Discuss mailing list

[Calum McAlinden <calum AT mcalinden.me.uk>]

On 16 February 2014 21:22, Jon Hebb <somebodyrocks AT gmail.com> wrote:
> It's a great improvement on searching OpenNIC Calum, and I, as well as I'm
> sure many other, members appreciate your contribution and work.

Thanks everyone for all the feedback and encouragement!

> My only "issue" (which isn't really even an issue) I guess is that I'm used
> to search on modern search engines (say Google, Bing, etc.) and this one
> seems to be lacking a few things.....

Yes, I am aware of all the issues you mentioned and will continue to
make changes and adaptations to include these features. I have just
make a change so that a page with a domain that matches (or is similar
to) the search query carries 10 times more weight compared to the
keyword being just in the body/title/url. This should ensure that the
homepage comes up if the query is just for the domain. Over time, I
want to get rid of the non alpha-numeric query striping, and implement
Google-like features and language recognition.

On 16 February 2014 21:48, Daniel Quintiliani <danq AT runbox.com> wrote:
> This afternoon I submitted all of them one by one to seek.geek. Should I
> not have done that? I can narrow it down to four if you want, just
> thought you'd want more entries.

No, it wasn't the URLs you submitted (I think I can work out which
ones they were from the logs - seek.geek does follow redirects btw :))
I mean URLs like http://www.elbasurero.geek/ (and there are tons of
them, check http://grep.geek/?q=www.www.www&cmd=Search ) that offer no
content whatsoever.

At present, a bash script is trawling through each OpenNIC domain and
sorting them into categories:
- Suspected parked
- Web Server default
- Actual real websites
- Offline

We'll see the results of this tomorrow.

Also, on a side note, should I be indexing New-Nations domains? At the
moment I'm not. Do people see New-Nations as "part of OpenNIC"?

Thanks.
-- 
Calum McAlinden
http://www.mcalinden.me.uk