Re: Query third party search engines

Richard Wheeldon <richard_at_rswheeldon.com> wrote in message news:<3B2D0BFD.15FB_at_rswheeldon.com>...
> Jitendra Agrawal wrote:
> > Are u talking about meta search engines.
>
> Both about traditional meta-search engines, i.e. those
> which request answers from other interfaces on demand,
> and those which try to mine as much information out
> of an interface as possible for indexing purposes.
> The paper to which I referred deals with the latter,

Hi,

One of the concept I found amazing in this paper, is that Stanford tried to get information from ANY web page, with only one crawler!

Personally, my target are embedded systems... Even if computing power on embedded systems is getting greater and greater, I do not feel fuzzy analysis of a web page by such systems affordable.

So, I see two solutions in this sense.
. First is to reject the computing on an external server. . Second is to focus on particular web sites, using single crawler for each of them.

Of course, encoding data in a common way on all sites would be more helpful. But this means common protocols are needed. And this is the background of Telematic Servers...

However, user doesn't seem to be ready to pay such big amount of money, for accessing data which are free of charge. Nevertheless, server would be necessary, even just for updating the crawlers upon web sites changes on all the embedded systems distributed all over the world.

Romuald Received on Sun Jul 22 2001 - 01:31:32 CEST