Re: Updating BCP 10 -- NomCom ELEGIBILITY

[Brian E Carpenter <[e-mail]>]

On 12/02/2015 09:48, Michael Richardson wrote:
> 
> Brian E Carpenter <[e-mail]> wrote:
>     > Being geeks, we have a tendency to invent solutions that
>     > are theoretically wonderful but quite impractical. The current
>     > 3/5 rule avoids that and is fairly easy to check algorithmically.
> 
>     > I think we need to keep that simplicity as much as possible. So
>     > something like the following might work.
> 
>     > Either
>     > 3 out of the last 5 meetings
>     > or
>     > 3 out of the last 10 meetings and (RFC author or WG Chair or AD in the
>     > last 5 years)
> 
> When you say "RFC author", I guess you mean something which went through the
> process to get a number.  I guess that means that having your name on the
> front page now has more significant value now.   If two IDs merge, now it
> matters who is an editor, and who is in "Contributors" at the end...
> 
> I would like to know why my previous list was too long or unclear?

For me there is an over-arching consideration: a relatively simple
script must be able to generate the list of all qualified volunteers
from existing databases. Otherwise the whole process will be
impractical.

Inevitably that results in a somewhat blunt instrument.

   Brian

> I specifically listed:
> 
>   0) attend the meeting in person.
> 
>   1) be a document shepherd or working group chair on a document
>      that entered AUTH48.
> 
> "WG chair" was listed above, and I added "that entered AUTH48", because I
> wanted to deal with the possible way to game things where a WG is created as
> a sinecure. No dead WG.  It's the act of going through AUTH48 that I care
> about, not who you were. [The authors of the document get caught by 2]
> 
> And I included shepherd, because I want to provide more opportunities for
> more people; if you attend meetings rarely, you are unlikely to be visible to
> ADs for selection as an WG chair, but shepherds are in short supply.
> And all the above things are trivially checked in the datatracker.
> 
>   2) be the document uploader (pressed submit) on a document that
>      was scheduled into a WG session.
> 
> So anyone can upload at any time.  If the WG takes your document on,
> then the datatracker clearly knows this, so that part is easy to check.
> If you draft-author- is scheduled, this will require some additional
> datatracker effort for the WG chairs to indicate that your document goes into
> the agenda... but, I we wanted to be able to track which documents were
> discussed into which WG anyway for all sorts of reasons, so I think that his
> work will already happen.
> 
>   3) opened a ticket on a document that was scheduled into a WG session.
> 
> While the Issue tracker is not well integrated into the datatracker, I know
> that many want it to be, and also that many want many other improvements.
> Again, knowing that an issue was discussed at a WG session is pretty
> important thing to track, I think... so really the datatracker OUGHT to know this.
> 
>   4) scribed for the I* telechats.
> 
> This requires secretariat/I* action to confirm, and is frankly probably my
> least interesting way.
> 
> Now, I can see how "AD" could make one eligible, and how an AD after a 6
> year term, could easily become non-eligible having taken a break. But:
> 
>      I don't like AD this way because it looks too much like the previous
>      IESG is selecting the new IESG. Yeah, random selection.
> 
>      Since former ADs are well known, often well spoken (thus causing them to
>      write IDs), and are often asked to join directorates and do reviews
>      (therefore opening issues), and might even be really really good
>      shepherds...  I prefer rules about what you do, not who you are/were.
> 
> --
> ]               Never tell me the odds!                 | ipv6 mesh networks [
> ]   Michael Richardson, Sandelman Software Works        | network architect  [
> ]     [e-mail]  http://www.sandelman.ca/        |   ruby on rails    [
> 
> 
> 
> 
> 
> 
> --
> Michael Richardson <[e-mail]>, Sandelman Software Works
>  -= IPv6 IoT consulting =-
> 
> 
>