[Search-l] (no subject)

[Peter Burden]

On 10/01/2008, Stephen Carr <stephencarr at blueyonder.co.uk> wrote:
>
>  Hi again,
>
>
>
> Well I'm still not sure how to reply to a thread so here I have cut and paste one reply to my request with my answers, I still have not seen a good reason for not having a DOMAIN name (NOT URL) wildcard search (along with extra search words if possible) is this really that bad an idea? Am I the only one that thinks wildcards will be a GREAT help finding what your looking is some instances?
>
>
>
> Forget all the techno jargon, all you need do if the user types www.help*.com is first return a list of ONLY those top level domains that do not fit this criteria then continue and apply any standard search parameter!
>
>
You are talking about a searching for a domain name. Search engines do not
(normally)
build and maintain searchable lists of domain names, they build and maintain
searchable
indexes of document content. I don't think any SE operator has felt the need
to provide
domain searching, you could try the search facilities that many domain
registration sites
offer, but I don't recall seeing one that offers anything other than exact
matching or possibly
looking in a limited set of TLDs.

Netcraft are currently reporting some 150 million domain names. I don't know
of any
way of organising such a list for efficient (meaning 100's comparisons at
most) searching
using regular expression (or wildcard) matching. Maybe somebody does.

[Searching for an exact match is a different matter, there are many
techniques that
can reduce the computational load to 100's or fewer comparisons.]

I've sometimes thought that regular expression matching would be a nice
advanced
search feature but the efficiency problems are so horrendous that I've never
seen any
SE offer such a feature - not since about 1995 anyway.

> Thanks
>
>
>
> P.S. Would also love it if anyone can tell me how to reply to a thread instead of keep creating another!
>
>
All I ever do is reply in the same way as a I reply to any other e-mail.  I
use either Google's web mail
or the excellent Thunderbird programme. I notice that you are using
Microsoft Outlook, that should be
OK, you might want to check the settings or try another e-mail client.

Stephen Carr wrote:
>
> >* Well not sure how to move this along any further or if anyone is *
>
> >* remotely interested in getting this feature included in the Wiki-Search *
>
> >* engine*
>
> >* *
>
> >*  *
>
> >* *
>
> >* All I do now is the following*
>
> >* *
>
> >*  *
>
> >* *
>
> >* Use of wildcards for domain name searching is not available currently*
>
> >* *
>
> >* I would use it as it would help me*
>
> >* *
>
> >* I believe it will get more and more use over time as more obscure domain *
>
> >* names appear*
>
> >* *
>
> >* The addition of wildcards and phrase searching really is a powerful filter*
>
>
>
> >>>>>>>>>The danger is that someone will decide to search for something like
>
> >>>>>>>>>'m*.com' or 's*.com' and end up with millions of results.
>
>
>
>         That's right, just like if I was in a folder on my machines with millions of files starting with S and did a wildcard search like s*.*
>
>         It should bring back millions just like when I type "test" in google it requires 900,000
>
>
>
> >* This could be expanded to all sorts like email addresses searching and *
>
> >* many more*
>
>
>
> This would raise spam and privacy issues.
>
>         If an email address is published then it's fair game as it is now, however I would not be bothered if it was only domain name wildcard searching and nothing else
>
>
>
>
>
> >* If anyone wants to reply go to my website and email me, forgot what it's *
>
> >* name is but it had "card" near beginning, a "-" in middle and ended in *
>
> >* ".co.uk", but I'm sure you will find it using "inurl:card inurl:- *
>
> >* inurl:.co.uk" as Google only returns 325,000 possibilities  :o)*
>
>
>
> I only see 138 possibilities here for 'card-*.co.uk' and 364 for
>
> 'card*-*.co.uk' and  but that's on a limited list of about 3.77M .co.uk
>
> domains used for hoster stats tracking and comparison with
>
> com/net/org/biz/info/eu/mobi/ie.
>
>
>
> Not sure what ur trying to say here?
>
>
>
> The problem with .co.uk and many other country code TLDs in Europe is
>
> that the registries do not permit zone file access. Thus websites would
>
> only appear if they are linked from others (the fundamental flaw of a
>
> link based approach to search).
>
>
>
> Not sure you get the gist of what I'm proposing, I'm not interested in country code's .co.uk just wildcard searching on do
>
>
>
> Regards...jmcc
>
>
>
> No virus found in this outgoing message.
> Checked by AVG Free Edition.
> Version: 7.5.516 / Virus Database: 269.19.0/1216 - Release Date:
> 09/01/2008 10:16
>
> _______________________________________________
> Wikia Search mailing list
> http://alpha.search.wikia.com/
> Change options or unsubscribe:
> http://lists.wikia.com/mailman/options/search-l
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.wikia.com/pipermail/search-l/attachments/20080111/45cd7b72/attachment.html