Date:         Thu, 18 Dec 2003 13:47:18 +0000
Reply-To:     "Z39.50 Next-Generation Initiative" <[log in to unmask]>
Sender:       "Z39.50 Next-Generation Initiative" <[log in to unmask]>
From:         Robert Sanderson <[log in to unmask]>
Subject:      Re: versioning for context sets and record schemas
Comments: To: "Z39.50 Next-Generation Initiative" <[log in to unmask]>
Comments: cc: [log in to unmask]
In-Reply-To:  <[log in to unmask]>
Content-Type: TEXT/PLAIN; charset=US-ASCII

> > I am half inclined to write this off as "out of band agreement".  I
> > don't think any of us wants to see more new syntax in CQL to have
> > queries specify what CQL version they're using.
>
> So let's leave aside how the version is indicated when cql is used outside
> of srw, assume it's not in the cql string, assume further that cql will be
> wrapped in something, and that it's that something's responsibility. So when
> that something is srw, that leaves us with the question of how srw specifies
> the cql version.

My preference would be to have a mandated CQL version for every SRW
version.  They don't need to always increment at the same time (and with
any luck, CQL 1.1 has enough extensability to never need to increment
again!) but SRW can only change CQL versions when it goes up a version
itself.
So we may have minor versions where the only change is to update the CQL
version, but I don't see that as a problem.


> > for the SRW case, the CQL version can be disovered from the ZeeRex
> > record.
> Is that sufficient?  The client would discover what? The highest version

Here's the scenarios as they play out in my head:

1.  In SRW version 12.9, we're using CQL version 4.1.  Someone sends a CQL
query which only complies with version 1.1.  The server doesn't know which
version of CQL it is, and just fails the query, even though it could
(potentially) process it.

 - Actually, I just don't care.  CQL is never going to change so
dramatically that updating will take more than an hour or three of work.
Supporting profiles and context sets is going to take MUCH longer than
supporting CQL versions.   Especially with several CQL parsers freely
available in a variety of languages, either to plug in or use as
reference.

2.  An old SRW server gets a new CQL query and can't process it.

 - This is handled already in the versioning system, if we tie CQL
version used to the SRW version as above.

3. Another protocol ...

 - Can solve it another way if it wants to.

Rob

--
      ,'/:.          Dr Robert Sanderson ([log in to unmask])
    ,'-/::::.        http://www.o-r-g.org/~azaroth/
  ,'--/::(@)::.      Special Collections and Archives, extension 3142
,'---/::::::::::.    Nebmedes:  http://nebmedes.o-r-g.org:8000/
____/:::::::::::::.
I L L U M I N A T I