Print

Print


On Fri, 2010-10-22 at 11:09 +0100, Edward C. Zimmermann wrote:
> On Fri, 22 Oct 2010 09:53:19 +0100, John Harrison wrote
> 
> > 
> > There would seem to be a clear use-case for displayTerm, although I
> > confess I've never figured out how to implement it tidily in our search
> > engine.
> 
> I would like to suggest that depreciation does not mean that the
> "functionality" would go away but could still live on in an optional
> <extraTermData> which can contain in addition to an element for alternative
> display

Yes, I got that.

> > 
> > At first I was horrified by the "anything goes" proposal for the search
> > term value, but then I read on a little further and came round to the
> > idea. However I would tighten up the definition a little to something
> > more like:
> > 
> > "A term that when used in a search of the same index, produces an
> > appropriate search response, consisting of 1 or more record(s). 
> > Where a numberOfRecords is also supplied by the scan response, the subsequent
> > search should return the specified number of records."
> 
> Was that not the implicit intent of the "exactly as it appears in the index"
> following an assumption that terms in an index are terms in records that can
> be found whence a term "exactly as it appears in the index" will return 1 or
> more records?

Yes, I guess that was the implicit intent. What I'd tried to do was to
make the reasoning explicit, while removing the requirement that the
term value had to be "exactly as it appears in the index", to cover the
examples already given on the list, and allow your anything goes
approach. However...

> Technically being in the index does not guarantee that a search response will
> return any records but we've come to assume that these terms when used as a
> search term would produce an appropriate search response consisting of 1 or
> more records--- which btw. a number of systems don't always deliver (and since
> my engine has a facility for search time stop words I can well see use cases
> of my own engine that don't deliver). Effectively its been "what the server
> that supplied the term deems appropriate" and we've had wishing thinking that
> the server would deem appropriate a search response, consisting of 1 or more
> record(s). I'd like to leave this ambiguity in.

... having spent a little more time thinking about this, I've remembered
that one of the services I've written would violate this on occasion,
through no fault of its own. So I would actually like to keep the
ambiguity in as well :) I guess this might also affect the wording of
the definition of numberOfRecords...?

All the best,
John


> I think the important issue here is that the term is supplied by the server
> and the server provides a search response as it intended. It might be the
> desire of the operator of the search server to provide service that fulfill
> their customers' expectations but these issues are "tar pits" outside the
> realm of protocols. 


> 
> > 
> > IMO this definition should also apply to term in facets, due to their
> > similarities in use.
> 
> Yes.
> 
> 
> > 
> > All the best,
> > John
> > 
> > -- 
> >               '.    ,'.         John Harrison
> >              '  `  '  '         University of Liverpool
> >  c h e s h i r e  |  3          e: [log in to unmask]
> >                   v             w: www.cheshire3.org
> >               `-..;.'           t: 0151 7954271
> >                 ..,     (c)
> > 
> > On Thu, 2010-10-21 at 21:04 +0100, Ray Denenberg wrote:
> > > The utility of displayTerm (vs. actualTerm) was illustrated  many years ago
> > > during early implementation of Z39.50 browse/scan.  You'd have to look up
> > > that discussion in the archive, but basically, an implementation had an
> > > index where terms where not very user friendly but were much more efficient
> > > for searching than their corresponding displayTerms. I don't recall who the
> > > implementer was, but there was a real implementation.
> > > 
> > > Of course this was for scan (or "browse" as we called it in Z39.50), not for
> > > facets.
> > > 
> > > My position on this is that scan and facets should be aligned in this
> > > matter. That is, they should both have displayTerm or neither should. That
> > > means the OASIS committee should consider depricating displayTerm from Scan.
> > > I don't say it SHOULD DEPRICATE it, it should consult with Z39.50 and SRU
> > > implementors to see if displayTerm is still necessary/useful. That was the
> > > purpose of Ralph's posting to the SRU list. If it seems that displayTerm
> > > should remain in scan then I think it should be added to the facet response.
> > > If it seems nobody cares whether it is retained or not then it should be
> > > dropped from scan 2.0 (and not added to the facet response).
> > > 
> > > --Ray
> > > 
> > > -----Original Message-----
> > > From: SRU (Search and Retrieve Via URL) Implementors
> > > [mailto:[log in to unmask]] On Behalf Of LeVan,Ralph
> > > Sent: Thursday, October 21, 2010 3:07 PM
> > > To: [log in to unmask]
> > > Subject: Re: <displayTerm> as a subelement of <term> in Scan responses
> > > 
> > > Pretty much all the conversation on that list has been about changes to the
> > > facet response and the only controversial change is my desire to add a
> > > displayTerm.
> > > 
> > > Here's a pointer to the achive for this month:
> > > http://lists.oasis-open.org/archives/search-ws-comment/201010/threads.ht
> > > ml
> > > 
> > > Ralph
> > > 
> > > > -----Original Message-----
> > > > From: SRU (Search and Retrieve Via URL) Implementors 
> > > > [mailto:[log in to unmask]] On Behalf Of Jonathan Rochkind
> > > > Sent: Thursday, October 21, 2010 1:50 PM
> > > > To: [log in to unmask]
> > > > Subject: Re: <displayTerm> as a subelement of <term> in Scan responses
> > > > 
> > > > Can you link to the thread in the archives of the list, or is it (like 
> > > > most of our lists, for no good reason) private?
> > > > 
> > > > I am curious what the arguments against it are. It seems like a good 
> > > > idea to me?
> > > > 
> > > > But I do not use Scan at all.
> > > > 
> > > > I do not use facetting through SRU at all, but if/when I do, I believe 
> > > > I'd use displayTerm if it was there.
> > > > 
> > > > LeVan,Ralph wrote:
> > > > > There's an overlong debate going on in the search-ws-comment mail
> > > list
> > > > ([log in to unmask]<mailto:search-ws-
> > > > [log in to unmask]>) about including a displayTerm as a 
> > > > subelement of <term> in a facet response.  The feelings against
> > > including
> > > > displayTerm are so strong as to suggest that it should be deprecated
> > > in Scan.
> > > > >
> > > > > I've been the one arguing for the inclusion of displayTerm for
> > > consistency with
> > > > Scan, but I don't use it myself in any of my implementations.  So, my
> > > question for
> > > > you all is: do any of you actually use the displayTerm in your
> > > responses?  If not,
> > > > I'll happily drop my arguments.  If so, can you provide a good use
> > > case?
> > > > >
> > > > > Thanks!
> > > > >
> > > > > Ralph
> > > > >
> > > > >
> 
> 
> --
> 
> Edward C. Zimmermann, NONMONOTONIC LAB
> Basis Systeme netzwerk, Munich Ges. des buergerl. Rechts
> Office Leo (R&D):
>   Leopoldstrasse 53-55, D-80802 Munich,
>   Federal Republic of Germany
> http://www.nonmonotonic.net
> Umsatz-St-ID: DE130492967
-- 
John Harrison <[log in to unmask]>
University of Liverpool