[WikiEN-l] "Articles that do not cite reliable published sources will be deleted."

[Bryan Derksen]

Cheney Shill wrote:
> --- Delirium <delirium at hackish.org> wrote:
>> In practice, yes, if:
>> 1) The information looks likely to be correct; and
> 
> Problem is that's subjective.  How does one determine what
> is or is not likely?

Subjectivity is not inherently a problem, and for a project like this
it's bound to happen at some point. In this case one determines whether
it's likely based on whether anyone actually challenges it. If nobody
challenges a bit of information then either everyone thinks it's likely
to be correct (that's good) or nobody has noticed it (not so good, but
also makes the issue not terribly urgent). Or perhaps nobody who's read
it feels qualified to judge it, but a citation probably wouldn't help in
that sort of situation.

>> 2) Even if it isn't, it doesn't matter all that much (not
>> potentially  libelous, etc.); and
> 
> So, now we have the same subjectivity as 1) along with a
> test:  If it causes legal problems, delete.  

If nobody has a problem with it then there's no problem with leaving it
in. Subjective but still quite useful.

>> 3) It looks likely that a source can be procured in the
>> future.
> 
> Back to subjectivity.  How does one determine whether a
> source is likely?

One just thinks it when one reads the information. eg., "The article
says Mr. Foo died in 1975. I don't have access to a library to check
right now but he was pretty famous so there's probably obituaries in the
newspapers if nothing else." That's all that's needed, one doesn't have
to actually look it up.

If someone comes along who _doesn't_ think a source can be procured then
he can go ahead and challenge it, and others can then dig up a source or
otherwise try to convince him that one exists. Just like with the first
point, above.

>> Sometimes an article that's lingered for months with an unreferenced 
>> banner will be proposed for deletion, since that's taken as evidence 
>> against #3.  But there are plenty of things that linger unreferenced, 
>> even though everyone knows they're basically correct---and you can even 
>> roughly verify them by doing some google searches---because nobody's 
>> taken the time to edit them into a properly referenced article.  A bunch 
>> of stubs of locations and figures of classical antiquity fit this 
> 
> So then it is no problem to put a warning that it will be
> deleted in x days if no sources are added.

You're very good at reaching conclusions based on comments that don't
seem to remotely support them, I have no idea how you got this from what
Delirium wrote. :)

The problem I see with this sort of hard-and-fast deadline is that
there's no pressing need for it but a very real chance that it's going
to remove stuff that could well have been improved instead if it had
been allowed to sit around longer. The sorts of stubs Delirium mentioned
as an example probably aren't on many watchlists or the focus of many
people's academic interests, so even if the sources are out there it's
going to be hard to force people to dig them up in only X days. It's
hard to "force" anyone to do anything on Wikipedia, it's a volunteer
effort after all.

> At this point, there seem to be 2 reasonable alternatives:
> 1) Create a time limited deletion warning permitting
> deletion upon expiration without reliable sources.
> 2) Create an indefinite banner that clearly states that the
> article/section is in a highly unreliable state should not
> regarded as accurate.

I've got no problem with the banner approach, though I think "highly"
might be overstating the issue.