Re: [security-services] proposal wrt "status"and"send comments to the editor"

["Eve L. Maler" <eve.maler@sun.com>]

At 11:48 PM 1/6/02 -0800, Jeff Hodges wrote:
>Overall comment/thought -- it may well be UNreasonable to make all of 
>these doc metadata changes to the doc set we're attempting to coalesce 
>this week.

You're probably right, and the metadata is already getting pretty shipshape 
compared to earlier drafts...

> > Finally, we should beware that the bibliographic references to other SAML
> > drafts that appear in all our specs go "stale" quickly because the URLs in
> > them mention draft numbers (e.g., bindings-08).  We need to check them all
> > before we go live.  If we were to copy over each latest published draft
> > into a generic "sstc-draft-bindings.pdf" file (etc.), it would be
> > convenient then to change all the bib stuff once more, for good.
>
>This is a hard one cuz it's messy any way we try to do it. If we do it via the
>canonical "this is the url to the latest draft" as you suggest, then we 
>have to
>remember to update the "generic file" every time we put a new doc rev up 
>in the
>doc repository. Otherwise, the doc editors need to remember to go through 
>their
>bib sections and update appropriately. My experience is having to do the
>latter, so I tend to lean that ways. I could be convinced to do the former I
>suppose (but it adds to the responsibilities of the web site maintainer).

I think it's better to have it be the explicit job of the website 
maintainer (really the person who "publishes" documents to the doc 
repository, so it could happen at that time) than have it be the implicit 
job of multiple other spec editors who don't necessarily track when a 
particular spec has changed.

         Eve
--
Eve Maler                                    +1 781 442 3190
Sun Microsystems XML Technology Center   eve.maler @ sun.com