[rfc-i] DOIs redux

[johnl at taugh.com (John Levine)]

>So, I think the instructions to authors need to state this. I have a lot
>more faith in the authors' already knowing the DOI for a reference than I
>have in an automated lookup. Certainly, when I look for a reference using
>almost any of the standard tools, the DOI almost always shows up. But there's
>then a human in the loop to validate the result. I suspect that just sticking
>this on a wish list for automation in xml2rfc masks a significant amount of work.

I suspect it'd be more effective to do it the other way, put a DOI in
the reference section, it can pull the requisite info out of the
underlying database.  As others have noted, this is not very hard to do,
an http request to retrieve the data and some munging to turn it from
their flavor of XML into ours.

In the spot checking I've done, the DOI data for the kinds of articles
we're likely to cite, such as ones published by the ACM and IEEE, is
quite good.

R's,
John