Processor profiles next steps

Hi folks,

Henry and I chatted a little bit about the processor profiles document
and next steps on same last week (er, a couple of weeks ago now, but
who's counting).

I went into the conversation feeling like we had a long way to go. I'm
happy to say that I left feeling like there's a much shorter distance
between where we are and victory.

My notes contain two concrete suggestions:

1. Amend the introduction to note that we're talking entirely about
batch processors, not interactive processors. There's no dicscussion
of reporting requirements in the case of "live" data models. As Alex
has demonstrated, there are definitely more things to consider if
you're talking about the effects of modifying a data model "in
memory". But the XML family of specs have always been pretty silent on
that point and I don't see us undertaking to fix that in this version
of this document.

2. Rename the profiles. After twenty minutes of head scratching,
here's the best we could come up with:

   Basic profile
   ID profile
   Not-self-contained profile
   Full profile

Clearly "not-self-contained" is a crap name. Better suggestions most
welcome.

I suggest we make these changes, respond to the last call comments
that we got (some of which will take time and energy, volunteers
welcome), and see if that gets us over the finish line.

(Henry, please let me know if you think I missed anything or
mischaracterized our discussion.)

                                        Be seeing you,
                                          norm

-- 
Norman Walsh
Lead Engineer
MarkLogic Corporation
Phone: +1 413 624 6676
www.marklogic.com