I think we would be talking about database identifiers rather than human
readable database names - so this may not be an issue.

I had thought that the last time we discussed this the proposal was

A) to have the database identified as part of the query, rather than the
URL (as discussed earlier in the week) - i.e. treat it as a sub
collection
B) add some meta-record meta-metadata such as which database it came
from.

Or is this oversimplifying?

Matthew

> -----Original Message-----
> From: Z39.50 Next-Generation Initiative [mailto:[log in to unmask]]
> On Behalf Of Eliot Christian
> Sent: Thursday, June 16, 2005 10:30 PM
> To: [log in to unmask]
> Subject: Re: Betr.: Re: Multiple Sources in SRU/W
>
> At 11:34 AM 6/16/2005, Mike Taylor wrote:
> >[...]
> >The obvious way to do this would be to allow the "database name" part
> >of the base-URL to be a separated list of database names rather than
> >just a single name. So:
> >
> >
> http://z3950.loc.gov:7090/voyager+oldrecords+mydb?version=1.1&
> > operation=searchRetrieve&query=dinosaur
>
> I'm afraid of this because I don't understand the implications. I do
> recall being nervous about URL stuff after reading the URL RFC over
> a year ago (see ftp://ftp.rfc-editor.org/in-notes/rfc3875.txt )
>
> FWIW, here are some "database names from Hell", selected from the
> registry of Z39.50 servers (yes, each "dbname" has multiple words)
>
> Robert Cliche (http://registry.gsdi.org/viewrecord.php?rec=687)
> SCI Program Metadata (http://registry.gsdi.org/viewrecord.php?rec=674)
> Bowen Island (http://registry.gsdi.org/viewrecord.php?rec=684)
> NATIONAL WETLANDS INVENTORY
> (http://registry.gsdi.org/viewrecord.php?rec=12)
> RADARSAT DATA OVER CANADA
> (http://registry.gsdi.org/viewrecord.php?rec=739)
> SPOT HIGH RESOLUTION VISIBLE IMAGERY - CANADA
> (http://registry.gsdi.org/viewrecord.php?rec=717)
> LANDSAT MSS SCENES COLLECTED BY CCRS
> (http://registry.gsdi.org/viewrecord.php?rec=690)
>