Re: [search-ws] Re: ATOM

[Farrukh Najmi <farrukh@wellfleetsoftware.com>]

Ray Denenberg, Library of Congress wrote:
>> I believe we need one mandatory response schema for interop that is
>> guaranteed to be supported by all implementations.
>>     
>
> I'll go along with that, however ...
>
>   
>> Choosing ATOM as that one mandatory response format has many advantages:
>>
>>     * We have less spec content to define and maintain
>>     * We get syndication (feed) capability for free
>>     * Any existing ATOM capable feed reader will work with a URL
>>       representing a search defined by our spec
>>     * ATOM syntax is extensible and therefor we will be able to add
>>       additional attributes to our response without requiring any one to
>>       update ATOM
>>     
>
> Some of these advantages are ATOM specific.  Others, you would get with any
> similar format.  For the ATOM specific advantages (syndication capability,
> for example), instead of saying "Choosing ATOM as that one mandatory
> response format" you could say: "supporting ATOM (in addition to the
> mandatory response format)"  provides these advantages.  Some implementation
> won't care about syndication capability. The ones that do would implement
> ATOM.
>   

But then the question is why reinvent the wheel for our response syntax?

We gain much by adopting the wheel already provided by the ATOM standard.

What do we gain by defining our own response syntax other than save a 
few implementations
some rework.

I am concerned about not doing what is right today because of decisions 
made in SRU several years ago.
This is a major new version and one that is being standardized and 
adopted by many other standards.
Lets make the best decisions based on latest realities.

Thanks.

-- 
Regards,
Farrukh

Web: http://www.wellfleetsoftware.com