Joel Lamotte

[Klaim - Joël Lamotte]

After some thought, I'm conviced this is an important feature:

An AOSL file should describe in one place a sequence for different interpreters which have different capabilities.

This is not true at the current time because you need either:

 - different AOSL files for each specific implementation;

 - one AOSL file that don't use specific capabilities;

So we need a third option that is 

 - one AOSL file that contain both a default description of a sequence, and capability-dependent variations too;

I have some ideas of how to implement this. I will write some drafts as soon as I have the AOSL 1.0 documentation ready (in a few weeks).

Joel Lamotte