On Jan 5, 2006, at 11:31 AM, John Levine wrote:
Quite frankly, I believe we can address the second step (which of
the requirements are not met today?) with a firm "none."
One is that ASCII doesn't permit adequately beautiful pictures. If
that's the problem to be solved, it seems to me that a
straightforward solution would be to allow authors to submit image
files along with the ASCII text. I'd suggest requiring that the
image format be GIF, since it's simple, stable, well documented,
widely supported in both freeware and commercial software, and the
patents have expired. (Or maybe PNG, any stable public format will
do.)
A minor problem with independent graphic files is they are difficult
to manage. A graphic image has also become a vehicle for Trojans, as
file extensions often do not take precedence within rendering
engines. This would impose a new risk for the editor. An interim
solution could be a drawing application (or a regular editor) that
uses Unicode "Box Drawing" characters rather than ACSII hyphens,
under-score, plus symbols, greater-than, less-than, and vertical
bars. For these to provide a clean output, the line spacing would
need to be controlled for a clean look. This can be done using HTML
and PDF outputs.
The other is that the editorial process is more tedious than it
needs to be, because RFCs have a mandatory structure that plain
ASCII doesn't express. RFC 2629 or 2629bis captures the structure
while being well supported in free and commercial software and, in
a pinch, legible without tools since it's ASCII underneath.
These tools can already utilize the "Box Drawing" characters, so a
utillity to assist with the creation of the box drawings would make
this less painful. The utility could also add the needed XML wrapper
to make this an easy cut and paste. If desired, these characters
could be translated back into the ASCII equivalent for the ASCII
version. It would also seem that bibliography and author names could
also include a Unicode element used by the HTML and PDF outputs, but
then not the ASCII versions. The alternative elements for titles and
authors would be assuming a desire to retain the ASCII only version.
If the ASCII version is not retained, then the effort would be even
more straight forward.
-Doug
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf