On 6/8/06, Steve Summit <scs(a)eskimo.com> wrote:
And of course, if the various Search options all got
their "Do
you mean?" functionality just right, we truly would *not* need all
those redirects. Many of them exist not because a commandment on
a tablet said we had to create them, but rather, as a workaround
for the fact that readers do occasionally type in article names
spelled wrong, and we want to make it easier to find what they
want, and the software hasn't always been smart enough.
I actually mostly make them to prevent anyone else redlinking them and
then the same person (or someone else) creating a duplicate article. I
probably create on average 5 redirects a day just through random
browsing. If my first search term doesn't hit on the right article, I
make it a redirect.
Totally agree that the software "should" be smarter about resolving
search queries. Totally understand that the developers have other
stuff to do.
So yes, if the software can tell that a user who typed
A probably
meant B, and if it can implicitly redirect from A to B, we don't
need an explicit redirect from A to B, after all. And having the
As long as the implicit redirect works in all cases, which it doesn't.
software do that automatically is obviously preferable
to the
nuisance of creating and maintaining all those thousands of
explicit redirects manually.
Yep.
The remaining case is that if the software is
implicitly
redirecting from A to B, but there is some distinction between
A and B after all, such that a later editor wants to create a
distinct article on A, there must obviously be a way to do so
without being trapped by the same implicit redirect.
Yes. You're also alluding to the basic problem of "a" and "A"
mostly
being the same, but occasionally being different. It would be "nice"
if there was a way of making "Nice" and "nice" different articles,
but
for many good reasons, that's the compromise that was reached.
(Apologies if I'm belaboring the obvious here.)
No, you're helpfully reducing it to its most basic elements.
Steve