Monday, January 31, 2005
Go Go Go
So I'm not sure if I mentioned, but I'm currently undertaking to write a conference paper. As most academics will be aware with the summer conference season just around the corner, the New Year submission deadline season is upon us. So I've been working feverishly to pull somethings together.
Ultimately, this is a good thing, because my work rate has been really good these past two weeks, I have made good progress, I will have something written, and I will hopefully get a publication out of it (in conference proceedings). Of course, on the downside, I have been neglecting the work required for my income, and so to make amends, I will have to focus pretty heavily on it for February.
I am of course, with all this progress, put in mind of my recent post concerning publishing results in blogs. I have some more results that are quite interesting, but I don't know if I can share them. Of course, some of my results kind of go against what I was hoping for, but that in itself can be a result. Of course, should I make the conference, then I can certailny let you read the paper from it, as it shall already be published work. What I can say is that I am attempting to put blogs in the context of cross-section of writing styles. Where do blogs sit? Only smarties have the answer.
Friday, January 28, 2005
That's a Lot of Code
I'm so proud of myself. I'm sure there are tools out there that you
can use, but I decided to write my own code to get what I needed from the
BNC. Not only that, but since I knew the code would be big, I wrote it with good programming form...I tested it as i went along and built it modularly.
Coming from a programming background you would think that would be second nature, but no. But this time I did it, and it saved me a lot of debugging time at the end. Every time there was an error, there was only so much it could possibly be, and they where all easily fixed. It took a couple days in total, and last night I ran it and examined the results.
Quite disappointing. A few things didn't print, so I made sure they were ok and ran again. Still results weren't impressive. I mean, there was kind of the pattern I was looking for, things were very roughly in the right place, but not everything was as I had expected, and with such poor strength. It was disheartening.
Then I spotted an anomoly. Where the should have been 5, 7 and 3 files, the results were saying there were 5, 12 and 15 files. The counter wasn't resetting. That could be it I thought. Turned out, none of the important counters were reseting. I had a little play, and on a small set, results improved significantly. Thank goodness. The programme is running for what will hopefully be the final time. Here's hoping it's everything it's meant to be.
Tuesday, January 25, 2005
No YOU Tool
So it's been a busy day learning about tools. Parsers, taggers, interfaces...all such fun. See it's all part of the next stage of my work, and I'm looking outside to a reference corpus...a la the British National Corpus. As with much research, I'm not sure I can really say what I'm doing with it, but that leads me onto a new thing...let's link and discuss...
So according to an article on the BBC, academics are really taking to blogs. Mostly they are being used as tools for teaching and research and sharing of information, much like this blog. Of course, no mention of those of us diligently studying blogs themselves. pfft. The article is all about Universities encouraging students and staff to blog. Anyway, they come with warnings:"This type of technology is very open and easy to instigate and that often means in the rush to use it, the bigger questions on the most effective ways to use the technology without creating legal and reputational issues for the institution are forgotten or end up being asked too late."Of course, the biggest worry of blogs is that someone might say something naughty. I just think that is a bit strange because staff and students have been able to make potentially reputation destroying webpages for years. The part that more concerns me is the publishing of sensitive information.
Not just sensitive I suppose, in the sense of telling the world your research ideas, but many journals and conferences do not accept previously published work. So, if one was to post their results, along with say an informal discussion, does that count?
One could argue that since this has not been a problem with personal/academic webpages, it would not be the same with blogs. However, who really updates their webpage every other day? An academic probably changes their webpage only to add publications, or brief summary details of new project. In blogs, they are being encouraged to express their thoughts far more regularly. Or am I just being paranoid.
I think this is quite interesting, but it's a great shame that you can't leave comments. If you are inspired to say anything, mail me, and I'll post it.
I just wrote an long and well thought out post...and it was really good...and Blogger just ate it. Blogger is absolutely infuriating sometimes. I though I'd be safe. I though it would be ok. PFFT!
Luckily, by the magic of the Unix middle button...I think i still have it...let's see...well...none of the formatting...just the raw text...this may take a moment...
Monday, January 24, 2005
Long Time Coming
So yes, it has been a long time since I posted anything, but not this long surely? Was October really the last time I posted? I could have sworn I posted at least once in November. I was sure that wasn't my last post.
And where have my comments gone? Does that mean my template has changed? Surely not since my page counter is still there. Does it mean the people that did the commenting tool have stopped...maybe. It's been so long I don't remember who I went with. That's annoying. I don't really have time to sort it out, but they are handy things to have.
Anyway, I just wanted to post and say that things are still progressing. It's been a busy time of late, and it doesn't let up. I'll definitely make more effort to post, once I work out what is going on. It's not like I can even ask for suggestions....pfft.