[rfc-i] [Tools-discuss] RFC Index Files Updates -- XML and HTML

Brian E Carpenter brian.e.carpenter at gmail.com
Tue Aug 18 15:28:25 PDT 2009


I agree re machine parsing. My concern is that when human eyes look
at the txt file, they won't find the stream name. And for some purposes,
plain text is still more convenient than *ML. I certainly see Bob's
point about compatibility with pre-1989 awk scripts though ;-)

BTW, while we're on the subject, does anyone here maintain a BibTeX
version of the RFC index?


On 2009-08-19 06:04, Fred Baker wrote:
> I would agree that people that are parsing rfc-index.txt should probably
> be using the XML version.
> On Aug 18, 2009, at 8:24 AM, Bob Braden wrote:
>> Brian E Carpenter wrote:
>>> Hi,
>>> Thanks for this.
>>> Is there any reason not to add the stream info to the .txt version
>>> of the index?
>>> Regards
>>>   Brian
>> Brian,
>> This is the result of a 10-year old policy at the RFC Editor. After
>> Joyce and I took over the job, the desire arose to add additional
>> fields to rfc-index.txt. However, input from various people in the
>> community convinced us that a fair number of hard-core Internauts had
>> written their own code to parse the obscure syntax of rfc-index.txt
>> and reformat it for their own purposes.  We feared that adding new
>> fields to rfc-index.txt would screw these people.  We therefore
>> decided to freeze the format of rfc-index.txt at the 1989 level, and
>> instead make an html version (and later an xml version) with extended
>> fields.  So all changes or extensions to the index format go only into
>> the .xml and .html versions.
>> Now that XML has taken over the world, maybe this policy should be
>> revisited.
>> Bob Braden
>> _______________________________________________
>> Tools-discuss mailing list
>> Tools-discuss at ietf.org
>> https://www.ietf.org/mailman/listinfo/tools-discuss

More information about the rfc-interest mailing list