OUseful.Info, the blog…

Trying to find useful things to do with emerging technologies in open education

Twitter Gardening – Pruning Unwanted Followers

Some people like to keep their Twitter follower network under control, whilst others are happy to accept all-comers (ahem), but using the default Twitter web page to track of who amongst your new followers may be a spammer is not all that convenient…

So here’s the simplest of gardening tools for helping keep those unwanted followers under control: simple Twitter gardening tool; to see your own friends and followers, you’ll need to edit the URL. Click through the link, then in the browser address/location bar look for psychemedia, and change it to your twitter name….

[UPDATE – reminded of this hack by Brian Kelly’s August 2012 post on “unwanted” Twitter Followers, I noticed that the single page web app described in this post has long since rotted… here’s a stop gap: Twitter gardening tool, though I may take it off that URL in the next day or two… ]

The tool is very simple (I’ll post the how to later) and simply pulls in details about the most recent of a named person’s followers, along with a count of their number of followers, friends and updates.

The tabular view lets you explore the ‘quality’ of your Twitter friends and followers based on 3 metrics:

  • the number of their friends;
  • the number of their followers;
  • the number of their updates;

A form based interface may be incorporated one day – for now, you have to go via the URI, which looks something like this:

e.g. http://ouseful.open.ac.uk/dataTableDemo2.php?
u=YOUR_TWITTER_ID&p=1&min_fr=200&min_fo=600&sp=5

The URI arguments as follows:

  • u: Twitter username
  • p: results page (0 gives 100 most recent friends/followers, 1 gives the 101st to 200th most recent friends/followers, etc; if less than 100 people are displayed in any view, it’s because Twitter has blocked them…)
  • min_fr: min no. of friends for ‘Min Network Size’ view
  • min_fo: min no. of followers for ‘Min Network Size’ view
  • sp: min fr/fo ratio for ‘Spammers?’ view

To order the rows by column, simply click on the column heading (once for an ascending view, a second time for the descending view).

By default, the app will only pull in the 100 or so most recent friends/followers (actually – it may be less than that; each 100 is based on following accession (I think) so there will be gaps corresponding to personally or automatically blocked followers).

If you click on ‘Get Next 100 friends/followers’, the next page of followers will be pulled in and added to the table.

A couple of spam follower detecting heuristics are included. The “Spam?” view will display followers whose friend/follower ratio is greater than the specified number (you can set that value via the URI). Another view shows people with a network size above a minimum number of friends and followers (again, those values can be set via the URI). Simply viewing the whole table ordered by decreasing numbers of followers can also detect spammers.

To block a follower, clicking on the name link will take you to their Twitter page. If you’re logged in to Twitter, you should be able to block the follower.

Note that if you’ve loaded lots of pages of followers in, click out to Twitter personal page, and back to the Twitter Gardening page, you’ll lose all but the first 100 or so followers from the table. So a tip: right click on the link and open the Twitter personal pages in a new tab ;-) Work through the table, opening potential spammer pages in a new tab if required, then review the tabs one at a time…

Here’s the link again: simple Twitter gardening tool. Remenber, all you need to do is click through the link, change psychemedia in the URL to your Twitter ID, and refresh the page to analyse your friends and followers.

See also: Brand Association and Your Twitter Followers and associated comments for a discussion about personal Twitter network management.

Written by Tony Hirst

September 24, 2009 at 4:46 pm

Posted in Tinkering

Tagged with ,

11 Responses

Subscribe to comments with RSS.

  1. Tony,

    This is amazing, I love the whole Gardening metaphor, and I need to be weeded. But my only question is why would you block @klgoiiebgoqe?

    Jim Groom

    September 24, 2009 at 4:51 pm

  2. @jim Britney told me to… She was embarrassed…;-)

    Tony Hirst

    September 24, 2009 at 4:59 pm

  3. How does this compare to http://www.twitblock.org and http://www.tweetblock.org ?

    Jeffery Lay

    September 25, 2009 at 8:48 am

    • I don’t know – how does it compare to twitblock and tweetblock? The app is just a throwaway, the simplest thing I could think of that provided me with a useful context for demoing how to use the Google visualization API datatable and table view…;-)

      Tony Hirst

      September 25, 2009 at 9:55 am

  4. The answer to the question of what the relationship to twitBlock is would be that your fr/fo technique is ONE of the parameters that Tim uses. twitBlock.org is worth a look! It has a handy jScript bookMark that checks an individual and leads to the option to check all.

    BTW; I am not affilliated with twitBlock. Is it relevant that I tend to read fr/fo as friends/foes? I detest those antiSocial followers that only harvest for merchandising.

    iRobt

    September 25, 2009 at 12:32 pm

  5. […] in common, “meeting” in another context (face to face or electonically, using Topify, using Twitter-gardening, using Tweetdeck’s new followers column and when they joined vs how many they […]

  6. I’ve been looking for a way to clean my twitter followers that actually works. I had been “pruning them” manually and quite honestly that really sucks!

    Dessius

    January 14, 2010 at 4:05 pm

  7. […] causes of viral Twitter posts, applying heuristics for spotting Twitter spammers (as Tony Hirst has described), […]

  8. […] curate your followers to any significant extent (for example, blocking spambots, and not doing your twitter gardening), then your personalised search results may not be as highly tuned as they might be… […]

  9. I wonder whether it also makes sense to include a factor relating to the age of the account, and maybe the ratio of tweets to days since it was created?

    As far as hashtag communities go, might it also be worth doing a plot of the distribution of things like fr/fo ratio to get an idea of the maturity of the tag community members? (We can also get more refined and consider the number of friends/followers within the tag community etc, cf http://blog.ouseful.info/2009/09/09/personal-twitter-networks-in-hashtag-communities/ and http://blog.ouseful.info/2009/09/15/finding-new-people-to-follow-in-a-hashtag-community/ )

    Tony Hirst

    February 14, 2012 at 2:01 pm


Comments are closed.

Follow

Get every new post delivered to your Inbox.

Join 865 other followers

%d bloggers like this: