> From: Bryan Derksen <bryan.derksen(a)shaw.ca>
>
> Daniel P. B. Smith wrote:
>>> From: "Steve Bennett" <stevagewp(a)gmail.com>
>>
>>> I don't think deleting accurate, high-quality, unreferenced material
>>> is in Wikipedia's best interests. Asking for a source, yes. Adding
>>> sources, yes. But *deleting* good material? No.
>>
>> Unsourced material is not high-quality material.
>
> It can be, depending on the circumstances. This isn't something on
> which
> categorical statements can be made (or at least one can't expect
> anything close to consensus on such statements). For example, consider
> the articles where we imported masses of unreferenced material from
> the
> 1911 Encyclopedia Britannica. That's not even high quality stuff by
> today's standards but it's served as a good foundation for further
> work.
Such articles _are_ sourced. They're sourced to the 1911
Encyclopaedia Britannica.
When we source something to _The New York Times_, we don't worry
about whether _The New York Times_ cited _its_ sources. The reader
knows where the material came from: not from an individual editor's
head, but from a specific issue and page of The New York Times. They
can make their judgement about its reliability.
When we source something to the 1911 Encyclopaedia Britannica, the
reader knows it cames, not from an individual editor's head, but from
the 1911 Encyclopaedia Britannica, and they can make their judgement
on its reliability.
"David Gerard" wrote
> Yeah. The nice thing about a preload template is that it can guide new
> editors without being a straightjacket on experienced ones - it's just
> text in the edit box, after all.
I've always felt the empty edit box was a big plus. I would almost prefer a 'required field' for Sources with prompts for a new page.
Plus it would slow down making redirects to cut boilerplate text. I do a lot of redirects ... I do mean thousands.
Charles
-----------------------------------------
Email sent from www.ntlworld.com
Virus-checked using McAfee(R) Software
Visit www.ntlworld.com/security for more information
"Steve Bennett" wrote
> On 11/30/06, charles.r.matthews(a)ntlworld.com
> > Trouble is: I have checked that my own Erdos number is at most 4. Knowing that it is exactly 4, rather than 3, is equivalent to knowing a huge amount about collaborative papers ... which is a needle-in-haystack search.
>
> No, it's really easy actually. Go to Wikipedia and check out
> [[Category:Erdos number 2]]. If you've worked on a paper with any of
> the people in it, you're 3. Otherwise, you're 4.
You're putting an awful lot of faith in the completeness and accuracy of our own information ... which in this case changes or should change every time one of a whole bunch of people publishes a preprint with a new collaborator. (By the way, do you update ENs at the preprint or journal publication stage, which might be years on?) It's a fun trivia game, but nightmarish, and EN2 is really the maximum we should allow on the site.
Charles
-----------------------------------------
Email sent from www.ntlworld.com
Virus-checked using McAfee(R) Software
Visit www.ntlworld.com/security for more information
Bryan Derksen wrote
> If there were some way to make it easy to cull out
"more finished" stuff from the "less finished" stuff there might be less
pressure to delete work in progress in order to make Wikipedia as a
whole "more finished".
My post, to which this replied, was actually trying to point out something about our metaphorical take, also.
I think our quality control talk usually does go along these lines: skim off or cull the layer of worst articles; cream off the best and call them Featured, or whatever.
What's wrong with that? Nothing in there about hypertext. It's still per article, not speaking to navigation, orphanhood, incoming links as a metric and so on. Imagine a recently-created but much-needed stub article. One would hope the incoming links were noted before it was speedied for lack of sources.
Charles
-----------------------------------------
Email sent from www.ntlworld.com
Virus-checked using McAfee(R) Software
Visit www.ntlworld.com/security for more information
_______________________________________________
WikiEN-l mailing list
WikiEN-l(a)Wikipedia.org
To unsubscribe from this mailing list, visit:
http://mail.wikipedia.org/mailman/listinfo/wikien-l
"Daniel P. B. Smith" wrote
> Unsourced material is not high-quality material.
