Re: [charter-tool] Fwd: I-D Action:draft-ietf-genarea-charter-tool-04.txt

[Paul Hoffman <[e-mail]>]

On 2/4/11 10:12 AM, Robert Sparks wrote:
>
> On Jan 29, 2011, at 8:22 PM, Paul Hoffman wrote:
>
>> On 1/24/11 1:25 PM, Robert Sparks wrote:
> <snip/>
>
>>>
>>
>>> The tool should make it easy to track information and make the right
>>> changes if the name of the group changes during, or especially at the
>>> very end of, the BoF and review phase.
>>
>> Not sure what you mean here. It already allows changes; are you asking
>> that someone can look up the earlier name and find the current charter
>> proposal?
>
> At least. There have been a few cases recently where the WG charter that
> got approved had a different name for the working group than
> the name in all of the proposals (and sometimes even the name that went
> through the WG review call to IETF announce. If the tool is supposed
> to facilitate the approved charter against the proposals leading to it,
> it needs to make following that change easy.
>
> This could also happen during the sequence of proposals before it gets
> to the WG review stage.

Got it. Fixed in the next draft.

>>> The charter naming convention was useful for stating search and
>>> compare requirements, but I would prefer that it be described as a
>>> way to be precise in stating the requirement rather than declaring
>>> that the solution (something meeting those requirements another way
>>> might make tracking name change I discuss above easier for
>>> instance).
>>
>> I'm hearing mixed messages. :-) I think a contractor implementing this
>> would be much more likely to get it right the first time if we are
>> specific. If you want something different for the specific naming
>> algorithm I gave, that's cool, but asking a contractor to come up with
>> their own isn't likely to work as well as us defining one.
>
> I was hoping to leave the door open to a solution that didn't assume the
> distinction and mapping would be done based on filenames.
> These things could authoritatively live as records in a database instead
> of in discrete files.
> This is a classic spot where we are defining a mechanism instead of
> stating a requirement.
> Perhaps that's the right thing to do when managing the vendors at hand.

I formalized on filenames because the IETF has not had the greatest 
record of getting things out of databases run by different contractors 
over time. A mirrorable filesytem, although not the most efficient 
database, fixes that future transition issue in many ways.

>>> "Initial IESG and IAB review" does not match the string "Internal
>>> Review" that shows up in the agendas now.
>>
>> Correct. "Internal review" does not describe whom it is internal to.
>> Part of the purpose here is to make the states more understandable to
>> the wider audience who will now be tracking charters. Is the
>> difference a problem in this case?
>
> It is if the phrases start showing up on the screens of the produced tool.

They would, of course. I will change this back to "Internal Review" in 
the next draft, and hope that people looking at the tool understand whom 
the review is internal to.

--Paul Hoffman