> Karl Berry <[log in to unmask]> writes:
>> little reason to continue even with XeTeX.
>> Just for the record, even in the hypothetical future where LuaTeX has
>> caught up to XeTeX in terms of system font access and Arabic
>> typesetting, Jonathan K once spoke to me about another reason why
>> XeTeX will very likely always have a place: good typesetting of other
>> minority scripts.
This is, imo, the big advantage of LuaTeX over XeTeX: the latter
relies on what the font provides, while the former allows changing
their default behaviour to fix and extend their capabilities.
>> Since everything in LuaTeX is being programmed from
>> scratch, it seems highly unlikely it will ever support the variety of
>> scripts that ICU et al. do.
> It would seem to suggest itself to me that one part of dropping XeTeX as
> a separate project would entail module loader and interfaces into ICU.
> Ideally, one would have built-in support for the more important LuaTeX
> targets, with seamless ICU fallbacks for the rest.
And... The aim is to provide tools to extract data from the fonts
in a generic way, so that they could be used in a similar way XeTeX
does -- no need to support scripts separately, and therefore no need
to write "everything" from scratch.
And... LuaTeX or XeTeX are essential for non English documents. The
age of active characters, at last, coming to its end.
(btw, I got a copy of Beginning Lua Programming, and in my free
time I studying how to use Lua.)