I'm not dead yet

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view

I'm not dead yet

Bryan Cholfin
Well, I think I see what's going on a little clearer now, actually. I don't
think I split the problem at quite the right plac the first time. Here we have
some e-text that has been tagged or coded according to SGML rules, and those
codes/tags are represented by ASCII strings (or a particular subset of ASCII)
so that any piece of common software should be able to read and write to files
containing those codes/tags. Now there are two seperate problems after that:

1) Many word processors and editors store their files in proprietary formats
with proprietary formatting codes embedded into them. This presents a problem
to people wishing to share SGML coded files, but who are not necessarly using
the same systems. This is not a new problem, this is an old problem rearing its
head in a new context.  Now it may be that this is not precisely in the realm
of the TEI or SGML, but it would seem that eventually you'd want to get away
from having to worry about compatibility across, say, different word
processors (do you, user of MS-WORD, for example, want to have to have a
filter for everyother word processor in the known universe?), i.e. the
goals of the software companies that give us these things run counter to the
goal of being able to universally share text. Now, all hope is not lost, since
most of these programs can read or write to plain ASCII text files, preserving
the SGML tagging but losing the proprietary formatting information (though,
I would guess that in this context that would be less important).

2) Will the word processors/database handlers/etc. have code built into them
to allow them to use the SGML coding, or will the software just perceive
random ASCII strings? As has been pointed out, in some cases this may
interfere in the normal operation of the word processors (though I would
guess in practice this would not be a major impediment for most). Presumably
it would not require *major* rewrites of software to allow searching, sorting,
etc. routines to dynamically make use of SGML coded structural information,
and I would also expect that in some wp or editing software (if this hasn't
in fact, already occured) that allows for user supplied macros/programs,
that each user could tailor the system to his or her own needs, which is the
whole point of those facilities.

This whole issue may be secondary to the main point of the TEI, but I would
guess that in the long run, if it isn't thought about here, software
developers will be handing out 'solutions' that will hamper the usefullness
of the whole project rather than increase it.

Bryan Cholfin
Broken Mirrors Press