[ANSTHRLD] Asterisk?

[Kathleen O'Brien]

>Various disjointed stream-of-consciousness blatherings on the
>submissions arm of the Ansteorran heraldic octopus.

Now _THAT'S_ an accurate description of it!


>After us were Kathri Asterisk and Mari Bordure, both skilled,
>hard-working (oy!), and willing to work things out.  I presume
>everything went smoothly there too; *I* never heard of any problems.

With a well-working pipeline, you won't.  Problems get noticed and fixed
before anyone outside the pipeline smells anything.  :)  Stuff happens.
Welcome to Real Life.  We're a volunteer organization and we're all human.
I think one of the biggest assets in an officer is the willingness to stand
up and say "Life exploded on me.  I need a replacement!" or "I need help
this month!"  There are always folks willing to lend a hand.  And, in most
cases, if a person bails when life explodes and does a clean handoff of
their position, folks are very willing to see them have another position
later - 'cause you know they'll communicate if something goes wrong.

Part of the trick of a good pipeline is having the sections be small enough
that they can be handed off without drowning the person willing to dive in
and take over.  I'll use the Bordure office as an example since I know it
best.  A one or two month backlog of stuff to do for Bordure could be
tackled in a single day on a weekend by a determined person.  The same
cannot be said for other tasks.  For example, Daniel could give a better
estimate, but it looks to me like it takes 3 people an entire Saturday just
to process the inbound submission packets for the Laurel office each month.


>Submissions have had some problems since.  I'm not that up on gossi^W
>internal knowledge, but I think it was the problem of pipelined
>processing: if one part blocks, the whole pipeline chokes off.
>I wonder how well parallel processing might work: two or more people
>with the same job, and they switch off?

I'd still vote for the pipeline approach.  Each stage of the pipeline is
small enough that most people with a full time job can handle their piece
of the pipe during most months.  There's one major advantage to the
pipeline style octopus: any burps in the flow are noticable inside the
pipeline almost immediately.  That gives the people in the pipe a chance to
fix stuff before it becomes a Real Mess (TM).

The other advantage of a pipeline is that it doesn't take that much
coordination with other people to get your job done.  No "decisions by
committee".


>(Gwenllian and I had an agreement that if life was hideous for either
>of us for a month or two, we'd take over the other's job.  In the
>event, we didn't need it: I just skipped two LoIs and put the
>submissions in next month, and she simply kept up with all the work
>while researching and getting a PhD, providing economic advice to Alan

I was Gwenllian's deputy for the second year of her tenure.  I lived about
10 minutes from her, had a scanner, and had the advantage of knowing
something about submissions.  :)  There was one month when PhD stuff fell
on her like a ton of bricks.  That month I drafted the ILoI and she just
did a final edit of it.  After that point, we decided that I should take
the "extra" name form and process all the name submissions into the ILoI -
to cut down the workload on her.  Those name forms are the ones that are
now forwarded on to Arbalest.  So we kinda did have an "in parallel" thing
going there.  If she got swamped, I could jump in (provided my team at work
wasn't shipping software that week...)


Mari