[rfc-i] URIs in references, was: Call for Review of draft-iab-styleguide-01.txt, "RFC Style Guide"

"Martin J. Dürst" duerst at it.aoyama.ac.jp
Wed Mar 26 18:57:00 PDT 2014


On 2014/03/27 06:39, Dave Crocker wrote:

> I'll express strong objection to leaving the issue to authors.
>
> Possibly trust the stream editor, but individual authors cannot be
> expected to have enough background or discrimination to judge this issue
> reliably.  And for documents to be useful 20 years hence, this issue
> needs to be handled well.

For a document to be useful 20 years hence (apart from its usefulness 
for historic studies), it has to first and foremost contain content that 
is useful for 20 years. If we look at RFCs from 20 years ago, then some 
of them are still useful in that sense, while others are not.

Also, the importance of supportive content (incl. references) tends to 
change over time. Often documents contain supportive content that was 
deemed important at the point of the creation of the document, but this 
importance may change over time. After 20 years, e.g. the reason why a 
particular design was chosen may be much less important.

If a document is still useful 20 years hence (e.g. because the protocol 
it describes is still in use), then I'd guess that the references it 
contains either are no longer needed (e.g. because the design has shown 
its value in practice) or if by bad luck they disappear from the URI 
cited, and they still are of value, then somebody somewhere went to the 
pain to publish them somewhere else, and it will be found easily again 
with a search engine. We can also publish an erratum; for an actual 
example, please see
http://www.rfc-editor.org/errata_search.php?rfc=6055.

This is not to say that we shouldn't work hard to get the "best, most 
stable references possible", but we should be aware that we are not 
living in a perfect world, nor in a world that needs to be perfect.

Regards,   Martin.


More information about the rfc-interest mailing list