> Date: Mon, 12 Sep 2005 10:55:17 -0500
> From: Eric Lease Morgan <[log in to unmask]>
> 
> > http://curtis.med.yale.edu/dchud/log/project/linkstack/more-unalog- 
> > search-with-opensearch-and-content-indexing
> 
> The three bullet points at the bottom of the blog summarize the  
> author's experience:
> 
>    * The mess that is the SRW/U specs slows adoption
>    * The lack of an OAI-PMH-esque testing tool makes "compliance"
>      difficult to test
>    * The absence of a public-facing a9-style SRW/U clearinghouse
>      limits its potential impact significantly
> 
> These points echo tasks the group articulated as "next steps" in  
> regards to adoption. See:
> 
>    http://www.loc.gov/z3950/agency/zing/forum-june05-output/steps.html

Yes.  The infuriating thing is that there is _nothing_ hard about any
of this.  Given the funding it would take maybe two months tops to
resolve all three of these problems -- nailed.  But because (unlike
OpenSearch) no single big company is sufficiently committed to SRU to
make it happen, we still struggle.

For what it's worth, I thought the blog was very good; spot on, in
fact.  It hurt to read, but stung worse because I felt its justice.
What hurt most was that the superficial problems with SRU completely
hid its order-of-magnitude technical superiority from the blogger.

 _/|_	 ___________________________________________________________________
/o ) \/  Mike Taylor  <[log in to unmask]>  http://www.miketaylor.org.uk
)_v__/\  "Not raw -- cooked" -- Monty Python's Flying Circus.