On Tue, Aug 26, 2008 at 04:04:02PM -0700, SM <sm@xxxxxxxxxxxx> wrote a message of 37 lines which said: > Anyone can write tools without undue cost to read, parse, search and > produce documents in that format. Read, search and produce, OK. Parse, no, unless you're joking. Parsing RFCs is incredibly difficult. If you doubt it, please write a Perl/Python/Javascript/Lisp/whatever program to extract from a random RFC: * author(s) * title * list of sections * normatively referenced fellow RFCs and post it here. Of course, PDF is even worse in that respect and so I would not support the idea of switching to PDF. What I wanted to emphasize is that the biggest problem with plain text (wether it is ASCII or RFC 5198 is irrelevant) is the extreme difficulty of parsing and therefore of extracting metadata. RFC 2629 rules :-) _______________________________________________ Ietf@xxxxxxxx https://www.ietf.org/mailman/listinfo/ietf