In response to the former question - at present ZNG supports both forms
but which form a particular implementation supports is left to that
implementor. I don't think we've discussed in detail conformance levels.

I *do* want to keep the SOAP requests in the spec. as I do see a need
for them if we expand ZNG to cope with non-textual queries (which is an
area I'm particularly active in) or complex queries (as it is
recommended that a URL not exceed 256 characters, although in our
context I think that this is not a major issue).

As regards returning the response as a SOAP request - I personally think
it less complex if the response format is the same regardless of the
request format. In any case the only overhead here is to add the lines

<SOAP-ENV:Envelope xmlns:zng="urn:z3950:zng_prototype1"
xmlns:SOAP-ENV="http://schemas.xmlsoap.org/soap/envelope/"
SOAP-ENV:encodingStyle="http://schemas.xmlsoap.org/soap/encoding/">
<SOAP-ENV:Body>

to the beginning of the response and

  </SOAP-ENV:Body>
  </SOAP-ENV:Envelope>

to the end. Which I don't see as a big hurdle!

Matthew



> -----Original Message-----
> From: Theo van Veen [mailto:[log in to unmask]]
> Sent: 31 October 2001 09:35
> To: [log in to unmask]
> Subject: Betr.: FW: Re: are there already zng-targets
>
> We have an ZNG-server without SOAP running in a very short time. It
was
> also no problem to write an in a very short time a HTML-page serving
as a
> ZNG-gateway. The root element of our response is
searchRetrieveResponse.
> The use of SOAP makes things more complex and it will certainly result
in
> a higher implementation barrier. Maybe it is a stupid question but I
am
> wondering now what is the added value of SOAP in this standard. I know
it
> makes sense in case on none-ascii input.
> Secondly to keep the implementation barrier low, isn't it a good idea
to
> respond only with a SOAP envelop if the request was also embedded in
SOAP?
>
> The power of simplicity can be of great value to spread the use if
ZNG.
>
> Theo
>
>
> >>> [log in to unmask] 30-10-01 21:54 >>>
> FYI - this is still flakey in places and only handles the URL form at
> present. Currently returns records in Yanks (Redsox ot whatever)
>
> Matthew
>
> > -----Original Message-----
> > From: Antony Corfield
> > Sent: 30 October 2001 17:05
> >
> >
> > Our ZNGServlet can be found at the address below, using the type of
> query
> > shown. Attribute mapping can be specified as bib1.author or
> openURL.author
> > etc. Default is bib1.
> >
> >
>
http://jafer1.las.ox.ac.uk:8080/jaferWeb/ZNG?query=author+%3D+smith+AND+
> ti
> > tle+%3D+fish&startRecord=1&maxRecords=10
> >
> >
> >
> >
> >
______________________________________________________________________
> > Antony Corfield Java Developer JAFER
Project
> >
> > mailto[[log in to unmask]] Libraries Automation
> Service
> > tel: +44 (01865) 284454 99 Banbury Rd
> > fax: +44 (01865) 278275 Oxford OX2 6JX
> >
______________________________________________________________________