![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
![]() |
The way I think about it, the current "ascii" rule makes it hard for the writer, moderately easy for the reader, and very easy for the archivist. It is hard for the writer now, because we are mostly using wysiwig tools that don't produce ascii very naturally. OTOH, it is not too hard -- for example, the extra cost of postprocessing a Microsoft Words formatted document to an internet-draft format is at most a couple of minutes, which seems a reasonable investment when you write for the posterity. As a matter of fact, it does not take longer than the nroff postprocessing that we used when nroff was the norm. Ascii text is moderately easy for the reader. Let's face it, the 72 character per line format has its roots in a time of 24x80 character screens, which itself had its root in the generic 80 columns punch card. That is fine when a screen is large enough to display a full line of text, but that is definitely suboptimal on small screens. A marked-up presentation would allow for better rendition on small PDA screens, and would also allow for a better support of those of us who have a poor eyesight. OTOH, the RFC format is very regular, with fixed size pages, regular indentation, etc. Many people have already written filters that convert this format into their preferred mark-up language, which is a reasonable compromise for viewers. We should note that the "ease of view" argument is an argument for mark-up languages, not for fancy page formatting languages such as Adobe's PDF: if you cannot reformat the page on display, there is no particular benefit for the owners of small screens or weak eyes. But the very benefits that make make marked-up presentation tempting also carry a big risk in a standard world: if the documents you view depend on how you view them, there is a risk that conficting view generate conficting interpretation. That risk is well known in international circles, when for example the translation of various UN resolutions can carry slightly different meanings in different languages. If we want standards, we want a simple reference. Printing an ascii text with a fixed font and a fixed page set-up gives us that. In my opinion, the hardship imposed on the authors and on the present readers is more than compensated by the ease of archival of the text format, the ease of read by future readers, and the adaptation to the job at hand, publishing standards. Mark-up languages have evolved in time, from nroff to tex and latex, from wordperfect to office XP, from sgml to html and xml. Markup languages tend to be highly parametrized, with various nroff or tex macros, various word templates, various html extensions and xml dtds. The macros tend to have an even shorter lifespan that the languages themselves, which make them even poorer candidates for an archival function. Frankly, the energy spent every fice years in questioning the ascii format would be better spent in writing transition tools... -- Christian Huitema
Note Well: Messages sent to this mailing list are the opinions of the senders and do not imply endorsement by the IETF.