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.

Entries in weekly (18)

Friday
Jul062012

That was week 90

Another week down, and I've made some good progress on the coding. Got to the stage where you can save a diet, but there are an awful lot more things to do before the nutrition is complete.

Got a poster done for the postgraduate poster show that I have to attend as part of my second year annual review. I'm struggling a little to come to grips with what needs to be done for an academic poster. I think (like everything) I need to work out the audience and then really think about what I want to present and how. I'm reasonably happy with what I've put together, but it will be interesting to see how it looks at A1 size. I also got feedback from my supervisor that I needed more words on it - apparently that's unusual advice but I have "a tendency for brevity"! I quite like that, but clearly need to strike a balance.

Been a quiet week with Jim on holiday. I've been a little tired too. Not sure what that's about! I think I've got all of the paperwork done now for my annual review (barring the stuff Judith needs to do) so that's good.

I really need to push the coding the next few weeks. Basically if I don't get almost all of it done before cricket week (mid-August) I've got a week holiday and 2 weeks (potentially) out at conferences. I need to get to the point of user-testing by the end of September, so that I can gather the right data to aim to be done by the end of my funding. I think this means the attempt at balance will have to go out of the window, and I may have to try for extra hours too. Try and keep the running ticking over, get enough sleep... Challenging, but it's a month to hopefully set myself up for the rest of my time. Worth it. Just got to stay on target.

Monday
Jul022012

That was week 89

I felt like I achieved a better balance last week between writing words and code, but we'll see.

I got a poster accepted to Fun and Games 2012, which should (finances permitting) see me in Toulouse for a few days at the start of September. However, chalk it up to experience too - I did use the wrong template (I followed the link on their page, rather than the wording in the text) and just because they accepted your poster doesn't mean you get free attendance. 300€ for the student price feels a little steep, but Pejman assures me it'll be worth it! I need to resubmit my abstract in the correct format.

I think I mostly got my student report written for this year, although the plan makes me feel a little like the Underpants Gnomes. (Step 1: submit thesis, Step 2: ?, Step 3: Get paid in October 2013.) I even managed to get all three academics to agree on a time and place for my review. So just the poster to go, and since I need to do one for Fun and Games I may as well kill two birds with one stone. The post needs to be done by Friday, for printing. Good pressure there then.

Now that the lab meetings are finished we've moved onto a series of 'people' posts, so those are hopefully writing themselves. Gareth was first to respond, so first to be published.

I saw Judith on Tuesday, which was a reassuringly brief meeting suggesting I'm moving in the right sort of direction.

The coding was slightly slowed down by a bit of a computer rejig. I've been having problems with my mac losing its network connection, which is a major problem when your home profile (including lots of application configurations etc) is stored on the network. It could take me 15 minutes to get the thing shut down, because of the force quitting that had to happen. So on Thursday Christian came and moved me to a local profile, with stern reminders to back everything up myself regularly. Moving stuff over seemed to take until halfway through Friday, and then everything needed setting up again. I'm not quite there yet actually. Still, made some progress. (Not enough! Don't panic yet!)

Jim is on holiday for the next two weeks, so I want to achieve stuff to show him when he gets back. We'll see...

Saturday
Jun232012

That was week 88

I am starting to notice with weeks that there is always an excuse as to why I didn't achieve what I'd hoped. This week we have plumbers in ripping out our bathroom, which has meant coming and going at different times to try and let them in or out, plus more cycling than I'm currently used to, and some really irritating network issues even when I'm in the office. So. Not as much done as I'd have liked. Maybe if I reflect on this enough I'll come to terms with it and start a) using the time I do get more profitably and b) getting more realistic with my expectations.

