Tailrank – the Attention Engine

 I owe Nathan Weinberg a big thanks for pointing me to Tailrank. He describes it as "like Memeorandum, but with a real search engine, ads and links to various other things."

There's one thing that Nathan doesn't mention that is the killer: it allows me to upload my OPML file. Memeorandum doesn't.

Once I did upload my OPML file, the relevance of the engine shot up at an almost infinitely higher level. It was almost magical to witness. Posts that I have reffered to on my blog this month rank highly (in case you are wondering - it doesn't know I'm the author of my blog, so the results aren't skewing the results for this reason.).

I'm going to play some more - what I'm hoping is that it meets the requirements I stated only a week ago:

I subscribe to people whose opinions and views I value, but when there is 'big news' (as per yesterday's Windows Live) or even 'little news', I want to do a number of things:

  1. know who is writing what about a topic
  2. navigate the discussions taking place (clustering / threading)
  3. find the smart thinking - valued by others
  4. define the topic I'm interested in
  5. filter by 'my A-list': by those whose opinion I value (that I'm subscribed to)

So far it appears to meet 1,2, 3, 4, and 5! Compared to the other services I mapped against the above requirements, Tailrank wins as an attention filter.

The magic is in the abillity for me to upload my OPML - it allows the Tailrank algorithm to be put to work on the RSS feeds I've been collecting, creating and pruning for over three years - my ever-so-valuable attention file.

Jason b has been playing with it for the last two weeks, he likes it too. Even Robert has uploaded his attention file.

Kevin Burton, co-founder of Rojo-Networks is the creator of Tailrank. He calls this class of application a Meme Engine, I like the term, however I think Attention Engine is more accurate - Slashdot, Digg and Memeorandum are about spreading memes - these are Meme Engines. They don't take into account your interests and who you value. Tailrank seems more like an Attention Engine to me - it promises to bring things to your attention that you want to know about based on your  OPML (attention) file. Your data, interests, posts, blogs and people that match your attention profile. Powerful stuff.

What makes it so 'rich' is not AJAX. None to be seen in fact. What makes it 'rich' are the number of pivots on the data and the views and results that emerge. You can see a 'recommended', 'everyone', 'social' or 'local' view and can cross-tab against date, search term and tags. These 'everyone' to 'local' filters appear to scope the results in and around your attention file - 'everyone' allows the inclusion of the network that touches your OPML file feeds on the one extreme, while the 'local' view seems to zoom in only on your OPML file at the other extreme. What wonderful idea. There plently of other things to toggle too.

I love it. Tailrank went live only today. I'm sure I'll hear more positive feedback before the week is up. Well done Kevin!!

PS, just before posting this I saw this post by Kevin, referring to this:

"Seems like Tara and Alex have been having some fun.

They figured out how Memorandum works and hacked it so that they could appear on the site (thus getting more traffic).

...All existing systems have suffered from this. Back in the good old days Blogdex and Daypop would get hit with spam every once in a while.

That said, TailRank was designed with spam prevention in mind. We don't have any of this technology currently deployed on the site but in the next week or so I'll be able to sit down and write the code. This should prevent Tara and Alex's spam attack within TailRank (sorry guys). "

I won't even try Kevin 😉


Comments (5)
  1. Sweet! Thanks for the great feedback.

    I’ll probably blog about this later tonight/tomorrow.

    WRT #2.. I think we could do a better job with threading. I’m working on a new threading system which I’ll probably be deploying sometime next week. This way when a meme has a tighter cluster you’ll be able to manage it better.

    Thanks again!


  2. Scott says:

    YES, this is what I was talking about during the OPML breakout.

    One feature that would make it even better. Rather than uploading an OPML file, let the user point to a url containing an OPML file. 🙂

  3. MSDNArchive says:

    Scott: a subtle point you make, but v.important – quite right!

  4. Jason says:

    <P>Alex, thanks for my first ever non-spam link. I knew I was getting the hang of this blogging stuff. At any rate, check out my post here&nbsp;: </P>

    <P><A href="http://spaces.msn.com/members/geekspeaker/Blog/cns!1pz3z7wnAJbmfD9ANOMBREEw!310.entry">http://spaces.msn.com/members/geekspeaker/Blog/cns!1pz3z7wnAJbmfD9ANOMBREEw!310.entry</A></P>

    <P>&nbsp;&nbsp;about the limitations of OPML when used in an "Attention Engine". It’s not a cure all, and this represents an area for you folks at Microsoft to innovate the specification I think. Talk to ya! -Jason</P>

Comments are closed.

Skip to main content