Printable Version of Topic

Click here to view this topic in its original format

_ The ArbCom-L Leaks _ Even the Arbitrary Committee realizes the system doesn't work

Posted by: MaliceAforethought

This has been a common thread I keep running into during searches. More added as I see it.
**********

From: (Kirill Lokshin)
Date: Sun, 4 Nov 2007 23:46:41 -0500
Subject: [Arbcom-l] Internal organization

I think it is becoming painfully obvious to the community at large -- and to
us, I suspect -- that the Committee is not functioning very efficiently (if
it ever did). Questions go unanswered; requests and motions sit for weeks,
if not months, without any response; cases stay open for ever-longer
periods; and so forth.

The chief problem, in my view, is not a lack of (potential) activity within
the Committee as a whole, per se; but, rather, the fact that nobody is
actually responsible for making sure that everything gets done.

This is most obvious with requests/questions/appeals/etc. that come to the
list directly -- in most cases, they tend to sit unanswered for weeks on
end, regardless of whether any discussion occurred, because everyone
apparently assumes that someone else will answer them. (In a certain sense,
this is also caused by the lack of any clearly defined means for issuing
"official" replies without having a month-long vote; nobody actually has the
explicit authority to do so, so anything that does go out tends to be
whatever the first Arbitrator to respond thought of the matter.)

The natural consequence of this is that little gets done unless someone on
the outside -- typically a requester or a party to a case -- keeps bugging
us about it over and over. (And, of course, since the clerks are too polite
to do that, their questions tend to be completely ignored a large portion of
the time.)

I suspect that, if we continue to operate in a completely anarchic fashion,
we will be increasingly hampered by this inability to manage both internal
work and external responses. I think that it's time to consider whether the
benefits of introducing some sort of more defined organizational structure
might not outweigh the drawbacks of doing so.

The simplest solution, in my view, would be to create some sort of chair
position for the Committee. The chair would be tasked primarily with
keeping track of things to be done internally and making sure things move
along and don't fall through the cracks; and also, perhaps, with issuing
responses to external questions. Obviously, the creation of any such
position opens up a can of worms -- if I'm not mistaken, it was considered
and rejected when the Committee was first created? -- but I don't think
we're going to be able to function in an effective manner if the current
state of affairs continues indefinitely.

Kirill
----------

From: misfitgirl(Rebecca)
Date: Mon, 5 Nov 2007 16:25:41 +1100
Subject: [Arbcom-l] Internal organization

Honestly, I think Kirill is right on the money on this one. The
committee has had problems with getting cases and inquiries processed
expediently and generally getting stuff done since its inception, and
many an attempt at changing that hasn't succeeded. Appointing a chair
to oversee the whole process, make sure questions are answered and bug
other arbitrators when necessary would be really helpful.

Rebecca
----------

From: (charles.r.matthews)
Date: Mon, 5 Nov 2007 10:02:34 +0000
Subject: [Arbcom-l] Internal organization

> The simplest solution, in my view, would be to create some sort of chair
> position for the Committee. The chair would be tasked primarily with
> keeping track of things to be done internally and making sure things move
> along and don't fall through the cracks; and also, perhaps, with issuing
> responses to external questions. Obviously, the creation of any such
> position opens up a can of worms -- if I'm not mistaken, it was considered
> and rejected when the Committee was first created? -- but I don't think
> we're going to be able to function in an effective manner if the current
> state of affairs continues indefinitely.
>
> Kirill

I think we have to move carefully onto this ground. It is perfectly true that the AC flags, often. It is also true that there is plenty of incoming mail, and it is no one's particular job to answer or track it.

The AC has traditionally functioned more like a committee than an executive body. In other words, we discuss, as peers; we take on work, as we wish. All very like Wikipedia, in fact.

Now, we have the new wiki up. Various tasks have been mentioned with which it could help. These include:

- drafting
- "filing", in other words keeping factual notes
- workflow
- collective memory, so that we can refer to older matters and characters as they crop up again.

