[rfc-i] Does the canonical RFC format need to be "readable" by developers and others?

Iljitsch van Beijnum iljitsch at muada.com
Fri Jun 22 01:42:54 PDT 2012

On 22 Jun 2012, at 3:26 , John R Levine wrote:

> I've now asked several times for an example of a way that an xml2rfc document might be broken, and the problem wouldn't be evident in the html or text versions, with no replies at all.  

When I started using XML2RFC some years ago I had big problems figuring out how to do certain stuff, such as references. The documentation was very incomplete at that time, so basically I was relegated to copying code from other documents without knowing what that code did and adjust it to my needs through trial and error.

Apart from that, it demands excessive amounts of tagging for things like even the author's name, then tries to be smart in composing a full name from the parts but doesn't allow for names that don't use standard English capitalization. So it makes me do more work for an inferior result (but that part is my own little annoyance). When adding a reference to something lesser known (not an RFC or draft) you have to encode enormous amounts of information that never make it into the text, like even the entire abstract of the document referred to. This is a waste of time, authors have better things to do.

Last but not least, humans forget to close tags once in a while. Debugging that is utter hell, because the tool doesn't help you find the possible location where this happened at all.

So in my experience, XML2RFC is just barely better than creating the RFC format by hand, but it is a constant annoyance and time waster on many levels.

More information about the rfc-interest mailing list