[Foundation-l] Splitting Wikipedia by Project

Sat Feb 26 23:15:07 UTC 2011

There are several drawbacks to the idea of splitting EN wiki by
project, and I suspect the drawbacks will be equally true in other
languages.

1 Not everything fits neatly once into one project. So an article
about a Chilean Volcano might be of interest to projects as diverse as
Vulcanology, Chile, Rockclimbing and Botany. Together that makes for a
much better general article than if each project was only writing
about its aspect of the mountain.

2 Gnomes are useful, and will work across all sorts of articles across
one wiki, whether it is resolving death anomalies, adding intrawiki
links or resolving obscure typos. If you split EN wiki into seven
hundred or so different specialist pedia I might stay involved in some
of them - but I have no real interest in Bollywood or anime; Yet I
have huge numbers of edits there dealing with actors who were
"staring" in particular movies and heroes who "posses" particular
abilities.

3 We need 24/7 cover for admins to delete attack pages and block
vandals, and though our number of active admins on EN wiki is falling
by 1% a month, at present we can still provide that cover almost all
the time. Divide us into several hundred projects and we lose that - I
have admin rights on a small wiki outside Wikimedia where vandalism
can be up for hours.

4 As for splitting off BLPs - that would be as arbitrary and
unsuccessful as if we split off a pedia about places, buildings or
articles beginning in R. An article about a Taiwanese Baseball player
is a biography, but more significantly it is about a Baseball player
and a Taiwanese one at that.

Wikipedia is an incredible example of how the sum can be greater than
the parts, and in some aspects of economies of scale. But there is
more than that it - having a general encyclopaedia interlinked and
organised the way we have almost inevitably lures people away from
their initial interest and into editing stuff they might never
otherwise have dreamed of getting involved in. If anyone had told me a
few years ago that I would voluntarily be editing stuff about sport,
weather or MilHist I would  think they were mad. But I love being
involved in topics as diverse as King John, the Somerset levels and
Wikipedia:Featured article candidates/Johnstown Inclined
Plane/archive1 If wikipedia had been fragmented by project I would
probably still be doing a daily Sudoku and my garden would be somewhat
better tended.

WereSpielChequers

> ------------------------------
>
> Message: 5
> Date: Sat, 26 Feb 2011 15:35:39 -0500
> From: David Goodman <dggenwp at gmail.com>
> Subject: Re: [Foundation-l] breaking English Wikipedia apart
> To: Wikimedia Foundation Mailing List
>        <foundation-l at lists.wikimedia.org>
> Message-ID:
>        <AANLkTi=rcoQVdkLa4mVFkaf2uEAUEOOJchx7os-EFvUp at mail.gmail.com>
> Content-Type: text/plain; charset=ISO-8859-1
>
> To the extent that the enWP is a project to build a practical
> encyclopedia, it seems to have been getting increased acceptance as it
> gets larger. There is no indication that this trend is ceasing or or
> even faltering.
>
> To the extent that WP is an experiment, the experiment has already
> succeeded beyond the limits of similar projects, and there is no
> reason to stop at this point. Predictions that there would be a size
> beyond which it no longer scales have so far all of them been wrong.
> Splitting the encyclopedia is irreversible--we can always decide to
> split, but it is very unlikely that after sections develop separately
> they will be able to recombine.  But there  is nothing to stop anyone
> from making a split if they desire while leaving the actual Wikipedia
> as it is. I think WP can only benefit from serious competition.
>
> I agree the role of the wikiprojects should be increased and perhaps
> formalized, but already over  the last few years at the enWP,   some
> of the various WikiProjects and less organized impromptu groups of
> people interested in various aspects have made decisions that the
> community has not supported.  There is an advantage in having an
> Encyclopedia with uniform policies that have general agreement--people
> read it as  a whole & have common expectations.
>
> And with respect to BLPs, the biographical information about living
> people permeates most areas of the Encyclopedia, not just the articles
> with a living person's name as the title.
>
>
> On Fri, Feb 25, 2011 at 7:33 PM, John Vandenberg <jayvdb at gmail.com> wrote:
>> Was: Re: [Foundation-l] Friendliness (was: Missing Wikipedians: An Essay)
>> Was: Re: [Foundation-l] Friendliness
>>
>> On Sat, Feb 26, 2011 at 10:17 AM, Ryan Kaldari <rkaldari at wikimedia.org> wrote:
>>> On 2/25/11 3:11 PM, John Vandenberg wrote:
>>>> On Fri, Feb 25, 2011 at 11:18 PM, ?<dex2000 at pc.dk> wrote:
>>>>> ..
>>>>> I think it could also be considered to divide our huge language wikis
>>>>> into smaller parts. The existing WikiProjects could be made virtual wikis
>>>>> with their own admins, recent changes etc. That way, each project is in
>>>>> fact like a small wiki to which the newbie could sign up according to
>>>>> 'hers' area of interest and where the clarrity and friendlier atmosphere
>>>>> of the smaller wikis could prevail.
>>>>
>>>> This is the best solution, in my opinion.
>>>
>>> Yes, the larger wikis need to become WikiProject-centric. First step in
>>> doing this would be to create a WikiProject namespace. Second step would
>>> be to make WikiProject article tagging/assessment part of the software
>>> instead of template-based.
>>
>> I can see how those would be useful steps, however I think those steps
>> are part of a 10 year plan.
>>
>> A 10 year plan will be overrun by events.
>>
>> We need a much more direct plan.
>>
>> I recommend breaking enWP apart by finding easy chunks and moving them
>> to a separate instance, and having readonly copies on the main project
>> like we do for File: pages from Commons.
>>
>> IMO, the simplest and most useful set of articles to break apart is BLPs.
>> The criteria is really simple, and those articles already have lots of
>> policy differences around them.
>>
>> By the time we have perfected this system with the BLPs, the community
>> will have come to understand the costs/benefits of moving other
>> clusters of articles to separate projects, and we'll see other
>> clusters of articles migrated to sub-projects.
>>
>> btw, this idea is not new, but maybe its time has come.
>> http://wikipediareview.com/index.php?showtopic=29729
>>
>> --
>> John Vandenberg
>>
>> _______________________________________________
>> foundation-l mailing list
>> foundation-l at lists.wikimedia.org
>> Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/foundation-l
>>