Hi Andreas,

> - Copyright should be "2005, 2006" since this is a revised version of GFD.58.

yes, thanks.

> - Sec. 1: The first sentence "The main purpose of OGF work is to produce 
> documents."  feels too abrupt. The mission of OGF is stated on the web site ("
> accerates grid adoption...developing open standards"). Producing documents is 
> one concrete outcome, for sure, but I think that it would be nicer if this 
> document  'eased' into this statement via (a variant of) the more general OGF 
> mission statement .

Well, yes. Basically, I never really liked it but couldn't really find a better alternative until now.
Thanks for pointing this out!

> - Sec 2.4: Though not a schema URI, OGSA-WG and a number of other WGs defining
>  Profiles have adopted the customs part "www"  as defining profile claim URIs 
> (with otherwise the same formation rules). It would be nice to see this added 
> to this document for documentation purposes. (Since this is a community 
> practice document I think it is appropriate.)

Yes. Though formally this is covered by the extension element in the customs part, I agree that the obvious adoption 
gives reason enough to add it as a separate literal value.

> - Sec. 2.5.1: "2 of August" date to obsolete the old .ggf.org namespace should
>  really be the day that this document is officially published, since it cannot
>  be binding before that or even retroactively. I would suggest deleting the 
> date altogether and stating instead that using .ggf.org is deprecated.

Again, yes. I changed the paragraph accordingly.

> - Sec. 2.6: It is unclear to me why 'project' is allowed a second time in the 
> 'specific' part. Surely for the JSDL spec the second occurrence is *not* the 
> project name but the specification name. Though it is true that often the WG 
> name is also the specification name I think it is misleading to call this 
> second occurrence the project name.

This is purely because of the namespace generation rules. 
Semantically, you are right that the second occurrence of "project" in that rule is actually the specification name. 
However, as this is an Augmented BNK rule, I simply reused the "project" rule instead of introducing another rule for 
the specification name.

> - Sec. 2.7: In the example there is no  statement that /rss and /graap are 
> correct

Thanks, changed that.

> - Sec. 2.8: I think it is too restrictive to require version be in year / 
> month format only. Though this is common style there are also recent cases of 
> specifications opting for a non-date version number (e.g., the WS-CIM schemas)
> . I would suggest changing this to a recommendation and perhaps mentioning 
> this second possibility.

The document states that Namepsace IRIs, which have been generated according to this document, are unique. Clearly, the 
year and month parts are intended to ensure uniqueness of Namespace IRIs within one group's specification history. 
Allowing almost arbitrary versioning schemes would introduce yet another level of complexity, which we would like to 
avoid.

> - Section 3. 'GGF'  should be 'OGF'
> -- A process should also be defined on how to place schemas in the online 
> repository. Do only formally (GFD) published schemas get uploaded, who can 
> upload etc. If drafts can also be placed who does that, who approves, etc. The
>  definition does not have to be part of this document but the requirement to 
> define a clear process should be stated.

Yes, I fully agree that such process need to be defined, but this is out oof scope for this document. However, I 
mentioned this explicitly in the revised introduction.

> - Full copyright notice: "Open grid forum" should be "Open Grid Forum"

Thanks, corrected.

> As a general comment  it might be worth looking into restructuring this 
> document in such a...