[rfc-i] URL checking
wayne at schlitt.net
Wed Oct 19 09:01:48 PDT 2005
In <43564C52.1050403 at cs.utk.edu> Keith Moore <moore at cs.utk.edu> writes:
>>>not clear at all. typing "Keith Moore" into google gives different results
>>>from one month to the next, whereas typing http://www.cs.utk.edu/~moore
>>>has given consistent results for several years.
>> I said "largely". We're talking about citations in RFCs; IMO, they
>> should be specific enough anyway, i.e., "Keith Moore UTK", which would
>> probably work just fine. If not, a better citation is in order (Keith
>> Moore University of Kentucky), rather than necessarily needing a URL.
> I think you have it backwards. The URL is a precise reference. If it
> resolves to a document, it's highly likely to be the correct document.
More over, things like the wayback machine are often quite effective
at retrieving documents when the URL no longer works. When documents
have been relocated and the old URL no longer works, there are often a
lot of componenets in the old URL that help you find the new
location. e.g. ftp://cs.utk.edu/pub/moore/bulk_mailer/bulk_mailer-1.13.tar.gz
Doing a google search on "bulk mailer" will lead you to very different
stuff. Without the original URL, I wouldn't be as confident that
the new URL is what I want.
> The citation examples you are giving are less precise. For example, if
> one is referring to a document, title and authors are not sufficient, as
> it's not unusual for authors to revise a document or rewrite it for a
> different audience but using the same title. There's a reason that
> print references are generally expected to have precise dates and/or
> volume and page numbers.
Right, and in the above example, I can see that the tarball is a
different version and I should look at the ChangeLog to see if it is
still relevant, or to use the new URL to help find the older version.
A working URL may not work forever, and I don't think *just* a URL
should be given, but the URL is often one of the most useful things.
More information about the rfc-interest