One thing that occurred to me as I was putting up a few pages on the wiki is that it was the first time I was really thinking about the various areas of work, as a whole and top-down. It is very different having things in email threads, versus wiki pages.

The point I would make right now is nagging could be more organized, but that is not the only issue to be addressed.

Charles
----------

From: (Dmcdevit)
Date: Mon, 05 Nov 2007 02:10:25 -0800
Subject: [Arbcom-l] Internal organization

I'm not sure if a Chair is the solution, though. Or if it is, I'm not
sure if it's workable. Basically, it's someone whose job is to nag
people, along with their other arbitration tasks? ;-) One thing that may
be helpful now that we have the wiki is to put things that need a
*decision* and used to be discussed haphazardly on the list in a wiki
page. Like our current CheckUser thread, or the odd thread about
unbanning/banning/resysopping/desysopping etc.: they could be private
votes on the wiki in and that way people could easily see what issues
were still open and comment accordingly.

Dominic
----------

From: (charles.r.matthews)
Date: Mon, 5 Nov 2007 10:31:27 +0000
Subject: [Arbcom-l] Internal organization
Message-ID:

> One thing that may
> be helpful now that we have the wiki is to put things that need a
> *decision* and used to be discussed haphazardly on the list in a wiki
> page. Like our current CheckUser thread, or the odd thread about
> unbanning/banning/resysopping/desysopping etc.: they could be private
> votes on the wiki in and that way people could easily see what issues
> were still open and comment accordingly.
>
> Dominic

You are suggesting things more like straw polls, on the wiki? I think that could be helpful for some issues, certainly. Probably information gathering and opinion gathering are better kept apart ...? Wikis can cope with that.

Charles
----------

From: (Timothy Titcomb)
Date: Mon, 5 Nov 2007 07:28:35 -0500
Subject: [Arbcom-l] Internal organization

On Nov 4, 2007, at 11:46 PM, Kirill Lokshin wrote:

> I think it is becoming painfully obvious to the community at large
> -- and to us, I suspect -- that the Committee is not functioning
> very efficiently (if it ever did). Questions go unanswered;
> requests and motions sit for weeks, if not months, without any
> response; cases stay open for ever-longer periods; and so forth.
>
> The chief problem, in my view, is not a lack of (potential)
> activity within the Committee as a whole, per se; but, rather, the
> fact that nobody is actually responsible for making sure that
> everything gets done.
>
> This is most obvious with requests/questions/appeals/etc. that come
> to the list directly -- in most cases, they tend to sit unanswered
> for weeks on end, regardless of whether any discussion occurred,
> because everyone apparently assumes that someone else will answer
> them. (In a certain sense, this is also caused by the lack of any
> clearly defined means for issuing "official" replies without having
> a month-long vote; nobody actually has the explicit authority to do
> so, so anything that does go out tends to be whatever the first
> Arbitrator to respond thought of the matter.)
>
> The natural consequence of this is that little gets done unless
> someone on the outside -- typically a requester or a party to a
> case -- keeps bugging us about it over and over. (And, of course,
> since the clerks are too polite to do that, their questions tend to
> be completely ignored a large portion of the time.)
>
> I suspect that, if we continue to operate in a completely anarchic
> fashion, we will be increasingly hampered by this inability to
> manage both internal work and external responses.

This is an important matter and needs addressing. Thanks Kirill for
stating the problem so well.

> I think that it's time to consider whether the benefits of
> introducing some sort of more defined organizational structure
> might not outweigh the drawbacks of doing so.
>
> The simplest solution, in my view, would be to create some sort of
> chair position for the Committee. The chair would be tasked
> primarily with keeping track of things to be done internally and
> making sure things move along and don't fall through the cracks;
> and also, perhaps, with issuing responses to external questions.
> Obviously, the creation of any such position opens up a can of
> worms -- if I'm not mistaken, it was considered and rejected when
> the Committee was first created? -- but I don't think we're going
> to be able to function in an effective manner if the current state
> of affairs continues indefinitely.

