Re: [provision] spec issues.

[Gary P Cole <Gary.P.Cole@Sun.COM>]

Darran Rolls wrote:

> 2.1 - UML diagram
> I'll take over this from Gerry 

Thank you.  I've updated the list of open issues accordingly.

> Re 3.4.5 Which password ops should be batchable?
> My take, all. validatePassword is the only questionable one but no 
> real reason to stop it just here... 

So noted.  I've taken the action item to post to the list a strawman of 
which operations should be batchable (or not batchable, whichever is 
shorter).

> 3.4.7.1 Neither search nor iterate is batchable (for reasons of scale)...
> What are the options for specifying this? If its not hard to use, then 
> why hot have every capability operation declare its ability to be 
> batched? 

We lost BatchableRequestType in an earlier version of the XSD,
so AFAIK this cannot be enforced in XSD.
I believe that we must specify normatively which requests are batchable.

> 3.4.8.1.2 & 3.4.8.2.2. May 'suspend/resume' return an error when the 
> object is already suspended/resumed?
> I say not too. If you ask for something and its already there - hey no 
> worries 

Done.

> Appendix K
> Yes we need a new notice, I'll send that in a separate mail. 

Received and applied.  Thank you.

> 5. Security considerations
> We need to give this a good amount of consideration. Our learned peers 
> at SSTC have much extended their documentation in this regard. We 
> should follow suit. We should look for a volunteer to head this effort 
> tat the next meeting.

This does seem important (and could be quite visible, given Sarbox et al).