Hopefully by now you’ve read parts one and two of the Wikipedia links series and are ready for the slightly more aggressive part three. I affectionately call this technique the Chumbawumba method because when my links get knocked down I put them back up again. Never gonna keep them down. Don’t pretend like you don’t know the song. This more aggressive method is not only great for keeping your Wikipedia links up but works just as well for many other sites. I would love to hear some brain storms in the comments for other sites that this would work for as well. There are more than you would imagine. Off the top of my head I can think of six.
This process will be ran from a timed cronjob (schedule server task) and will aggressively ensure that your links stick and stay on all your Wikipedia submissions. This technique will, once again, require knowledge of some form of server side language and will assume you have already finished and have a working version of Wikipedia Links Pt. 2. I will also assume you have been keeping a database record of all the Wikipages you have submitted to.
1. Write a script that pulls (LWP) all the edit pages that you have submitted your article links to. Parse out the actual article, the EXTERNAL LINKS, and everything below the external links.
2. Check the EXTERNAL LINK for your domain. This is easily accomplished through a simple regex command =~ /$domain/.
3. If your link exists than ignore the page and move on. If it’s gone then just readd it. Someone mentioned in the previous posts’ comments that they are in a link war with some of their competitors on Wiki. Their competitors are removing their links and adding their own and visa versa. I could not imagine a worse waste of time. Plus the links disappearing and reappearing does no one any good if neither one can’t get any trusted link age on the pages. I suggest just stomping them out for good by also parsing for their links and just automatically removing them as they appear. They will eventually tire of fighting your automated system and bow down to the king. I know this sounds mean and unfair but fuck ‘em if that’s the way they are going to be.
4. When the script submits back to Wiki be sure to use a solid and tested proxy. That way you can protect yourself from being banned. I’ve never heard of Wikipedia banning domains from the external links but I suppose it is possible. In which case if this happens just buy a second domain and do a 301 path to direct path redirect to your real domain. No big deal problem solved.
5. Set the script on a cronjob. I recommend setting it for once every 48 hours starting out. Then keep a log of all the times your script has to actually make a change. This will give you a good way to gauge the link loss. Then just adjust the timing accordingly. If you find your links never getting removed than go ahead and just set the cron to run once a week or so. If you got some competitors than set it to every 24 hours during nonpeak hours.
Dun dun dun dun…can’t touch this! Btw today is national gay song day. So don’t mind me. I’m just having fun with it.