Date:         Thu, 27 Jun 2002 10:09:11 -0400
Reply-To:     "Z39.50 Next-Generation Initiative" <[log in to unmask]>
Sender:       "Z39.50 Next-Generation Initiative" <[log in to unmask]>
From:         Ray Denenberg <[log in to unmask]>
Organization: Library Of Congress
Subject:      sort: well-known keys
Comments: To: "Z39.50 Next-Generation Initiative" <[log in to unmask]>
Content-Type: text/plain; charset=us-ascii

Robert Sanderson wrote:

> > > I'll support his alternative proposal.
> > I don't completely understand the proposal. I understand the part about loading
> > the sort-key names with the auxiliary parameters, but what are the names?
> > Well-known sort-key names?
>
> Up to the server to support or not as it chooses. Eg a server can
> interpret a request to sort by 'Title' in any way it choses, but you would
> expect that if there is a title field or index then it would sort by that.

Sorry I'm still confused. Yes of course it's up to the server to interpret and
support or not-support the sort key. My question is are we going to distinguish a
list of well-known sort keys?  A client might ask to sort on author and the server
might understand "authour" but not "author".  So if we just say the client can supply
any sort key it likes and hope the server understands, I don't think that's going to
work well.

I'm further confused by the discussion, on the one hand, to load sort parameters into
the sort-keys, resulting in potentially highly structures sort keys,  and on the
other hand to just use magic keys.

--Ray