Ray Denenberg writes:
> Getting back to this.....
>
> Where we left off:
> We want to define an extension, whose name would look something like:
> x-info-99-metadata
>
> So the parameter in the SRU request would look something like:
> &x-info-99-metadata=[some value]
>
> (We have ruled out the more elegant but also more shaky alternative of
> omitting the value altogether, i.e. "&x-info-99-metadata".)
>
> One suggestion was:
> &x-info-99-metadata=true
Yes.
> I suggest instead that we maintain a list of reserved values.
>
> So the parameter could look like:
>
> &x-info-99-metadata=rec
> or
> &x-info-99-metadata=dc
> etc.
>
> How does that sound?
It sounds the same as it did the last time we went round this subject
:-) overly complicated, and a bar to interoperability since servers
will have to provide all listed metadata formats in order to work with
an arbitrary client and clients will have to understand all listed
metadata formats in order to work with an arbitrary server. For all
the same reasons as previously, I advocate One True Metadata Format
which this extensions mandates. Clients that really and truly want to
retrieve metadata in some other format for reasons of their own will
still be able (as now) to do so using a simple searchRetrievel with
the appropriate value of the recordSchema parameters.
So I think the request parameter should be:
x-info-99-metadata=true
or
x-info-99-metadata=TRUE
or
x-info-99-metadata=t
or
x-info-99-metadata=yes
or
x-info-99-metadata=1
or something similar. I am guessing that there must by now be a
convention for representing "true" as a URI query parameter value --
does anyone know where it is (and, better still, where it's
documented?)
_/|_ ___________________________________________________________________
/o ) \/ Mike Taylor <[log in to unmask]> http://www.miketaylor.org.uk
)_v__/\ "Me, I'm just a lawnmower: you can tell me by the way I walk"
-- Genesis, "I Know What I Like"
|