Tuesday, August 26, 2003
Don't you just hate people who don't update their blogs.
I shouldn't say that though. I've had a number of people respond to the "How do you define a blog" question by saying they don't like the word blog. Interesting.
Anyway, as could be expected, interest in my experiment has waned fairly quickly. I've yet to have a day of zero hits on the experiment page, but we are clearly down in the single figures. Yet still the results are coming in. Only 1 or 2 a day, but it's still something.
So what now. Well, with respect to my blog work, data analysis comes next. The first stage of this is to decide on a coding scheme for my data. Apparently this is quite important, because I don't want, at a later stage in the project, to have to come back and recode everything to include something that I didn't think about before. This is going to be tricky.
This is made even trickier by the various formats my data is in. Ignoring the fact that I have text files, Word files, and even XML files, just the different styles of HTML is going to be a bother. I should be able to process the markers though, using a Perl. Why Perl as opposed to Python? because as i told my supervisor, I've never even heard of python. I don't even know if python is what he mentioned, but i'm sure it began with a P. Anyway, I can use Perl to process text, so I shall do that.
The most important data is the text. The entries. The journal logs. However, text can be littered with non-textual elements. HTML code for italicising, for including pictures, links, etc. People also quote things. A lot. How, using simple text processing can I automatically NOT include any songs/poems etc in the text written by the author? I a, seeing the potential for a lot of manual work. It will be worth it, so as not to include non-author penned text, but it's gonna be a pain.
But before any of that comes some writing. It's six weeks till i offcially start my third year, so we have decided that I should use that time to catch up with all the writing I should probably already have done by now. So I have a number of things to write up:
1) Stylistics. This is straightforward (hah!) literature review writing.
2) Write up the work I did with a smaller corpus at the beginning of the year.
3) Write up everything I know about, and have done with blogs so far. This part maybe tricky, because there aren't many books I can really reference. A lot of what is written about blogs is written online, often in blogs. If anybody knows of anything useful about blogs that I can check out, please let me know. Thanks.
While i am writing this i shall be thinking about my coding and what have you. I shall keep you posted on all the exciting developments.
Tuesday, August 19, 2003
Hi, my name is Scott Nowson, and I am a PhD student in Informatics at The University of Edinburgh. I am currently studying writing style in blogs.
you can find out more about me, personally and professionally at my website, and you can email me right here.
That'll do for now.
Hits were for sure an all time high yesterday, and i'll tell you why: blogdex For those who don't know, which included me until this moring, blogdex is "a research project of the MIT Media Laboratory tracking the diffusion of information through the weblog community." Which means that they keep a track of very many blogs, and they look at the links these blogs contain. The front page gives you a run down of the top fifty things across the internet being linked today. And yesterday, i was on the front page it seems.
But now I have a quandry. I don't put links to anyone who has done my experiment. But if I link to the page bearing information regarding those person registered on blogdex who are linking to me, I am kind of doing the same thing. Not that linking me means one definitley did the experiment...but...i'm not sure. i'll consider this.
Which is something I was going to talk about anyway. A recent completer of my experiment had some questions about why I didn't link to people. Their main, and valid, point was that blogs are quite public, so why would anyone not want to be linked. Also, blogging is often about links, and these are normally reciprocal. you link me and i'll link you. So there are all these people linking to me, and they are getting nothing in return.
My argument was that I've told people it will be confidential, so I have to keep it like that, i'm sure I could get into trouble with the University ethics commitee if I didn't. And blogs may well be public, but the person behind them isn't always. I have data from people who write anonymously, from people who have expressly reminded me of my pledge to anonymity. And Whilst I would never reveal ANY of the information in the survey part of the questionnaire, the two (survey results and blog text) are linked as far as I am concerned.
What I should really have done is put a question in the survey asking if respondants wanted me to link to them. I could have had a big list on my website or something. So for know I won't link to my blogdex entry. You can find it if you want to, but then there are various ways you can find the people that link to me, I'm just not going to tell you who they are myself. This would be a good topic for discussion, if I had one of those commenting systems that I see on so many people blogs. Which brings me to my next topic of conversation...
A number of people when responding to my survey have said that they read this blog, and they wondered if I was going to keep it up. The plan was just during this "project" but this is my PhD, so I guess I could continue, and regardless, I will be doing work on my data (analysis, stats, etc) for some time. I did actually start up a diary of my PhD, long ago, before I evne started. Before I knew about blogs (meaning it was all done shoddily in basic HTML by hand) I started my diary of a PhD: I started with the application process, I was going to talk about getting funding, that sort of thing. I had grand plans that it would be the ultimate insight into ALL the work involved in getting a PhD, but I didn't really keep it up, it stopped, just as the one I did for my MSc project. I resurrected it as a place where i could keep all the details of my supervision meetings, but again that didn't last. I shall try and keep this up though. At the very least posting once a week so that I can say what I've been doing.
