[WikiEN-l] The Economist on "notability"

[Matthew Brown]

On Sat, Mar 8, 2008 at 12:36 AM, Ian Woollard <ian.woollard at gmail.com> wrote:
> On 08/03/2008, Matthew Brown <morven at gmail.com> wrote:
> >  Firstly, we honestly should not give a damn whether our Google
>  >  rankings are high or low.
>
>  Um. I think we want people to read the wikipedia though. Why would
>  people bother contributing if nobody reads it? Do our audience want us
>  to be high in google rankings? I would think so. A *lot* of people
>  find wikipedia articles that way. You more or less seem to be saying
>  you don't care about what our audience wants. Well, it's an argument.

If Wikipedia has good content people will read it.  We have had little
trouble finding readers without doing anything in particular to play
to the search engines, and I suspect this will continue.

There's a reasonable case to be made that our popularity has exceeded
our readiness for it.

>  There's subtle problems with abandoning notabilty, like every corner
>  store in the entire damn world would want and would be able to get a
>  wikipage. And that would push them way, way up in the google rankings.

I'm not, by the way, against having standards.  I just believe that
our standards are often set wrongly.

There's an argument to be made (as is done by Phil Sandifer among
others) that 'Notability' isn't even required to prune out the crap.
Insisting on verifiability and NPOV seems to do pretty well, for one
thing, and enforcing rules on conflicts of interest helps as well.

>  That's one of the important functions that notability deals with,
>  without it, every single tiny company in the whole world will have an
>  article for business reasons, and you know that many of them will be
>  forced for business reasons to anonymously sabotaging each other.
>  Could that ever be a good thing? How could we deal with that kind and
>  scale of in-article advertising? And once they're all in the database
>  it would be horrible to try to delete them; millions of companies, you
>  would have to go through one by one by one...

Strangely enough, people find having an NPOV article about their
things to be a lot less satisfying than they think it is.

I have no opposition to deletion reasons that include 'blatant
advertising', either, so long as they're correctly applied.

>  No, that's not really true, because the wikipedia implicitly (not
>  deliberately) uses SEO techniques; google juice enters and never
>  leaves. Basically almost any article in the wikipedia ranks higher
>  than almost the rest of the web for that reason.

I'm pretty sure that that's not the reason, since Wikipedia's ranking
doesn't seem to me to have improved since nofollow went into effect.

>  So ANY junky article in the wikipedia, is BIG in web terms. Do we have
>  a responsibility to the rest of the web? Not per se. But the rest of
>  the web decides how big we are and they can diminish us; that could
>  well mean that our current best articles become a lot less
>  significant. But apparently you say above we shouldn't care about
>  google. Uh huh.

I'm frankly unsure what you mean here.

>  Not if the wikipedia has an article on almost every word in the
>  English language, which it soon will have, and has effectively SEO'd a
>  bunch of non notable articles on any particular topic up above the
>  rest of the web. I say that it's really not a good idea at all for the
>  wikipedia to do that; they trust us, and we must not abuse that trust.

What are they trusting us to have?  Good information.  I'm also
confused by your use of 'non notable articles' here - generally the
deletionist argument is about topics, not articles, here, but perhaps
you used them interchangeably here.

If Wikipedia can have a good, encyclopedically written article on a
subject, I rarely see a problem with us having one.  I agree that a
very sucky article on something is little better than nothing at all
and may even be worse, but I'd tend to prefer improvement to deletion,
or if deletion, a deletion that does not preclude someone else
recreating a good article in its place in future.

>  It doesn't sound to me like they're following policy if it's ad hoc consensus.

"Consensus" on WIkipedia is often very minimal.  Have you seen how
many votes in total the average AFD discussion has?  Perhaps half a
dozen all told.

>  Particularly if you call 'odd' any meaasures that delete items deemed
>  'non-notable' you are personally fond of i expect.

I'm personally fond of useful encyclopedia articles, even if they're
on specialist subjects that would not be covered in Britannica.
Wikipedia is a specialist work as well as a generalist.

>  >  Yes, it's embarassing when some blogger points out that we cover some
>  >  random Pokemon better than we cover a head of state.  The solution to
>  >  that isn't to delete the Pokemon.
>
>  You'll need to point me to where the policy says that that's the process.

Policy is descriptive, not prescriptive.  It's my personal reading
that a lot of the pressure to delete certain subject areas comes from
editors who find it embarassing that Wikipedia covers 'unserious'
topics.

>  But there's the opposite problem as well; deleting notability because
>  somebody used it to delete Pokemon is *serious* overcorrection. The
>  right answer is to work out a criteria that allows Pokemon (if we
>  decide that Pokemon is desirable) and add it to the policy. NOT delete
>  the policy. Big changes to solve small problems are rarely a good
>  idea. You're suggesting removing a core policy that would create
>  millions of articles including small business and people's cats.

We can get rid of the flaws in the current 'Notability' idea without
throwing the baby out in the bathwater, I agree.

However, the problem is that the people who work on deletion are a
small subset of Wikipedia contributors, even regulars, yet they have
an outsize effect on our deletion policies - especially since policies
and guidelines are constantly changed.

That's because only deletionists spend so much effort on deletion
discussion.  People like me are out writing stuff.

>  >  I also see no reason why we have to use any automated measure of
>  >  quality that creates 'losers'.
>
>  Then you just haven't thought it through fully; which is not
>  surprising, removing notability would be significant and complicated
>  enough, that I'm not sure anybody CAN think it through fully. And if
>  that doesn't scare you then it really, really, really should.

I remember Wikipedia before 'Notability'.  It didn't work so bad.

In general, though, I see in 'Notability' a very CS-major computer
nerd way of trying to go about things.  An attitude that says
consistency beats getting things absolutely right.  A belief that we
can create policy as a computer program that ensures the right outcome
- and even if that outcome isn't perfect, it's better that it be
automatable than involve human judgment.  A desire to make policy with
a primary emphasis on it being hard-edged and definitive, because it
makes the job of the deleters easier.

I have problems with that.

-Matt