jayjg wrote:
On 12/22/06, Ilmari Karonen <nospam(a)vyznev.net>
wrote:
jayjg wrote:
On 12/21/06, Daniel P. B. Smith
<wikipedia2006(a)dpbsmith.com> wrote:
You can't prove a negative, but you can
certainly say "his book is
not in the Cornell University Library" or whatever, and cite a link
to the search or a description of how to do the search. This doesn't
seem very different to me from a citation.
No, you absolutely cannot do that, for reasons eloquently stated
elsewhere. The claim that it is not in the Cornell University Library
is a novel conclusion based on your own original research;
I would disagree, and say that it is a claim made by the library
catalogue, and thus properly sourced to them. The claim may be one made
by omission, but, insofar as the catalogue claims to be comprehensive,
it is nonetheless an unambiguous claim.
But that catalog has made *no* claims; it can't, it simply exists.
Rather, some Wikipedia editor has made a claim *on behalf of the
catalog*.
By that logic, books don't make claims either, they simply exist.
A official library catalog constitues a very clear unambiguous claim by
the library that "our library contains these books." If the catalog is
described as a comprehensive one, it includes the further claim "...and
no others."
If the
catalogue does not claim to be comprehensive, or if there is
doubt about its accuracy, it may be better to phrase the statement
explicitly as "his book is not listed in the Cornell University Library
catalogue."
And who do we source that claim to? Zero0000? Daniel Smith?
We source it to the catalog, as published by the library, as it existed
on the date it was checked. Just as we would source claims about the
inclusion of a company on the Fortune 500 list to the list itself, as
published by _Fortune_.
What does
the footnote say: "Wikipedia editor Daniel Smith ran this specific
search on the catalog on December 11, 2006, and it did not return any
results"? Ridiculous.
<ref>[[Cornell University Library]] [
http://catalog.library.cornell.edu/
catalog]. URL accessed on December 11, 2006.</ref>
For bonus points, include the direct link to the search results. This
would be comparable to mentioning a specific page when citing a book:
recommended, but not required. For extra bonus points, include a link
to an archived copy of the results on
webcitation.org.
Assuming you'd searched through the cards
properly, one card hadn't
stuck to the back of another, or been temporarily removed for some
reason, or...
Books can have errors too, and it's possible to misunderstand them.
Does not mean we have to stop citing them.
Or would you
also consider the statement "IttyBittySoft was not listed
in the Fortune 500 list for 2006" to be OR?
Why on earth would one ever want to put such a statement into an
article? It's clear that the only reason would be to make some sort of
argument, push some sort of POV regarding IttyBittySoft. Why not just
stated the obvious facts, and leave it at that? "IttyBittySoft
reported annual revenues of $36 million in 2006". That's all we need
to say; let's not beat our readers over the head with our agenda.
Presumably because some other source (most likely IttyBittySoft
themselves) did in fact claim they were "one of America's 100 largest
corporations" or something to that effect. That would be absurd, of
course, but you'd be surprised what claims people might make. (See
f.ex. [[Office of International Treasury Control]].)
Or maybe an otherwise credible source confused them with Microsoft (and
never bothered to publish a correction).
--
Ilmari Karonen