URI fragments in IVOA specifications

[Norman Gray]

Rob, hello.

On 2012 Apr 4, at 19:13, Rob Seaman wrote:

> Typo in section 1.1: "fining" -> "finding".

Thanks!

> Since you emphasize the limited shelf life of URIs (which might have seemed to be the whole point of URIs), perhaps we should consider some other kind of permanent identifiers for VOEvent notifications.  In a strange fashion a report of a transient phenomenon has a permanency that many other archival holdings may not.  Tycho's Stella Nova has the stature of an historical document, as to a lesser degree do GCN, ATel, CBAT and MPC notices.  The refereed literature already cites IVORNs.  If there is an issue we should address it.

I think they have a finite life, but wouldn't want to commit myself to anywhere more precisely than 10^(1--3) years.  I do believe they are for all practical purposes long-term things; indeed they're such important things that when they are finally replaced on the transgalactic interinternet, the successor technology _will_ have a mapping from them.  That's a type of very secure permanence.

This is a good opportunity to note in this forum that (contrary to the impression which the prescriptive language might give) I'm not predicting imminent IVOA disaster.  It's simply better to go with the grain of a technology than against it, and the three issues are intended to demonstrate that this is not a purely theoretical or aesthetic objection.

All the best,

Norman


-- 
Norman Gray  :  http://nxg.me.uk
SUPA School of Physics and Astronomy, University of Glasgow, UK