One of the things I've come across in blogging world, is that it is only polite, when somebody links to you, for you to link back to them. Previously, in the context of my experiment, this was not possible. Quite a few people linked to me while I was collecting data, but due to the fact that I promised anonymity, I felt I couldn't return any of the links, lest it become clear who has done my experiment.

However, I'm no longer collecting data, and with this recent BlogShares thing, some new people are finding me. And these kind people are linking to me. So it is only fair that I return the favour:

• First up, is someone I've previously mentioned, whose paper on blogs I had read, Torill Mortensen. Here we have a fellow academic with an interest in, amonst other things, Blogs.
• Secondly, another academic, Stephanie Hendrick, who hopes to study blogs in her PhD.

Go check these nice people out.

So I was using Google for vanity reasons earlier. I was looking up Blogademia to see if I was listed, to see if anyone else was using the term, and I made a shocking discovery...

This blog, my blog is listed on BlogShares. I don't really know what this means, but it certainly wasn't me. Apparently it was listed June 23rd this year. How? Who?

I don't appear to be worth very much, and I have no idea how to do anything about that, but it's certainly something interesting that I need to investigate further.

So finally I have some results. "Excellent progress" says my supervisor. "Finally!" says I.

I have indeed begun to analyse my corpus. I'm starting on some fairly straightforward linguistic stuff, which basically requires me to run my text through some software, and play with the statistics.

Oh...I should mention something on my files. Traditionally results of analyses such as I am conducting of average over an authors texts, so we had decided to split each subjects blog into the individual posts. I had forgotten this, and was already to go when it came back to me. This process of splitting the files delayed me slightly...not least because it has left me with almost 2000 individual files to work with. Which is a lot.

So anyway, my first stages of analysis are replication studies. A colleague did his studies on a corpus of e-mails, and I am looking to see if I get similar results. His first work was actually replication of the work of a well known academic in the field, so I'm currently doing replication². Apparently, either together or separately, this work may be publishable. Yay!!

So far I've just got a whole load of number, I still need to look at them closely to see what they actually mean. I also have to write them up...well I don't have to but I should. Which is a big problem for me: I like doing, not writing. I just want to get on and run more and more analyses and get more and more number. I don't want to take time out from getting somewhere to make lots of table and write lots of text explaining the statistics. I am aware, however, that eventually I will have to for my thesis, and it's good to get it out of the way at the time when I know what I'm talking about, and it's a tangible way to show others my work. So I will do it.

That's today's main task: spend time explaining the numbers I already have before I run off getting more. Hopefully this will help me get to grips with what it all means, and I can start reporting some results here.

Ok, so I'm just having a play with Blogger, to see if I can make my post look like the last couple, without having to copy and paste the HTML.

So, finally, the spelling is done. There were a couple of problems, such as unreadable UNICODE character popping up in files, and there were a lot of people who would occasionally just forget about the spacebar, but other than that, I think a little over a week is so much better than my anticipated 3 months of spell checking. And that was a really long sentence, the kind I really need to stop writing, because it doesn't look good in official things. So let me re-write it:

So, the spelling is finally done, with only a few problems. I encountered some unreadable UNICODE characters which had to be removed. Also, a number of people would occasionally forget to use the spacebar. All in all, it took little over a week to check all my files. This is a marked improvement on the 3 months I had calculated it would take.

Now, technically the re-draft is longer, but it's not one sentence, so it reads better, and is more formal. My formal writing is pretty poor, because I have a very informal style by nature. My supervisors are always getting me to re-write sentences, because I just love commas, and I put them everywhere. As awful as it would have been, they could have taught us a lot more of this sort of thing in school. I was terrible at English at school, but mainly because it was all about in-depth studies of Shakespeare and poetry, and I could just never see the point. So I blame the education system for my poor writing skills.

Anyway, what was I saying? Oh yeah, the spelling is done, and I'm almost ready to go. A quick intermediate stage before the formal linguistic analysis is correction of some informalisms using a proprietary system I have developed. And then...oh joy of all joys...I extract those, lump them there, run that, stick it in there, click this and vol-au-vent...results.

I probably won't be able to tell you in detail the exact results, because it will hopefully be publishable work, but I can tell you if my results are good or bad, and how they compare. It's so exciting.