On 20/01/2009, at 8:22 PM, Martin Schröder wrote: > 2009/1/20 Joseph Wright <[log in to unmask]>: >> I've had a go at summarising the key aims of LaTeX3, as I understand >> them, on my blog (http://www.texdev.net). I'd be interested to know > > Very interesting, thanks. > > However, most of these aims bring no direct benefit for the user, > while other areas where LaTeX is sorely showing it's age, e.g. > hyperref, are missing. I think any LaTeX3 release without a direct > integration of hyperref's features will not attract many users. The > same goes for document classes (e.g. KOMA) and a much better > integration of utf8 and fonts. Definitely, of course. It's just that most of the work done so far and available at the moment is in the area of the underlying groundwork (i.e., the expl3 modules and the xpackages) before this sort of thing can be attempted. Designing the user-level interfaces will be one of the last pieces of the puzzle, when we know what features we have to play with. There will probably be several iterations between specifying what we want the user interface to be and writing the code to implement it. The vast scope of material on CTAN gives us a pretty good starting point for working out what should be included by default. As Frank has mentioned before, the functionality described in the entirety of The LaTeX Companion could be considered a first approximation of what a LaTeX3 user interface might be. Writing a specification for a LaTeX3 interface is a huge task for a small team to tackle, perhaps greater in scope than writing from scratch the entire HTML and CSS specifications. Somewhat of a daunting task. It's probably a foregone conclusion that we'll need further help from members of the community when the time comes for doing this. But it's never to early to start... Cheers, Will P.S. (While I am working on some LaTeX3 code right now, these are obviously entirely my own opinions and might even be in opposition to the ideas of other members of the LaTeX3 team!)