Re: LANGUAGE/COMPARATOR review

[Arnt Gulbrandsen <arnt@xxxxxxxx>]

Cyrus Daboo writes:
> Lets keep this simple and go with (2) as that does not require the 
> server to remember more state than it already does. And in all 
> likelihood clients will issue LANGUAGE before doing NAMESPACE anyway.

I changed it in a different way, equally simple. No server state.

Ken's suggestion to mandate NAMESPACE made immediate sense to me. 
Specifying what happens if LANGUAGE is supported and NAMESPACE isn't, 
or if the client hasn't issued NAMESPACE, or, or, was too hard and 
matters in too few cases to be worth much prose. So I said: Client must 
support NAMESPACE (at least enough to ignore the NAMESPACE response) 
and server always sends the response if it has translations. If it 
doesn't have translations - fine.

Thank you very much. This helped.

Arnt