The Next Web

Twitority’s Jon Wheatly about the hype and debates

Earlier this week, Loic Le Meur asked for authority-based Twitter search. Jon Wheatly, a young coder from the UK, only needed 12 hours and numerous cups of coffee to make this possible. He developed Twitority, with which you can search within tweets from people with different levels of authority (based on how many followers you have). A hype was born. Almost every major tech blog picked it up.

Home › Jon Wheatley
Jon Wheatly

And now we report about it, but not without giving you some background. A story about a young enthusiast like Jon needs more than just a discussion about vanity and the value of Twitter authority.

In an email, Wheatly described the hype: “Things have been pretty crazy since we put the site online on Sunday. After Loic blogged about us the site started getting pick up by other bloggers and just snowballed from there.”

He also responds to some complaints about the somewhat slow search: “Because of the limitations of the Twitter API we were forced to hack around it and scrape the follower numbers for each user, per search that is performed. This is why each search takes so long. It really isn’t ideal but it’s the only way a third party can build any kind of “authority” based search. The server was really moaning last night but it survived, just.”

And what about the debate about the idea behind of Twitority?Erick Schonfeld for example, argues that not the number of followers matters, but how many times someone gets retweeted. Wheatly: “The debate is still raging about whether or not follower numbers are a good way to measure authority and honestly I don’t know. This was just something that Loic (and numerous other people) wanted so we decided to see if we could make it happen.”


  • sixbit
    Or you could do twitter rankings like the app store rankings, based on current volume/rate of new followers :-).
  • Another method of working out if someone has more authority will be if they are grouped a lot in the new twitter groups.

    One thing I can get from crowdstatus data is how popular people are by the amount of crowds certain people have been added too. I will make these figures available when I get the update up and running.

    Also wouldn't it be quicker if they cached each users data when they are queried? From my experience its better to hit twitter as few times as possibe. I don't think it would effect the results if the user meta data is just updated every 24 hours from the last time it was requested.

    I think we need a twitter developer network because the google group does not cut it for passing on tips etc.
blog comments powered by Disqus
 


TwitterCounter