hit counter

I'm currently working on a DPhil in HCT at the University of Sussex. This section of the website is for an on-going 'learning diary', for me to write my thoughts and notes on various courses and my thesis.


Dreaming of questions

A couple of weeks ago I had a meeting with my supervisor, who asked me where my research was going (probably not unreasonably!). After I'd burbled at her for what felt like an age about all the different stuff I was reading, she sat back, and did the "hmm" noise, that I know means I've not exactly done something wrong but she's about to suggest something. 

She suggested that I had a few different angles going on, and that maybe I needed to focus some more. She then suggested a way of getting that focus: write my dream thesis abstract. Basically, I had to imagine myself at the end of my DPhil, and write down what I did (in general terms), what results I got (assuming it all went well) and what that meant. 

So, I sat down and tried to remember where I'd started from on my reading odyssey. 


(This is how I think best. Pen, paper, arrows, circles. It works for me.)

From there, I came up with a short little abstract. I emailed it off. I waited. I was told to put more detail in. I thought a bit harder. I wrote a longer abstract:

"I looked at group communication in small, newly-formed groups in a massively multiplayer online game (MMOG) environment.

I did a study to examine elements of communication style used in the group formation stage, including use of paralanguage (e.g. smileys), in-world vs. external references, self-disclosure by individuals and the development of shared terms. I concentrated on text chat amongst the groups in both 'social' mode and when they were 'on-task' e.g. conducting a raid. I looked at how those communication styles contributed to the formation of group norms around the way that the group played together. I used these results to formulate some design guidelines to facilitate the group communication and ran a smaller study testing these in an online game environment written as part of the project.

My results demonstrate that group communication in the early stages of group formation are affected by the design of the interface and can be positively affected by design decisions taken to increase the salience of the group identity. The design decisions I made were to include a clear group-name next to the individual's own in the screen, in a similar font size. I provided a separate group chat window, which could have the colour changed by any member of the group to indicate that they were now 'on-task'. Some groups used this feature frequently, whilst others only turned it on if a group member felt the level or tone of chat was inappropriate.

A successful group is one that was still active at the end of the study period. The successful groups formed hierarchies ranging from formal to informal, but in all cases the members had developed clear roles in the groups. They had a collection of group terms, and understood how the rewards of team activities (such as raids) were divided amongst the group members. Many of them had standardised the level of paralanguage used, although different groups used different amounts. Many groups also established clearly different patterns between social and on-task communication styles, often with a marked increase of abbreviated terms and the proportion of in-world comments.

My original contribution is to look at online group communication in a visual but not video (MMOG) environment. "


Today I had another meeting with my supervisor. Apparently that one was fine. So from there she wanted to work backwards to a research question. We batted it around for a while, asking what kind of question would actually be interesting. After a bit (not that long actually, it was an hour meeting in total), we came up with a question that actually, I rather like. The wording may change, but at a basic level I am asking: 

Given that there are benefits to working in groups in online games, what kind of interface elements and design help groups to form and work/play together?


It feels good for me, more informatics than sociology, and less nebulous and woolly than some of my other ideas. I can imagine a shape for the next two years coming from that, and while I don't expect my dream thesis abstract to come true it has definitely been a very useful exercise. 

Also: what a relief! Not having a specific question has actually been bothering me quite a bit. I know that's not etched in stone, but now I don't feel quite so much that I'm making up the question as I write the answer. Bring it on!


Poster presentation course

Yesterday I attended a poster presentation course. I kept hearing about posters and how PhD students normally start out by presenting them at conferences, but I wasn't entirely sure what it meant. Surely they don't just mean you make a poster that shows your work, do they?

Erm. Yes. Yes, they do. A poster is a graphical representation of your work to date. Just like in primary school. 

Unlike primary school, there should be a reasonably clear structure to them. In fact, the information presented should be rather like an exceptionally brief paper, with snappy title, background, methodology, results and conclusions. The order of the names on the poster (and apparently putting only one name on the poster is very bad form - your supervisor's name should go on there at the very least) is critical. The acknowledgements section is important. And don't forget the (brief) references section (ideally less than 10). 