Of course people who ticked the "be updated" box on my survey will receive emails whenever there is anything worth saying.
The problem is, my blog, this blog, sucks. As you can imagine, I've seen MANY blogs. There are many that look like mine, just using the simple blogger templates, but there are so many out there that look so good. What's worse is that many of those people are considerably less technically oriented than geeky ol' me. Loads of people has blog rolling. I don't know what that is exactly, other than a list of blogs. I see a lot of RSS around, and it seems like it might have something to do with XML, but again I am clueless.
I had a look at the code for this place, and I'm really not sure what to do with it. I've said before that I want this page to look more like my own homepages, but I don't know if i'm capable of it. I want to have links. I want to have all sorts of exciting things. I came across a site detailing the 10 habits of highly annoying bloggers a while ago. I'm going to waste some time and space discussing these. Just for fun.
1. Bloggers who don't enable comments on their blogs.
I don't. i don't know how. Sorry.
2. Bloggers who rarely produce original content, instead simply aggregating links to other blogs that I already read.
So far, since I am talking about my work, everything is original content. Except this. ironic huh.
3. Bloggers who spend more time blogging about blogging than anything else.
Well dur. I guess this applies to me, because I'm blogging about my work on blogs. Meta-blogging. Sorry.
4. Bloggers who are FontBitches and don't care.
Well that's not me. I just use the font provided by Blogger. I have seen some teeny tiny fonted blogs on my travels. They're annoying.
5. Bloggers who don't provide any "about me" info on their blog, or pointers to it elsewhere on their site. I like to know more about the people I'm reading.
i'd like an about me bit, but until i know what i'm doing, you'll have to wait. Sorry.
6. Bloggers who don't provide a blogroll.
7. Bloggers who post excuses for not posting. I don't care if you're busy today. If you're not posting, fine. I'll just assume you had other things to do.
I might have done this. Thing is, it dpends on the type of blog. If It is a diary style blog, you are reading about someones day to day life. If they don't post for a while, you should be equally interested in why they haven't. Not sorry.
8. Bloggers who react but rarely act. Commenting on what other people say or do is interesting, but I'm annoyed by folks who never seem to have original material. (Yes, this is like #2 but it's not quite the same.)
Again, i'm original material boy. Most of the people I want to link to I won't for the anonymity reasons stated above.
9. Bloggers who don't provide obvious RSS links for their blog. Yes, I know that RSS auto-discovery is great, but not all tools do it.
RSS, schmaressess. Sorry.
10. Bloggers who have TrackBack but don't use it.
I don't use it, but i don't have it, so that's ok right?
well, that's at least 5/10. I'm 50% annoying. Which makes a change.
So, I need to learn a little bit more about making my blog, so I can make it all pretty, and have comments and blogrolls and all these trendy things. One thing I can do however, is have a temporary "about me" section...
Monday, August 18, 2003
Lots to say, so I'll probably make a couple of posts, just so it's easier to read. By way of introduction, i'm going to bring you up to date on the present, talk about the future, and some other things.
When last I checked on Sunday afternoon, there had been minimal action here at the experiment corral. However, there was a big surge last night and my inbox was once again full of data. And the website hit counter is currently on a big surge, today could be the biggest day yet. Which is nice.
But this is all good, because it's pretty much been a month. My supervisors and I decided that the experiment would run for a month, then we'd leave it alone and start the analysis. It did'nt take off quite as we had expected, it's harder to reach bloggers than we thought.
"There are 100s of 1000s of blogs out there, imagine if only 1% did the experiment. That would still be loads..."
Those were the days. So it has taken a lot of work, surfing for blogs, mailing bloggers in order to drum up business. It's taken too much work, I 've had little time to do any other work. the work i had planned on spending the summer doing. So we decide that i'd go on a final offensive last week, and just hope I could hit gold.
That's not to say the experiment is over. Far from it. DO NOT stop sending data...don't stop telling your friends. All this means is that I am going to be less in recruitment drive, and more of a middle manager. I'm still available to answer questions, and i'll still be dealing with all the data I get. And I'll still come here and tell y'all how it's going. I'm just not going to be emailing any more people. Today's surge was good. The constant flow of data that I'll be getting for the rest of the week...hint hint...will also be good.
