Franc Zabkar wrote in
:

I confess that I find Google's search algorithm incomprehensible also.

+C
About 13,100,000 results --- OK

"C"
About 25,270,000,000 results --- eh?

-C
Your search - -C - did not match any documents --- OK

+C -C
About 255,000 results --- eh?

BTW, I, too, thought that the plus operator had been withdrawn and
replaced by quotes.


I think search engines are maybe all failing, because they all depend on
trying to filter the unfilterable.

Machines can't do this. When confronted by a page of 'posts' all cut crudely
out of a forum, snipped mid-word, no less, then pasted on a page with fake
names, only to fool Google into ranking it high so they can get people to
load them and see the ads on them, then even if Google use all kinds of fancy
heuristics, they will NOT eliminate this crap, and they aren't all that
interested in trying. They only do as much as to keep the masses from howling
at their incompetency. I suspect it IS a failure to grasp the complexity
adequately that causes them to return rediculous assessments of how many
pages have appropriate matches.

Giving us EXACT substring searches is the only sure way to prevent this, so
we can use OUR brains to filter what they evidently cannot and will not
filter. But they never give us that power, despite it being the easiest thing
code can do. They never will, because we are NOT meant to be given that kind
of power. We'd have had it long ago, as we do on our own machines, if they'd
ever intended to help us that much.

Tonight I found some quiet high SNR pages, mostly details of C and API
coding. I mostly got them from links they shared, not directly from Google,
once I'd got a hit mostly by luck. This was how the net use to be I think,
before search engines. You either knew where stuff was, or you didn't, and
web rings, links pages, kept a trail from one good place to another. If a
page was ****e, no-one bothered to tell anyone where it was.

I remember someone saying that search engines will never be a substitute for
a good library. At the time I didn't want to beleive that, I hoped the net
would be a better and faster source than any library, and I've known several
good town and city centre libraries to compare with, too. For a while it WAS
better, I could find more, better, and freer info on lasers and computers and
pretty much anything else than I ever found before. I think that's why it's
so disappointing now. The saturation has become so noisy that it's like
trying to watch TV through so much static it's hard to tell if the picture is
even in colour!

Those who hoarded good books were wise. There are several great sites on the
net, but I won't be relying on search engines to find them. From now on I'll
be saving links, and maybe asking the owners if I can download a lot of pages
for private storage, in case I can't reach them any other way. Places like
the 'Battery University', Sound.westhost.com, and such. Places that like
libraries had more answers than questions, until an enquiring mind came to
them. The internet is full of questions, mostly from UNinquiring minds.
Usenet is an oasis, but the web is becoming a desert. The more people fill it
the emptier it ghets, somehow.

I never thought I'd say this, but if there's to be a two-tier internet, bring
it on.