The results section is the most important - that's the bit you can't put any spin on, what you will actually be judged on. Even if the results are preliminary, which many are. The methodology should give an idea of what you are doing, but not spell out every step clearly to prevent intellectual theft (as posters are unpublished and therefore not protected). The conclusions should remind of the problem and the results, and show why the results are interesting, along with indicating the relevance to other published work and what your future work may be. 

Quite like primary school, a bit of polish goes a long way. Of course, what that means has probably changed since I was 10. It's no longer a lovely double-border with a decorative pattern made from joined up letters. Now it ideally means you create a large pdf file and pay someone to colour print it for you. Printing multiple A4 pages and glueing them onto some backing paper is apparently not a good option, and looks like you haven't really prepared. 

There was a link given to studentposters.co.uk for poster templates which may come in handy - apparently starting from a template is an easy way to make sure you get the right size when it's printed out. Also, different conferences have different maximum sizes and may have different requirements for anonymity, so it is really, really important to read the guidelines for the conference. And finally, work out what you want the result of the poster to be before you write it: if you want a job, focus on how your work is relevant for industry. If you want collaborators, structure the information to show where people could help. 

So, all in all a potentially-useful course. A good overview on posters, and what to expect. The venue left a little to be desired, but that building (Fulton - it's brand new, but I don't like it) has some technical issues. I'll leave those for another rant!


Questioning processes through games

As a part of my DPhil, we are creating an educational game. We are aiming to build on a tradition of board games that have been made to provide policy makers some simplified insight into the daily lives of those that will be affected by their policies. In our case, we are looking at African farming communities. 

A while ago I came across the work of Ian Bogost on persuasive games. Bogost suggests that games are in a unique position to explore processes because by their very nature they are procedural. They can therefore be used to investigate the procedures that surround a particular situation, either by parodying this situation (for example, there are games written that try to expose the burger industry by forcing you to bribe officials into deforesting in order to produce more grazing for cattle etc.) or by making explicit some of the hidden items (e.g. a game written for Howard Dean's campaign for Iowa which exposed the mechanism of political campaigning - unfortunately while not really supporting Howard Dean's message!). 

My feeling was that there were some good examples in the book - the two I use above I think show where games can make things apparent that would be difficult to demonstrate in other ways. However, I felt that in places some of the examples that Bogost uses were very stretched. I had some doubts that the players would notice the things he was talking about in the game, let alone question the underlying social issues they were supposed to illustrate. 

Reading Lawrence Lessig's Code v2.0 started with a reminder that in the online world, code is law. Everything that you can and can't do online is defined by code, and all of that code was written by someone, somewhere. Someone made a conscious (or unconscious) decision to make it that way, and that code can be changed. However, Lessig points out numerous predictions and statements that suggest the online world is immutable and the way it is is a natural thing. People don't question the underlying structure, or wonder how it could be changed to do things that it currently isn't possible to do.

This lack of question of the structure for me exposes a problem with Bogost's argument. In order to use a game to make a system transparent and open for questioning, people have to be aware that the structure of the game is manufactured. They have to ask why the game was built in that way in order to then go on to think about the structure it represents. In fact, in some ways it could serve to reinforce the system represented. Players may just go 'oh, that's how it works', rather than really thinking about it.

Adam Greenfield in Everyware raises that point exactly in Thesis 23 (pg 84-87) when he discusses the JAPELAS system for teaching the Japanese language. He points out that the Japanese language has a system of hierarchy built into it, and worries about designing technology that reinforces this without question. Granted he is talking about a ubiquitous system, and suggests that PC-based learning systems may have less of that effect given how tied they are to the PC. But these systems do still have an authority, and we need to be wary of that. 

I still think that Bogost is right and that games do present us with a unique opportunity to explore processes. However, I think the other reading I'm doing suggests that a game on its own still requires reflection in order to really internalise the message (as with all learning). How to induce that reflection? The most obvious answer is to let a teacher point out the areas for discussion with the group. In fact, all of the board games we have looked at have a component of debrief after the event.

I guess I'm concluding that we have to be extremely careful with our game. We need to be very conscious of any figures we use about how different tactics (e.g. if we implement inter-cropping options or even certain types of fertiliser) are valid, so as not to suggest certain tactics 'always work'. And we also need to keep the teachers who will make the system most useful in mind when we are designing. 


