Lists Home |
Date Index |
On Sat, 2002-02-02 at 21:29, Rick Jelliffe wrote:
> Simon's Regular Fragmentations seems to provide a nice reworking
> of Short References with the XML zeitgeist: no entities and
> synchronous with the containing element. (This probably makes
> Regular Fragmantations only suitable for inline short-cuts, such as date fields
> and case citations, so it is not as general as Short References
> which also would handle paragraph breaks, lists starting with "*",
> etc., better, I suspect)
That's a good description of Regular Fragmentations' current state, but
I'm hoping to go well beyond that. MOE is capable of representing
content that isn't well-formed (even as trees, with some extra work on
their construction), opening up a lot of other possibilities.
I'm hoping to eventually make it easy to build parsers which take in
text information and report some interpretation of that text as markup
(you can do that with SAX today), and to provide multiple levels of
processing on that content, kind of a combination of parsing and
fragmentation. Applications shouldn't have to notice this work, and will
likely think they're getting XML input.
Doing this consistently will likely be tricky, but that's also a lot of
why I value making these processes accessible to human intervention on a
> But is XML used by "casual users"? Or is XML more like
> programming languages, where there is arguably no such thing
> as a "casual programmer" (i.e., someone who starts
> programming without any specific knowledge of what they are
> doing) in the way that there is a casual user of a bus (who only
> needs the skill of sitting down).
There's been an astounding amount of casual HTML over the years, much of
it by people I'd never have expected to want to look behind the GUI. I
can't possibly write those people off, even if they're a small minority.
I'd much rather encourage them and encourage their numbers to grow.
Ring around the content, a pocket full of brackets
Errors, errors, all fall down!