Archives/2007-03-03

From LangCom

This page is no longer maintained and may be outdated: moved to m:Special projects subcommittees/Languages/Archives/2007-03-03.
Archived discussion
This is an email discussion about archival and transparency, scope, and processing method. A consensus was reached on scope and processing method, and everyone except GerardM agreed to public archival.


1.

Jesse Martin (Pathoschild)
03 March 2007 14:13


Hello,

There are a few points I'd like to clarify.

1. Archival and transparency.
Our charter<1> states that the whole set of our activities are public, but virtually all discussion is through private email or on IRC. I would like to archive our email discussions on the subcommittee wiki. Does anyone object to this?

2. Scope.
Re-reading our charter, we do not have the explicit authority to make decisions (although it is implied: "a clear step-by-step policy [...] for evaluating the feasibility of new language wikis with an automated procedure for project development"). Should we forward every request to the board, regardless of how unfeasible the request is? Or does the charter suggest some leeway in filtering requests, and we should only forward feasible requests?

3. Processing method.
How will we process requests? There are currently 39 open requests, of which two are in the test project phase. One method would be to propose decisions through email, and if there are no objections after a delay (perhaps 24 hours) the decisions are implemented and the next batch is proposed. If anyone objects but needs more time to respond, they can just say something like "I object to the second proposal; I'll explain later, no time now".

<1> charter: http://langcom.wikimedia.org/wiki/Main_Page#Subcommittee_charter

Yours cordially,
Jesse Martin (Pathoschild)

2.

Berto 'd Sera
03 March 2007 14:29


Hi Jesse!

1. Archival and transparency.

Okay for me :)

> 2. Scope.
> Should we forward every request to the board,

No. It will take ages to deliver a new wiki even if filter them. And if we do not what exactly is our function? Our archives are public, none can say we produce wiki-desaparecidos :) But we exist to make sure that requests are well-formed, so it should be us to decide what is well-formed and what is not.

3. Processing method.
>of which two are in the test project phase.

You mean in the incubator? If so it's only two. No work in the incubator = no wiki. There is no point in making requests when it's not clear whether the project has the strenght it takes to become a wiki.

> "I object to the second proposal; I'll explain later, no time now".

I object in principle to anyone not having:
- an ISO 639 code (but I'm willing to help to get one, if they complain with our definitions)
- a decent work done (with no robots) in the incubator. Say 5 users and 500 articles or whatever we stated in the official policy (can't remember now), a fully translated UI and a basic policy (at least the "Neutral educational content" clause should be there in a clear form).

I say "no robots" because it's unfair. This way a geek with 3 native speakers can have a wiki in two days while a community of 2 million people from Africa would wait for ages. A community is not made by python slaves, it's made by living people. Once they have a wiki I'll happily spend my time to help them with robots, but that's AFTER they got approved as real humans.

Bèrto ‘d Sèra

3.

Jesse Martin (Pathoschild)
03 March 2007 14:54


Hello Berto 'd Sèra,

The policy requires a code before a proposal can be approved. There are some cases where a language deserves a code but doesn't have one yet; it would be perfect if you can help users get one.

There's no explicit requirements in the policy that define a 'successful' test project, because that is best left to judgment. Trying to define every case where it is successful or not would be difficult and wouldn't work anyway. It would be a good idea to write some basic guidelines for successful test projects, but we shouldn't codify them as inflexible requirements.

For example, 500 articles and ten users is a failure if there are ten bot accounts automatically importing text; on the other hand, it's somewhat successful if the proposal is *supposed* to be mainly bot-edited, as with some of the wikis that automatically convert text from another wiki's writing system.

In general, I agree that we want a human community, not an army of bots. If there are both, all the better— bots are excellent tools.

Yours cordially,
Jesse Martin (Pathoschild)

4.

Jon Harald Søby
03 March 2007 20:15


On 3/3/07, Jesse Martin wrote:
> Hello,
>
> There are a few points I'd like to clarify.
>
> 1. Archival and transparency.
> Our charter<1> states that the whole set of our activities are public,
> but virtually all discussion is through private email or on IRC. I
> would like to archive our email discussions on the subcommittee wiki.
> Does anyone object to this?

I have no objections. However, I think archiving would be more practical if we got our own mailing list; this would also make communication easier (pressing just "reply" instead of "reply to all" ;-).

> 2. Scope.
> Re-reading our charter, we do not have the explicit authority to make
> decisions (although it is implied: "a clear step-by-step policy [...]
> for evaluating the feasibility of new language wikis with an automated
> procedure for project development"). Should we forward every request
> to the board, regardless of how unfeasible the request is? Or does the
> charter suggest some leeway in filtering requests, and we should only
> forward feasible requests?

I believe the board has has more important things to do than to judge on language matters, and frankly, I think it doesn't fall within the board's scope - what language editions exist or don't exist of Wikipedia really has little with the Wikimedia Foundation itself. But of course, this is up to the board itself - we should probably ask.

> 3. Processing method.
> How will we process requests? There are currently 39 open requests, of
> which two are in the test project phase. One method would be to
> propose decisions through email, and if there are no objections after
> a delay (perhaps 24 hours) the decisions are implemented and the next
> batch is proposed. If anyone objects but needs more time to respond,
> they can just say something like "I object to the second proposal;
> I'll explain later, no time now".

The e-mail approach would be easier with a mailing list; each new language proposal should have its own thread, perhaps marked in a specific manner (e.g. with a [new] tag or something). The delay should be considerably longer than 24 hours, however, I'm thinking 1 week - 10 days.

--
Jon Harald Søby
http://meta.wikimedia.org/wiki/User:Jon_Harald_S%C3%B8by

5.

Jesse Martin (Pathoschild)
03 March 2007 20:22


Hello,

I agree on all points. If the delay is longer, though, we'll need to consider several requests simultaneously.

Yours cordially,
Jesse Martin (Pathoschild)

6.

GerardM
04 March 2007 11:33


<this user has not agreed to public archival.>

7.

Berto 'd Sera
04 March 2007 15:42


Hoi!

<this text is quoted from a user who has not agreed to public archival.>
Yes. My impression is that people simply don't give a damn about it. Or maybe they are afraid to speak a clear word about it. It takes 5 minutes to see the material and say yes/no. Why on earth is WMF becoming more twisted than the former Central Committee of the Soviet Communist Party!?

<this text is quoted from a user who has not agreed to public archival.>
Yes. We need real data to make a decision, not just the usual list of names that may be fully invented (anyone can call up friends to register as native speakers). Let's see real stuff befre we spend real words. But once we have it let's be quick, or the people will be upset.

BTW, if I wanted everyone upset against this Commitee I'd do exactly what the Board is doing. I'd have everything paralyzed, and let the Committee be the escape goat... Maybe for some people this is a very good reason for them to become REAL SLOW :) Childish, isn't it?

Bèrto ‘d Sèra

Navigation