| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 372.1 | | AUTHOR::WELLCOME | Steve | Fri May 15 1987 10:38 | 1 |
| TECO could probably do it; do you know any TECO wizards?
|
| 372.2 | �Who is fluent in TPU? | MUNSBE::RUEDEL | Wilfried R�del, Munich, 757-0226 | Fri May 15 1987 14:31 | 2 |
| No, but I would have thought that there is a nice TPU procedure
around somwhere. So, if someone knows of something like that...
|
| 372.3 | Possibility | BUNSUP::LITTLE | Todd Little NJCD SWS 323-4475 | Fri May 15 1987 15:37 | 6 |
| In the note on DECspell dictionaries in this conference I posted a trivial
SCAN program that will locate all tags in a file and extract just the
tags and place them in another file. It loses all the arguments, etc. so
I'm not sure how helpful it would be for your translation needs.
-tl
|
| 372.4 | translation tools | VAXUUM::KOHLBRENNER | | Mon May 18 1987 10:24 | 28 |
| Extracting tags from a file is a non-trivial task.
Tags have arguments and the arguments can contain other tags.
Sometimes the arguments contain text that would need to be
translated, other times the arguments contain keywords that
probably should not be translated.
The <comment> tag and the <literal> tags come in two formats,
and they may or may not contain other tags in the text that
they contain. Is this text, with or without its tags, intended
for translation?
Separating the tags from the text seems to me only half the problem,
if there is an intention to later merge the tags back into the
translated text. How will the merge be accomplished? What are
the "markers" that tell how to put the two pieces back together?
Won't some tags disappear as part of the translation? The writer
of English may want to add emphasis to a word or phrase, using the
<emphasis> tag. The translator may find it easy to convey the
emphasis in the target language without bolding or italics.
Computerized aids for doing the translation sounds like a worthwhile
project, but it seems that it will require more than a simple TPU
procedure to be very useful...
bill
|
| 372.5 | separation of syntax and semantics (form and functions?) | ATLAST::BOUKNIGHT | Everything has an outline | Mon May 18 1987 11:26 | 8 |
| What it should really have access to is the exact same parser used
in GUTENTAG. With the work to be done in adopting SGML, maybe the
DOCUMENT folks could consider breaking the front end up into separate
syntax and sematics handling code, making the syntax handing code
available to other users such as translation aids, a "pretty" formatting
program for SGML, etc.
jack
|