Coding-wise I tried a new tack this week. With Jim out on Wednesday and Friday and Katy off on a well-deserved post-thesis-submission break, I went out for coffee twice this week. I took my iPad, and spent the time offline (deliberately picking spots that I don't think have wifi!) thinking about my code and writing quite specific to do lists. "I need this class creating, it needs this function, this class must call it in this function with this information" and so on. What this meant was I then went back to my desk knowing what I needed to do, and just got on and did it. So I got more done than I have for a while. I do think it only worked quite as well as it did because of the hours I've spent staring at the code so I know the problem inside out, but at least it felt like (pleasant) progress. More next week hopefully.

Writing was still a little sporadic. I started my student report for my annual review, but then hit problems trying to organise a date for the damn meeting. Trying to get three specific academics in the same room at the same time is really hard. I wrote up my thoughts on the Athene SWAN meeting I had on Wednesday, and I followed up on our lab meeting last week a little. So that is writing, just not for a paper.

I didn't make it to Tea at 3 because of the plumbers. The lab meeting didn't happen because apparently I really do need to send out a reminder the day before or no one shows up. (I was irate/disappointed, but I'm over it. I'll eat the cake on Monday.)

No meeting with Judith this week, so I'd better book one for Tuesday.

And that was week 88.

Saturday
Jun162012

That was week 87

Another unbalanced week! This time it was all about the writing. I submitted a 4-page paper to the HCI 2012 doctoral consortium, so this week has been all about writing that paper. Unlike the last time I submitted to a doctoral consortium, this time I've actually done a study. So the vast majority of this paper was about my playing of the Green Revolution Game and processing a bunch of stats from that.

Stats are proving to be a little thorny. I kind of stopped studying stats after GCSE. All the mechanics options seemed much more useful when I was doing engineering, but my lack of stats is slightly coming back to bite me now! Means and so on were pretty easy to generate from my Likert scale data, but then the inferential stats were more complicated. I decided my data was non-parametric (I know. Plenty of psychology papers seem to disagree, but I just can't believe that Likert scale data is parametric), so ended up doing some nice Kruskal-Wallace tests. In Excel. Because I didn't have time to learn how to do them in SPSS. Fortunately when I was told that was the wrong test for at least one thingummy Katy stepped in and helped me process them with SPSS. The paper got submitted around 5pm on Friday, so I had 6 hours 59 minutes to spare.

Met Judith on Tuesday, and mostly talked about stats!

Got asked if I'd help out with Sussex's Athene Swan application too. I'll need to find out what that's all about.

Had a bit of a problem with Dropbox on Thursday night. The file that I was working on all day showed up as unchanged from the day before when I got home. I'd been intending to do a bit more work, but that kind of scuppered me. With hindsight, the network at Uni had been flakey all day and I should have taken a copy, but honestly I was knackered so an evening off was welcome. Does mean I'll be taking more regular backups though, as it looks like I can't entirely trust Dropbox.

No code even glanced at this week. Still searching for that balance. Maybe next week?

Saturday
Jun092012

That was week 86

That was a very short week, and I mostly felt a bit rubbish. I spent a large chunk of the long weekend asleep, and Wednesday mostly I wanted to continue with that. Fortunately I perked up a bit on Thursday!

I've been struggling with a chunk of code around diets, food and nutrition for what feels like forever, and this week didn't bring any kind of resolution unfortunately. I'm coming to the conclusion that I need to really rethink my model for assets. I need to subclass an asset class for the various types of assets, like crops, food, crop enhancements etc. They have very different properties and functions, but enough overlap (e.g. People and households own a number of them, the market needs a buy and sell price for them...) that subclassing looks just right. The only problem was the database side. I think I've worked that out with Hibernate though, so it's ok. I'm a little concerned about the overlap between some crops and food, but not all crops and food (horticulture is food, but it's not really a plantable crop!). I'm starting to think I might have to let that slide for now. Might need to check with the sponsor whether people save seed from stuff like tomatoes or just re-buy it every year.

No writing again. Really need to do something about that.

Lab meeting went well, but I need to come up with something for next week. No ideas at the moment. Two more to go!

So yeah, not my most successful week. Onwards and upwards I guess.