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).
There’s even a cool screencast (I love these things)!
Mashable takes a very pessimistic view here and asserts that algorithms don’t work:
As you may already know, my feeling is that humans are generally better at editing than algorithms, but by the same token you could say that the main algorithms in use today (Google PageRank, Memeorandum, Google News) are largely based on human decisions, where a link generally counts as a vote. At the other end of the spectrum, you have sites like Digg and Reddit, which are entirely edited by humans in a distributed way.
It’s also possible to game Digg and Reddit this way. You just have to create a lot of accounts (which is why Digg has captchas).
TailRank is both. We take into consideration both human and machine factors. We use our robots to crawl the web looking for interesting patterns and them promoting then to the site. Once on the site users can share a story with their friends. The more people who share the story the higher the ranking will become.
The difficulty is that there are often edge cases where your algorithms will fail and humans have to come in and patch things back together again.
Tara and Alex seem to have found one with Memeorandum.
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).
-
1
Trackback on Nov 8th, 2005 at 9:55 pm
Tailrank - the Attention Engine
I owe Nathan Weinberga big thanks for pointing me toTailrank. He describes it as”like…
-
2
Trackback on Nov 8th, 2005 at 11:18 pm
Memerorandum hack - the return
This is getting embarrassing. The Memerorandum hack story has returned once more to the make thefront…
-
3
Trackback on Nov 9th, 2005 at 5:51 am
Examining traffic from Digg, Slashdot and memeorandum
Darren Rowse over on Probloggerlinks to a post by davak at Tech-Recipe that analyses traffic to blogs that comes from Digg, the social bookmarking site.
The summary of the post is that:
1. Digg users do not click ads2. Digg u… -
4
Trackback on Nov 11th, 2005 at 6:26 pm
H2.0 and ‘the party to end all parties’
It is with dubious pride that I announce that this blog has joined the new Hack2.0 Workgroup. The…
Leave a Comment
-
Spinn3r News- Cornell Researchers Launch Memetracker Powered by Spinn3r
- Spinn3r Sponsors 2009 International Conference for Weblogs and Social Data Challenge
- Spinn3r 2.3.1
- Spinn3r Hiring Director of Sales
- Feed Update Protocols and SUP
- Spinn3r 2.2.5
- Spinn3r 2.2.3
- Spinn3r 2.2.2 Escapes From the Laboratory
- Spinn3r Hiring Senior Systems Administrator
- Spinn3r 2.2.1 Released
-
Subscribe
-
Flickr
More Photos
November 9, 2005 at 12:50 am
Kevin, your robot needs to judge human intent to handle this kind of spam. It needs to peer into the hearts of men and women. Good luck!
November 9, 2005 at 12:53 am
Gabe…
It doesn’t.. you’re thinking about the problem in the wrong way :)
Besides.. who says my robots can’t pass turing tests :)
Kevin
November 9, 2005 at 7:36 am
“The difficulty is that there are often edge cases where your algorithms will fail and humans have to come in and patch things back together again.”
That’s the point I was trying to make. I’ve clarified that in the comments:
http://mashable.com/2005/11/08/hacking-memeorandum-more-proof-that-algorithms-dont-work/#comment-206
I probably shouldn’t have quoted that previous post, since that makes it sound like there’s some kind of distinction to be made between humans and algorithms. It’s more a case of stepping in when the algorithms don’t work so well.
Anyway, I’m off to give Tailrank a spin. Ciao for now!
November 9, 2005 at 10:22 am
See, Gabe has it right…there was nothing malicious…just a little experiment. ;)
Truly, my alterior motive was to knock that other business off of the radar of Memeorandum with a much sassier, hackier, thus geekier impression of myself to the Memeorandum readers. ;)
And…to tell you the truth. I had no idea it would actually work!
November 10, 2005 at 10:42 am
some suggestions:
1. drop all the ads. There are way too many of them and it looks awful. You aren’t going to make enough from them at this stage anyway, and they clutter the page. Wait until you actually have the traffic and content to suppor them, and then use them sparingly. Same goes for the Amazon block.
2. don’t allow URLs to be titles. First of all, it’s breaking your template and second of all, it just looks weird regardless
3. Clean up the design. Right now it’s a clusterf*ck. Once it’s clean, make sure that the important stuff is emphasized. Right now, all the links are the same color and font-weight, so it’s hard to tell where to click and what’s important.
November 10, 2005 at 10:51 am
Who is this suggestion for? TailRank?
> 1… ads.
Right now I’m experimenting. They might go.. they might stay. I hear your point about not adding them at an early stage.
> 2. don’t allow URLs to be titles.
This is a bug.. it’s going to be fixed today.
> 3. Clean up the design. Right now it’s a clusterf*ck.
Yeah.. I’d agree. I think it’s OK just not perfect. I’m not a designer. That said we’re looking for someone good to do a refresh…
I hear you about the emphasis on ranking. We’re still experimenting.
Thanks!
Kevin