Traffic from somewhere…

April 25, 2005 by Oliver
Filed under: XHTML Friends 

XHTML friends suddenly had a lot a of traffic… Not sure where from though. But when I checked today the number of sites added jumped from 900 to 1100 and the number of queued sites went from 200 to 1000! I’m making it crawl them all now :)

Tags:

Comments

No Comments on Traffic from somewhere…

  1. Oliver on Mon, 25th Apr 2005 4:19 pm

    2. It was most likely from Photo Matt I now realise :)

  2. Nicholas Shanks on Mon, 25th Apr 2005 9:44 pm

    3. < ![CDATA[Yes, I was just about to say "Matt Mullenweg" :-)

    You should take a look at XFN Graph 1.1 (http://web.nickshanks.com/blog/archive/000023.html) which I updated on Thursday. It understands “me” links and has a better URL parser than your site seems to have (for instance, you list http://www.tantek.com/log and http://tantek.com/log as two separate sites). It’s open-source (GPL) so take what you like.

    You also are not parsing the HTML correctly. Some of my XFN links got missed, some links got assigned “me” or other values when they had no rel attribute (and were linking to other sites), or the rel attributes of different sites got mixed up.

    Anyway, I hope your site gets us more publicity!
    - Nick.

    p.s. I’m in the UK too :-) ]]>

  3. Sören Kuklau on Thu, 28th Apr 2005 9:31 am

    4. < ![CDATA[Heya,

    nice site (also love Nick Shanks's XFN Graph tool, but in the long run would prefer something that's not a stand-alone tool :) ). Nick already mentioned that there's some parsing errors; for example, with my links page ( http://chucker.mystfans.com/links/ ), it appears to consider myself an “acquaintance” ( # acquaintance : http://chucker.mystfans.com ).

    Finally, I’m unclear as to whether I’m supposed to have it re-crawl my entire list when I add links, which I do every few days as I (re-)discover blogs.]]>

  4. Oliver on Thu, 28th Apr 2005 10:08 pm

    5. < ![CDATA[Hi,

    Thanks for the comments. A lot of the parsing errors have been fixed, or at least worked around for listing duplicate site but some still exist in the database.

    I *think* I've fixed the problem with incorrect attributes. It was a regular expression problem; any link without a rel attribute was getting the rel attribute of the next link that did.

    As for updating, at the moment the site won't automatically recrawl any site already in it's database since it has enough new sites in it's queue to cope with. You can however use the crawl feature in the sidebar whenever you want (that crawls immediately; the queue constists of links it's found).

    One thing I immediately used from XFN Graph is recognising "dating" as an XFN attribute :P]]>