On Jul 2, 2009, at 9:22 AM, Marshall Eubanks wrote:
On Jul 2, 2009, at 12:16 PM, Ted Hardie wrote:
At 10:19 PM -0700 7/1/09, Douglas Otis wrote:
for wanting more than just plain text documents is to permit
inclusion of charts, graphs, and tables, for a visual society
It seems to me that where this discussion has faltered before is on
whether this is, in fact, a requirement.
You are exactly correct, and I can recall several interminable
discussions of this.
To save time, I would suggest adopting the Patent Office rules on
Perpetual Motion. People advocating for a change to facilitate
figures (or to allow complicated math, such as tensor analysis)
should have an existence proof, i.e., a document that requires the
change to be published. (A document that left the IETF to be
published elsewhere for this reason would also do.)
What appears to be missed in these conversations represents a
dissatisfaction of the generation tools and output quality, which is
easily shared. There is good reason to avoid closed source generation
tools, however the IETF has already employed and permitted the use of
roff inputs and outputs, which appears to offer a reasonable means to
satisfy the many requirements already in place.
A suggestion to use Word XML outputs as a means of providing WISIWYG
operation misses what is currently in place within xml2rfc needed to
generate tables, state diagrams, and graphs. Yes, these elements are
_currently_ contained within existing RFCs, but in ASCII form. Even
though these elements are structured using ASCII, textual processing
must still accommodate special handling of these clumsy visual elements.
Although I am not blind, the simple instructions required by roff
tools should allow those visually impaired a superior means for
understanding the intent of visual graphics, rather than guessing what
a series of white-space and characters are attempting to convey within
diagrams or equations.
In addition, there are currently several RFCs already created using
roff, as were my first attempts at writing I-Ds. Due to IETF's
current level of support for xml2rfc, this mode of input now offers an
easier means to generate acceptable output. IMHO, roff tools can
still offer higher quality output that is more compatible with various
presentation media than outputs generated from xml2rfc.
Perhaps the IETF may wish to better retain the older roff methods by
offering better boilerplate and processing support for this currently
acceptable method for generating I-Ds and RFCs. A wiki style web-page
with IETF custom roff pre-preprocessors could facilitate roff inputs
for the creation of ID and RFC documents. The availability of roff to
html output should also make creating previews as a type of iterative
WISIWYG mode of creation possible. This would be no different than
the steps used with xml2rfc.
IIRC, .ps generated from roff tools are still acceptable inputs as
well, although I expect the current publishing automation is likely to
balk at output from these older methods. Too bad though.
-Doug
_______________________________________________
Ietf@xxxxxxxx
https://www.ietf.org/mailman/listinfo/ietf