Martin
TFF's Token Imp

Yarp. I still work there, but I've moved to Middleton now. I'd say that was a good memory remembering that, not a bad one haha!

Thankyou its ok!!!

Define horrible? We had our first actual frost yesterday, although I think with the winds we've been having it's probably like the 15th sub-zero morning of the year.

Hmm. I hate this question. I'd refer you to Facebook, but I'll try explain it. I work part-time (around study) for The New Zealand Institute for the Study of Competition and Regulation (ISCR). I work as a research assistant, utilising GIS to support a variety of econometric studies. Essentially, the organisation tries to convince public and private sector entities to become (paid) members, and we do research for them. Often, we do the research first, to convince them of how badass we are, and get them to be members. We supplement this by publishing parallel bits of research in appropriate academic journals. At the moment, I am working on a project using data from the New Zealand Transport Agency (NZTA). It's an econometric model of residential choice and car ownership. The best way to think about it is this: cities across the world are trying to grapple with the external harms caused by road traffic. They try things to reduce them -- like changing relative prices of different modes of travel (bus and train tickets, petrol...) or adding new services or what have you. But the problem is, we often don't know what the effect of these are. Do some interventions do nothing but cost a lot? Do some interventions make things worse? Which are 'good', and for what purposes? the data I have from the NZTA is an on-going travel survey (this). I extracted commuter information from it (because it covers all kinds of trips, but the ones we consider relevant for policy are regular trips to work that are amenable to change, unlike your trip to your aunty's two towns over), and now have a commuter database. Therefore I know all about everyone captured in the survey who travelled to work: the location of their house (secured information, I had to get ethical approval, and it was the first time it was released ), their age, occupation, family structure, the location of their work (and all intermediate destinations travelled to on the way), etc.

The econometric model (and I'm not an expert on it... I'm the (geographic and other) data guy, not the econometric guy (that guy is my boss, who has a PhD in Finance)) requires that we analyse the 'observed decision': you live where you live, you own two cars, and you commute by bus. You live where you live for a number of reasons--the price, it's close to the bus route which takes you close to work, it's sunny, it's zoned for a really good school which happens to be (let's say) Catholic, like your children are. You own two cars because you're wealthy enough, and have children who need driving to sport on a Saturday.

Then we need to evaluate other alternatives that you could have chosen. Why didn't you live in the next suburb over? Why don't you only own one car? If you keep asking these questions, it becomes: why didn't you live in any one other of the 150,000 residential addresses across Wellington city? Why don't you own 0,1,2...6+ cars? This is called a conditional logit model. It's my job to get the GIS to be able to evaluate the potential travel decisions of each (real) person in my database from their actual home to their actual work, and then again from each of the (unchosen) possible residential alternatives across Wellington to their actual work. We need to know how long it would take them (if it is even possible), and how much it would cost on foot, by bike, by car, and by public transport (train, bus, ferry and cable car). Then, we can use this ****ing huge matrix of data to compute utility functions and work out which aspects of the built environment are the most important determinants of decisions relevant to policy.

Maybe those who live only a short distance from relevant public transport own very few cars? This is important to know; and not only will we know it, but we'd have quantified it quite precisely.

The existing literature (and this is probably the best in it) has a limited array of alternatives and a piss-poor attempt at GIS. That paper only considers 10 other residential alternatives across the entire NYC... I am considering 4504 alternatives across Wellington. Their GIS used estimates of travel time. I am constructing a comprehensive multimodal network dataset (think Google Maps, except instead of a little web interface, I can change absolutely any parameter I want, and get it to automate route solves for (literally) the 800,000 iterations I need to do per travel mode, and write the output to a database.

...OK that was probably a lot more than you were expecting; I'm just really passionate about it. I really like going to work haha. I've been meaning to write that into my journal too, so thanks for le motivation.

That because he brought it upon himself. You weren't here for all that drama he brought out. It was pathetic.

Thank you, Mr. Sir. I like you too. If you aren't a member of the Goddamn Eastwood Academy of Awesomeness yet, feel free to join. I mistakenly overlooked you during my initial recruiting.

We do! I really need a day of nothing where I can just catch up on the Internet in general. I've finished classes for three weeks, only to be pulling 10 hour shifts. And the last thing I want to do after spending a day at a computer is spend more time at a computer.

How's winter treating you?

I know, I just had to tell him some way.

No, I do not have a Twitter.

I most certainely do not use @Aerif, and randomly tweet people concerning the use of RuneScape. Nope. Never.

Aye, Ron.

just watched it there, haha yea looks like they still have it

You love an Aussie, that makes you an Aussie! Yay!

kinda scared in case it disappoints... like how could they top the first one

must check out the trailer, havent seen it yet.

Hahaha you are a member! Welcome! It's a pretty crappy inactive group btw. But oh well! Glad to have you on board you bloody Aussie!

This ... is mostly why Prestige+ exists haha, just so you get the idea

OMG HI MARTIN! <3 How are you!?

Haha yeah I still listen to them some times!