|Elbow – Grounds for Divorce|
|Found at skreemr.com|
(play song while reading)
Technorati has always had it's problems, but like in any relationship, you can overlook them for a while. However, they've been adding up lately, and I think we've reached the stage of irreconcilable differences. Here are the grounds for divorce:
- It's broken – speaking just as this user, I know it doesn't work reliably. Exhibit A) – my authority hasn't changed for 20 days and even if I ping it manually, it still says 20 days ago. Exhibit B) – there are several people who I know have linked to me that don't show up in Technorati. As it works by calculating the linked references to your blog, this renders it pretty hopeless. Exhibit C) – my authority when it was working seemed to fluctuate randomly. There was a two months period where it dropped from around 160 to the current 79.
- It is pre-Twitter – Technorati works on references from other blogs. The problem is that a lot of the referencing has moved on to Twitter (and Friendfeed, etc) now. A lot of that quick linking people used to do in short blog posts they now do via microblogging. And Technorati isn't picking these up (is this because of the use of shortened urls, or just that it doesn't look at Twitter accounts?). It's therefore operating with only half the information.
- It lacks granularity – there isn't even an education channel. By lumping us all in together a small, individual blog is 'competing' with output from large companies with teams of writers. Take a look at the Technorati top 100 and you'll see that they are mostly either online magazines or professional bloggers. This is fine, but it needs a layer of categorisation over it to become meaningful.
- It measures the wrong thing – this isn't really Technorati's fault, more an issue of the blogosphere. I would like to do, or find, research that confirms this, but my hunch is that what people link to is different, or at least not necessarily the same as, what they find interesting and valuable. People often link to funny stuff, or technical tips. Some of the posts that I have gained the most links to have been short ones that say 'I found technology Y, it looks interesting.' Others might link to this and say 'I found this (via Martin)…'. But your 2000 word discussion on Wittgenstein gets less linklove. However, it may still be useful and relevant to people. This is particularly problematic for higher education where you want to promote the Wittgenstein skits and not the tech trainspotting.
- It's subject to gaming – all too easily really. As it measures link references the best thing to do is to engage in a real ding-dong fight with someone, preferably someone with a higher authority than you. This gets good links.
Now, you could say, 'who cares?'. Tony often chastises me for mentioning Technorati as he thinks it's all about ego (before pointing out that his rating is higher than mine ;)). But my interest has always been about whether such bottom-up metrics have a place in the world of the digital scholar, and Technorati has been the exemplar for these. If we have to consign Technorati to the tech bin then the question arises: does this mean any such approach is fundamentally flawed or do we need better metrics?
Doug Clow (who is rapidly becoming a digital scholarship guru) gave a talk recently where he concluded that we can really only rely on peer review. I don't fully agree – I think as scholars peer review will be the main component when assessing a colleague's contribution (say, for promotion), but given all the rich data we have online, surely there must be some reasonably reliable use we can make of it to at least add to the mix. I think PostRank, for example, does a better job of measuring your impact, although some of the same criticisms still apply.
So, I won't give up on metrics completely, but me and Technorati – we're through.