Right...i'll be back in another post a bit later.
Wednesday, August 13, 2003
what an unprofessional thing to say. AND multiple exclamation marks?!?! how awful.
Anyway, got in today, checked my inbox, and it's chocka. It's packed with people having done my experiment. How the...? What the...? Not that I'm complaining but normally, I email some people, a third to a half of them do it, and one of two more get it from them. But this response level was simply unprecedented.
How did it happen. The computers were down when I got here, so I couldn't check. But now they are back. Cool. Of course, privacy and confidentiality are assured with my experiment, so even if other people are saying they did my experiment, I will not name them, so I can't tell you exactly what happened. But...someone I emailed on Monday has linked me. Got a good few links there. One of the people who came through that blog linked me, and from them my greatest number of links has come. More even, certainly in the first day short term than my mention on the BBC website. Wow. If you are reading this...and one of you certainly has been :D...thank you so much. You are absolute stars.
So is everyone who has done my experiment and/or linked to me. I show no favouritism.
Word of the day: YAY!!
Tuesday, August 12, 2003
Wow. It's been a whole week since I posted. Well...err...never mind eh?
So I was doing a whole other experiment last week. As you know...maybe...I am emailing bloggers to ask them to do my experiment, but I am also hoping that word spreads. Last week, i tried to let word spread. It didn't. I have one of those 'how many people are looking at my page things' and the graph kind of starting going down...fast. So more mailing for me. In discussion with my supervisors, who are quite keen to put a cap on the time i spend on the actual data collection...this makes great sense really...we reckon, step up the mailing campaign this week, and then leave it. If any more data comes in great, but don't go looking for it.
How can anyone who writes like me honestly be studying writing style. Look at the last sentence. Well, the one before that one. You know the one I mean. It's long, full of commas, elipses ('...'s) and probably could read way better. But that's writing for you...it doesn't have to be good to have style. I'd be a perfect subject for my experiment.
Only I wasn't blogging in May. And it's not a personal journal style blog. Do you know how hard it is to find bloggers who match these requirements? AND have email addresses?? Harder than you'd think.
Tuesday, August 05, 2003
What I really want to do, is change my template, because I have seen so many cool blogs, nice layouts, funky stuff, you know the drill, and mine is so simple. I'd like some links, i'd like a different background. I'd like it to have the same format as (oops, clikced the "post" button instead of the "URL" button. yeah, i know they are a long way apart, but you know...) my webpage. Sadly, whilst I do know a degree of HTML (see my webpage for proof) it's just a bit beyond me right now.
But maybe in the long term I'll get around to it. not least because someone (you know who you are) who just did my experiment, did so because they read my blog. This blog. And they asked if I was going to be keeping it up. I wasn't planning on doing so beyond however long I am doing this particular work, but since this is really the crux of my PhD at the moment, I don't see why I can't continue long into the night. So thank you.
Which is a point. I only read the email because I wasn't sure which experimental data it came with. (When subjects fill out my questionnaire, the results are emailed to me. The subject is always the same, and contains the blog name. Then I get an email with the blog data, assuming the person has sent it like I asked. some don't. Just so you know). I don't normally read the mails that come with the blog data, because i assume they don't say anything. This one did. So I should really read them, incase they have any questions or what have you. Stupid me.
Also check that the data they have sent me is in order. Because that is important. so I'm going to have to go back and do that. Which is another thing to do along with mailing more bloggers, because the rate at which people have been doing my experiment has dropped since I just left it to run alone. People ARE still finding it, there are a few blogs out there which are getting me many links. But it's not enough, so i'm going to have to try and track down some more bloggers to pester. Which is proving really hard. today, I are being mostly finding blogs that started in July, and blogs that are punditing things, blogs that are lniking, and giving comments on issues, as opposed to the diary style journals I am looking for.
So if you know anyone who has a blog and hasn't done my experiment please send them to me.
Does it seem like i'm desperate? How unprofessional. Actually, that makes me wonder...how much will my informal style have on my official research when it finally surfaces. Am i going to be less credible because of the manner in which i have written here? hmm.
Friday, August 01, 2003
Well. Hello there. Things seem to be slowing down. I had to do another manual mailing yestreday, which generated a good number of responses. The BBC link is getting me hits, as are links on other blogs. Thank you to those people. But please could more people do my experiment. If you are a blogger, and don't mind answering a few questions and giving me your May archive then please do my experiment. If you know any bloggers please tell them about my experiment.
do it for science.
do it for me.
do it for the kids.