On Thu, 2003-03-20 at 22:46, james duffy wrote:
One problem with relying on google. It still does not
carry all wikipedia
pages. ....
That's why it's a temporary emergency measure, not a permanent
deliberate change.
I've restored the regular search during off-peak hours (04:00-16:00 UTC;
20:00-08:00 US west coast; for the aussies, this should be 14:00-02:00
Canberra time if I'm reading my timezone chart right).
We should be able to dump the google crutch entirely soon. We've a new
second server on the way to spread out the work over two machines; a new
stable MySQL release is out which should be able to search faster (if
I'm to believe what I'm told) which we will be testing shortly; and load
is down a bit thanks to PHP Accelerator, which I've installed on
Marumani's advice. (It seems more stable than the Alternative PHP Cache,
which we had used off and on in the past, but which we had some problems
with and had to remove.)
And if those two pages don't exist according
to goodle, what other wiki pages does it not recognise and why?
It won't recognize anything that didn't exist when google last crawled
our site. It might not recognize things that had errors when it last
checked them.
-- brion vibber (brion @
pobox.com)