Discussion points for tomorrow's call

["Ray Denenberg, Library of Congress" <rden@loc.gov>]

Based on phone discussions with Farrukh and Matthew over the past several
days, I want to propose a few items for consideration during tomorrow's
call.

-------
OpenSearch.
I propose the following strategy. The standard should:

1.explicitly recognize the existing OpenSearch spec, as a legacy spec, and
describe how to interoperate with that spec.

2. In addition, incorporate a new OpenSearch spec, which cuts the cord with
the legacy version.

I propose we keep the current draft annex, make it non-normative, and add a
new normative annex, blank, except for an explanatory note, explaining this
plan.


----------
Rest.
I believe that the only significant issue is the operation parameter.  I
would support consideration of eliminating that parameter, incorporating it
instead in the base url, in some fashion (this needs some discussion).


-------------
Atom/response schema
There is a suggestion to replace the current response schema with Atom.  I
support consideration of this idea.   However, we need to do quite a bit of
analysis. For example, why Atom (instead of RSS)?    I suggest we add a note
in the strawman which describes our thinking on this. Where the current
response schema is detailed, add a note saying "here is the information that
has to be represented in the response schema, but it might not be the actual
schema shown, the actual schema yet to be determined".

--------------
Discussion among SRU Implementors
When we consider anything for this standard that might be disruptive (or
otherwise important) to SRU and current implementations, we need to alert
the SRU list and get them involved in the discussion of that issue.  I want
us to adhere to the following procedure. One of the Chairs (Matthew or I)
will post an initial message to the SRU list. After that, there can be
free-for-all discussion over the SRU list. (Everyone on this committee
should subscribe to that list.) But Matthew or I should initiate that
discussion. The first such discussion initiated will provide context and
descrbe this process.  We will  run draft message by this committee before
posting when appropriate (not necessarily all the time). All of the above
issues would be appropriate for such threads on the SRU list. We will
initiate separate threads for each issue and Matthew and I will try to keep
discussion within a thread focused on the single issue.


--Ray