> I think the concept's ok and I like the interface, which is very minimal,
> but the results and the way to search is not quite there yet ;-)
>

You're more than charitable. :-)

When you aim to topple the planet's most efficient money making apparatus
disguised as a search engine (which Cuil unfortunately positioned themselves
as, in hopes of generating buzz), the simple logic says you must do what
Google does, do it better and then offer stuff Google does not and, more
importantly, cannot. IOW, if Google can look at what you do and spend a few
months to emulate it, plus add their insurmountable advantages in scale and
distribution, you're toast. So the concept is not OK, if you're not running
an experimental charity organization.

As to results, things get weirder. I've read that the reason why people get
different results from Cuil for the same search terms executed at different
times is that their architecture is constructed on knowkledge domains. For
instance, there is a server just for, say, astrology, where all appropriate
search queries go to be processed. But when that server is down, as was the
case for many such servers on their launch day, all queries for astrology go
to the general server pool which, apparently, doesn't have the same chops to
process such requests. So you get inconsistent, unrefined returns
contaminating the top results.

For some reason, notions such as massively parallel processing, sharding,
shared-nothing architectures, etc., escaped the attention of cool folks at
Cuil.

Fail, written all over.

-- 
Kontra
http://counternotions.com
________________________________________________________________
Welcome to the Interaction Design Association (IxDA)!
To post to this list ....... [EMAIL PROTECTED]
Unsubscribe ................ http://www.ixda.org/unsubscribe
List Guidelines ............ http://www.ixda.org/guidelines
List Help .................. http://www.ixda.org/help

Reply via email to