Re: Proposal to close ISSUE-67: Full container membership without pagination, saying no.

On 28 May 2013, at 19:51, John Arwe <johnarwe@us.ibm.com> wrote:

> > a priori ok with that. I'll try to implement pagination very soon so
> > that I can make sure practically I agree. 
> 
> In this case, I'm not sure what implementation experience adds. 

Just a better understanding of the problem really. I was just pointing out that I had not looked at
this in very great detail, which making an implementation forces you to do.

> 
> You can't force a server to serve a representation of every resource it could possibly store (or calculate), in the general case.  Even if the spec (and code) tried, in the "too large" cases you'd just end up with eternal 5xx responses.  "Eternal" not because there is anything wrong with the request, but simply because the server implementation does not scale as large as it requires.
> 
> In the case where a server recognizes that it cannot process the request, such that it would *like* to redirect to a paged equivalent, if you remove the paged option the server is no more able to fulfill the request than it was before.  You just force it to try (if the implementation values compliance above all else), but you have no guarantee it will succeed.
> 
> 
> 
> Best Regards, John
> 
> Voice US 845-435-9470  BluePages 
> Tivoli OSLC Lead - Show me the Scenario 
> 
> 

Social Web Architect
http://bblfish.net/