I would like us to use the new wiki to help address this problem.


Paul August
----------

From: (Kirill Lokshin)
Date: Mon, 5 Nov 2007 10:29:29 -0500
Subject: [Arbcom-l] Internal organization

Yes, the wiki will definitely help with this stuff; but I don't think it's
going to solve everything in and of itself. Even if we copy all incoming
traffic to the wiki for processing there (and it's not clear who'd be
responsible for doing that, or whether the added work would be worth it),
we'd still need someone to keep track of it on-wiki and make sure it got
handled in a reasonable time. I don't think simply leaving a discussion
open there and assuming that people will show up is going to be very
efficient; it doesn't work all that well for the various requests on RFAR.

Fundamentally, the normal Wikipedia way of doing things -- people picking up
work as they choose -- is very much an eventualist one at heart.
Interesting things get done quickly; boring ones slowly, or not at all.
This works well so long as you have no concept of deadlines and don't
actually need to handle everything; but I don't really think that either of
those is true for the bulk of what we deal with. We can't really afford to
have a "Questions unanswered since August 2007" category and the like, after
all -- even if creating it would mean that the questions would *eventually*
get answered.

Kirill
----------

From: (jayjg)
Date: Mon, 5 Nov 2007 10:49:22 -0500
Subject: [Arbcom-l] Internal organization

The main reason I kept trying to set time-limits and acceptance-limits
on various aspects of the cases was for this very reason; the large
number of inactive or barely active arbitrators meant that many cases
were doomed to months of "hanging around" on the Arbitration page,
following by a fitful petering-out. Yet these limits were resisted.
I'm not sure why Newyorkbrad kept saying "well, we've hit the 10 day
limit, and not enough arbs have voted to accept, but I'm reluctant to
delete the case..." The Arbcom *always* gets burned out, with some
arbitrators completely uninvolved, and others barely involved, and it
usually starts in February. There's no point in dragging out cases
where a number of arbitrators don't even think there should be a case
to begin with. If the active committee is 5 people, so be it; put
everyone else on the inactive list, and resolved accordingly.

On a related note, there are a number of people on this list who
haven't e-mailed the list in months, perhaps years. I think that this
list should be for active contributors; if someone hasn't mailed the
list in 3 months, then drop them. Thoughts?
----------

From: (charles.r.matthews)
Date: Mon, 5 Nov 2007 16:54:36 +0000
Subject: [Arbcom-l] Internal organization

"Kirill Lokshin" wrote

> Yes, the wiki will definitely help with this stuff; but I don't think it's
> going to solve everything in and of itself. Even if we copy all incoming
> traffic to the wiki for processing there (and it's not clear who'd be
> responsible for doing that, or whether the added work would be worth it),
> we'd still need someone to keep track of it on-wiki and make sure it got
> handled in a reasonable time. I don't think simply leaving a discussion
> open there and assuming that people will show up is going to be very
> efficient; it doesn't work all that well for the various requests on RFAR.

Well, I floated "straw polls". We do this anyway, really: some query or proposed action is raised on the list, comments are called for. The wiki method is similar:

A creates a wiki straw poll page, says so on the list;

B, C, D who are interested chip in on the page;

In the end everyone has said their piece, and if there is a decent consensus then that counts as decided.

The advantages of the wiki method would be

(i) A can announce a time limit, say 5 days;
(ii) Anyone who feels a limit is too long/short edits it on the wiki.
(iii) Informal closures can be done and tracked on the wiki, with a nag email sent round 24 hours ahead.
(iv) Side discussions and diffs etc. will help to build up the wiki as general documentation.
(v) Contentious points come back to the list.

I think this type of workflow is OK; if A is concerned enough to set up a page, A will presumably nag and close to get a decision.

