bibtex-1.0, or maybe not

Norman Gray gray at nxg.name
Thu May 23 17:03:15 CEST 2024


Greetings.

On 22 May 2024, at 5:35, Schneider, Thomas (NIH/NCI) [E] via texhax wrote:

[quoting Paulo Ney]

>> The problem Thomas is that URL is ephemera.
>
> I agree!!  (You can call me Tom.)

I do not!  Or not necessarily.

>> If you have a DOI, you have something, if you don't --
>> there is not a point in even using it. It may disappear before that
>> guy that printed the article tries to get to it.
>
> Fortunately the biomedical scientific literature
> (https://pubmed.ncbi.nlm.nih.gov/) now has DOIs for almost everthing
> and so I use those when I can.  For readers of my papers it's very
> convenient - they just click through the DOI to read the reference.
> For example, https://doi.org/10.1371/journal.pone.0279758

I think this is too binary a conclusion to come to.  URLs are generally useful in bibliographic references -- yes, in print as well as online -- but it might need some thought about which URL to use.

A DOI such as 10.1371/journal.pone.0279758 is not a URL (it's a Handle ]1]).  However the DOI-URL formed from that in the standard way, as above, is an example of a URL that one can reasonably rely on as a persistent one.

A URL which points to a random wordpress site, or one which points into a black-box CMS (ie ...?blearghhhh) is one where I'd be reluctant to rely on persistence.

In the middle are URLs such as pubmed ones, or arxiv.org ones, or loc.gov ones, or purl.org ones, or... a fairly wide variety of others where there is a clear 'institutional commitment to persistence', in the standard phrase.  But there's nothing about a URI that syntactically indicates this commitment: it's a matter of judgement on the part of the person recording the bibliographic detail.

Thus I think it would be a mistake to in any way deprecate the use of URLs in references, printed or not.  If a moderately stable URL exists for a resource, then it's an act of discourtesy to the reader to fail to include that in the reference, since all it obliges them to do is to spend time searching for that URL themself.

Best wishes,

Norman


[1] https://en.wikipedia.org/wiki/Handle_System

-- 
Norman Gray  :  https://nxg.me.uk



More information about the texhax mailing list.