On 13/03/07, Thomas Dalton <thomas.dalton(a)gmail.com> wrote:
Which is why
any correspondence we use is footnoted as such, and filed
with the Foundation. You don't have to know the background to be able
to tell that when we have a footnote saying "correction from subject,
filed at X", there is probably a reason why it doesn't correspond to
the public record...
But then, the source the actually WMF, and WMF is not a reliable
source. Why should anyone trust some random person on OTRS to have
correctly interpreted the email and copied the information across
without making any typos, etc?
a) No, the foundation is *not* the source. It is simply somewhere
which happens to archive the source for the purposes of using it to
write an encyclopedia. The source is the, well, the source; the person
who wrote to us with the correction. (I write articles and give the
book I worked from as the source. I don't give the library I read the
book in, which seems to be the appropriate analogy here...)
b) Right now we trust thousands of those random people to write the
entire encyclopedia, interpreting and synthesising like crazy without
even requiring the minimum sanity checks we'd impose on the "handlers"
here. I don't see how entrusting them with turning "good article,
thanks, but I got divorced two years ago" into a correction is
dangerous.
If it's complex, ask them to clarify.
The whole point of citing sources is so
readers can go and check for themselves to confirm it's right - citing
a private source is pretty much useless because it's really just
saying "I know I'm right, just trust me on it." Just because I'm the
foundation doesn't make me trustworthy.
*One* reason to provide sources is so that people can go and
double-check, but it's a minor one. By far the most common reason is
to indicate the reliability of the information, what kind of accuracy
and currency you can expect from it. In this case, we'd be attributing
We're not saying "we should institute a method for any random person
to make unqualified, unassailable, assertions". We're talking about
letting *the subject* point factual errors out to us. If we're
automatically assuming that subjects are untrustworthy, I look forward
to banning the use of autobiographies and corporate histories as
sources.
--
- Andrew Gray
andrew.gray(a)dunelm.org.uk