> Fundamentally, the normal Wikipedia way of doing things -- people picking up
> work as they choose -- is very much an eventualist one at heart.
> Interesting things get done quickly; boring ones slowly, or not at all.
> This works well so long as you have no concept of deadlines and don't
> actually need to handle everything; but I don't really think that either of
> those is true for the bulk of what we deal with.

Ummm, we aren't providing a service.

>We can't really afford to
> have a "Questions unanswered since August 2007" category and the like, after
> all -- even if creating it would mean that the questions would *eventually*
> get answered.

I don't entirely agree. Some issues - not all I grant - are really from deeply boring people who want something fixed for them. We should be quicker to deal with issues that can cause damage if they are left. "The community" does unravel certain things over time. Basically we should be quick when our remedies are likely to be a big improvement, but not pretend that is always.

Charles
----------

From: (Kirill Lokshin)
Date: Mon, 5 Nov 2007 13:10:16 -0500
Subject: [Arbcom-l] Internal organization

> A creates a wiki straw poll page, says so on the list;
>
> B, C, D who are interested chip in on the page;
>
> In the end everyone has said their piece, and if there is a decent
> consensus then that counts as decided.
>
> The advantages of the wiki method would be
>
> (i) A can announce a time limit, say 5 days;
> (ii) Anyone who feels a limit is too long/short edits it on the wiki.
> (iii) Informal closures can be done and tracked on the wiki, with a nag
> email sent round 24 hours ahead.
> (iv) Side discussions and diffs etc. will help to build up the wiki as
> general documentation.
> (v) Contentious points come back to the list.
>
> I think this type of workflow is OK; if A is concerned enough to set up a
> page, A will presumably nag and close to get a decision.


So long as we get some A that cares enough about the matter to go through
this, yes. I'd suggest, however, that a substantial portion of the matters
brought before us do not interest anyone on the Committee enough for this to
happen reliably. A large number of queries spur no discussion on the
mailing list; I doubt changing the location for such potential discussion to
a wiki will really alter that.

> Fundamentally, the normal Wikipedia way of doing things -- people picking
> up
> > work as they choose -- is very much an eventualist one at heart.
> > Interesting things get done quickly; boring ones slowly, or not at all.
> > This works well so long as you have no concept of deadlines and don't
> > actually need to handle everything; but I don't really think that either
> of
> > those is true for the bulk of what we deal with.
>
> Ummm, we aren't providing a service.
>
> >We can't really afford to
> > have a "Questions unanswered since August 2007" category and the like,
> after
> > all -- even if creating it would mean that the questions would
> *eventually*
> > get answered.
>
> I don't entirely agree. Some issues - not all I grant - are really from
> deeply boring people who want something fixed for them. We should be quicker
> to deal with issues that can cause damage if they are left. "The community"
> does unravel certain things over time. Basically we should be quick when our
> remedies are likely to be a big improvement, but not pretend that is always.


I think there's a distinction between acting quickly and responding
quickly. When someone asks us a question, I think they ought to be able to
expect *some* response within a reasonable timeframe; and I think it
harms our standing within the community when we consistently fail to provide
one.

Kirill
----------

From: (charles.r.matthews)
Date: Mon, 5 Nov 2007 19:23:46 +0000
Subject: [Arbcom-l] Internal organization

> I think there's a distinction between acting quickly and responding
> quickly. When someone asks us a question, I think they ought to be able to
> expect *some* response within a reasonable timeframe; and I think it
> harms our standing within the community when we consistently fail to provide
> one.
>
> Kirill

Well, you may get a better class of questioner. I've been having the pedophile stuff on my Talk page. And they don't just stop at one question. And they come back. I don't see that I'm unresponsive. Some people think they have a right to pose as many queries as they want, independent of what else one might have to do. I suspect that if we established some sort of expectation that we always answer mail, teh main effect would be to create more mail.

Charles