Learning the ropes

We have a weekly lab meeting, during which people in the group take it in turns to practice talks, or get group feedback on projects, or share knowledge, along with keeping the group in touch with the latest university news. In one of these last term Ben du Boulay took us through a process of reviewing a couple of (carefully anonymised) papers aimed a the young researcher stream for an upcoming conference. The benefit in this is obviously two-fold - the new researchers in the group get to see how the people reviewing papers think, and the other researchers get a chance to compare the way they review papers.

Whilst the styles of writing in the two papers differed quite a bit and a couple of us found that a little distracting, most of the reviewing process did concentrate on the content. That is quite reassuring in some ways! 

I found it interesting that both papers were on experiments that were planned, rather than had already been done. They were therefore talking about proposed methodology, and there was a greater focus on situating the research. One thing that was strongly recommended by the researchers in the group but had been done by neither person was to spell out the potential difficulties they see with their methodologies. It does look as though the word limit was quite stringent, but with no discussion section I think that is a useful point. It could also be a brave move to point out the potential problems with your proposed project!

At the end of the meeting, in spite of a huge number of questions and holes picked in both papers, all the senior researchers present agreed that they would recommend the two papers for acceptance. The reason given was that the two students would really benefit from the feedback that they would undoubtably be given, whereas a 'perfect' paper would have been less likely to produce that sort of useful feedback. Apparently this was a reflection on the two papers being for the young researcher stream rather than the main conference, which is an important distinction to understand.

I've not yet tried writing a paper, but with this and the weekly reading group we are being exposed to things that our fellow academics see as being good or bad practice. I think in some ways this is a fascinating example of a community passing on the norms to the new members, whilst at the same time confirming those norms with each other. Attendance at these meetings therefore seems to be an important part of preparing for a career in research, by understanding the standards we will be expected to adhere to.

There's an awful lot more to this getting a PhD lark than first meets the eye!


A question of balance

As the weeks pass, it feels like I'm getting more and more different chunks of stuff to be working on. I've got the game project to write code for. Then there's the reading for the DPhil. And the writing of some sort of literature review. And the game project website, and (although guiltily this keeps getting shunted to last on the list) the research group website. On top of that there's the task of doing the whole 'raising my profile as a researcher' - which this blog is now forming a part of, and finding the right conferences to aim to attend, discovering the interesting people in my area, working out what 'my area' is, etc, etc!

I am finding that it is difficult to work out a priority. The project obviously has the big external pressures, and correspondingly it has clearly defined tasks and obvious milestones and progress indicators. This makes it temptingly easy to spend a large proportion of time on. There's a whole raft of tasks, each of which can be satisfyingly ticked off as complete when I've done them and shown off to other people. 

In contrast, the work on the DPhil is nebulous. It takes as long as I have to give it. The more I read, the more there is to read. There is a lack of feedback that I have completed tasks, and the tasks sometimes don't really feel like they are achieving much. Still, it's easy to spend all my time on the reading at the expense of the game project.

If I favour any one thing, I end up feeling rather guilty about the lack of progress on the others. For example, this week I've been working on the project website (although it still isn't up - I need to chase people again for webspace), so I haven't been working on the game code. Since I finished the website, I've been trying to get on with some reading and writing up what I've read. I've also thrown in a bit of blogging, thinking about the conference I'm going to in July (my first academic conference!), and today I've been to a lab meeting (with an interesting presentation by one of our research group) and a presentation by a visiting academic. Most of that has next to no visible outcome, so I feel like I haven't really achieved very much this week and I'm feeling a little guilty about that. 

One of the books I've been reading (and I think I have at least three work-related books on the go - another source of guilt!) is 'Reality is broken' by Jane McGonigal. She lists feedback as an important part of any satisfying process, and while I'm not sure I agree with gamification as a concept I think that is an important observation. I think I'm going to try to make myself some feedback systems - not points or prizes or anything, more a list of what I've done. I don't want it to take too long to do (and I don't think I want to go as far as a time-sheet), so I'm going to try breaking down the DPhil tasks and keeping a tally of them each week. I'm also going to ask my partner to kindly listen to me give a summary of what I've done each week, and maybe make appropriately encouraging noises. 

It may not make me a better time-keeper, or help me prioritise the work any more easily, but it might just take away that feeling of not really having done anything or getting anywhere. Worth a shot, I reckon!