"The Battle of Hastings was the most decisive Norman victory in the Norman conquest of England." From [[Battle of Hastings]], and unsourced. This is certainly of higher quality than it might be, in that it doesn't imply that there was no other fighting.
Charles
-----------------------------------------
Email sent from www.ntlworld.com
Virus-checked using McAfee(R) Software
Visit www.ntlworld.com/security for more information
> From: "Steve Bennett" <stevagewp(a)gmail.com>
> I don't think deleting accurate, high-quality, unreferenced material
> is in Wikipedia's best interests. Asking for a source, yes. Adding
> sources, yes. But *deleting* good material? No.
Unsourced material is not high-quality material.
"The Cunctator" wrote
> Huh. Just because one can't think of a reason that it would be of interest
> to have a list of mathematicians of Erdos Number 3 doesn't mean that noone
> can. Wikipedia ain't paper.
Trouble is: I have checked that my own Erdos number is at most 4. Knowing that it is exactly 4, rather than 3, is equivalent to knowing a huge amount about collaborative papers ... which is a needle-in-haystack search.
Charles
-----------------------------------------
Email sent from www.ntlworld.com
Virus-checked using McAfee(R) Software
Visit www.ntlworld.com/security for more information
Ray Saintonge wrote
> The most effective solution for dealing with a bad thread is the onset
> of boredom. Eventually they all come to an end.
Actually, they don't, and trolls don't always go back under bridges. It only takes two folk who won't concede the last word.
Ray, where does this bottomless pit of generalisations come from?
Charles
-----------------------------------------
Email sent from www.ntlworld.com
Virus-checked using McAfee(R) Software
Visit www.ntlworld.com/security for more information
> charles.r.matthews(a)ntlworld.com wrote:
>> What is bad, to look at what WP:RS itself, is having material 'likely
>> I don't agree that a page of
>> college-level calculus, known and uncontroversial for two centuries,
>> should be deleted for the pedantic reason that it isn't referenced.
Obviously, what's needed is a middle course.
a) Most of the unreferenced material in Wikipedia is accurate. What
do I mean by "most?" 90%? 95%? 99%? Something like that.
b) Most of the accurate-but-unreferenced material in Wikipedia
_could_ be referenced. What do I mean by "most" here? A somewhat
smaller percentage, but still "most." And the amount depends on the
topic area. Yes, there is a substantial amount of material in
Wikipedia that is "original research" or original observation or
direct personal experience, backed only by the testimony of the
editor that inserted it. But most of Wikipedia's content is
verifiable. The editor read it somewhere, even if it was in a
classroom years ago or even if he or she doesn't remember exactly where.
c) Everything in Wikipedia should eventually be referenced or
removed. And by "eventually" I mean in a time frame shorter than the
"eventualists." Not like "Possible-Probable, my black hen/She lays
eggs in the relative When/She doesn't lay eggs in the positive Now/
Because she's unable to postulate how." But its taken years to put
the material into Wikipedia, and it will take a long time to get it
referenced.
d) So, the unreferenced material should be tagged. That calls the
reader's attention to the fact that the material is untraceable, and
its accuracy is hard to judge. Equally important, it also calls
everyone's attention to the fact that verifiability is policy, and
that it is taken seriously.
e) Once tagged, there should be no big rush about deleting the
material, but it should not remain indefinitely, either. How long?
Assuming that there's no specific reason to doubt the material,
months and months.
The _only_ objections to this I can think of is that that the tags
are ugly--which is true but susceptible to a technical fix--or that
we are not serious about verifiability and don't truly want to
restrict Wikipedia content to things that are supported by published
material.
It should also be noted that deleted material is not lost or
suppressed or destroyed or gone. It's in the history and can be
restored at any time if someone finds a reference. And in most cases
courtesy suggests copying the unsourced material to the Talk page to
call attention to the deletion and to facilitate others in finding
references if they want to.