Robots and humans

South Hinksey

Still catching up on work that emerged while in Scotland, I did take some time to follow a robot through Jericho today. It was collecting information about the world using GPS, a video camera, and a three dimensional laser rangefinder. While it wasn’t the cleverest of robots, it was a worthwhile experience. My fears about the coming robot uprising have been temporarily allayed; so long as throwing a blanket over one is a completely incapacitating act, we should be able to endure.

PS. Today involved an especially surprising unbloggable event. I need to think about it.

Hitchhike to Morocco

One very cool thing I learned while in Scotland is that there is an annual fundraising event for charity in which students hitchhike from England to Morocco. Mark did it last year, and I am seriously considering doing it during the Easter break. On average, it takes 4-5 days and it is done in groups of at least two. Each participant is meant to raise £500 in pledges for community development projects in Africa.

Seeing somewhere outside North America or Western Europe is a travel priority of mine, and I am especially keen on at least setting foot in Africa. This might be an especially interesting way to do so, and one guaranteed to generate some good stories.

Unlike Kilimanjaro, which will take a lot of planning and quite a bit of money, this could be done fairly easily and cheaply. All I need is a bit of registration money, a bit of food money, and an adventurous friend or two to come along.

On modes of reasoning

Electricity danger sign

One major tenet of liberalism is the idea that greater awareness of the world gained by people collectively through science and individually through education can improve overall human welfare in the long term. Firstly, the idea is that people will gain a more accurate understanding of the world and how it works. Much more controversially, they might improve the way they reason.

A game much loved by economists illustrates the controversy:

There are two players. The first player is asked to divide $10 into two parts and offer one to the second player. If the second player accepts the offer, each player gets to keep their share. If the second player refuses, nobody gets anything.

Standard economic logic would call upon player one to offer exactly $0.01, which player two should happily accept. Both players are made better off and should thus be willing to make the deal, and each player has maximized their earnings, given the rules of the game.

Of course, the game doesn’t work this way with real people. Hardly anyone will accept an offer of less than $3. This is entirely logical if you view the game not as an isolated occurrence, but one event in a life. Over the course of a life, it pays to develop strategies that keep you from having advantage taken of you. Likewise, over the course of repeated interaction, it pays to have strategies by which other people can be compelled to give you a better deal. This one choice may not offer the scope for such development, but the existence of such heuristic devices (like rules of thumb) can be extremely efficient where people have limited information and thinking power.

Economists, on the other hand, are about the only people who make offers of less than $1. More tellingly, they are also about the only people who accept such meagre offers. Through exposure to economic theory, their mechanisms of logical thinking have been altered. It is probably fruitless to speculate on whether they have been improved. Economists can understand the importance of factors like those listed above, so playing this way isn’t obviously a sign that their thinking has worsened. Of course, if economic trailing makes them less likely to anticipate that people might reject a $0.01 offer, perhaps they are worse off overall.

What is more interesting than the consideration of whether the economically optimal strategy is inferior or superior is the consideration of how frameworks of understanding affect decision-making and, furthermore, what effect that has on the liberal conception of welfare improvement through improved knowledge. The previous blog entry, for instance, portrayed the costs of global warming in terms of how much it would cost people to deal with (a very common economic representation). Drowned polar bears and damaged ecosystems only matter insofar as they affect people. Personally, I find such an approach reprehensible – for the same reason I think the wholesale denial of animal rights is morally unacceptable.

One can defend that position on pragmatic grounds: human beings with a reverence for nature have a better chance of living good lives and/or not wiping ourselves out. Saying we should cultivate the belief on those grounds is similar to Rorty’s conception of ironic liberalism. By contrast, the belief that the integrity of natural systems matters for its own sake has an intuitive appeal of a sort very un-chic and difficult to defend in a world full of poststructuralist rejections of firm ontological foundations to moral truth. Anyone who can devise an argument for the inherent value of nature not subject to such criticism will earn my appreciation.

PS. Inside nested padded envelopes, the dust-infested Canon A510 is en route to a registered service depot. If they decide to cover the problem under the warranty, I expect they will replace the camera outright, rather than trying to open and clean it. Doing so would take a fair amount of some technicians time and, if the camera isn’t properly sealed, it would only be a matter of time before parts of my sensor would start getting opaque again. Hopefully, it will come back in time for my trip to Ireland later this month.

PPS. I just upgraded to WordPress 2.0.4. Please report any bugs you come across on the bug reports page. Note also that, due to a barrage of spam comments, I have tightened the comment filtering settings. My apologies if any of your comments get zapped by the filters.

Global warming damage curves

Perhaps the biggest question about global warming is what environmental economists refer to as ‘the shape of the damage curve.’ I would say that the scientific evidence that global warming is taking place is essentially ironclad (though the relationships of some events, such as more severe hurricanes, to it are rather more tenuous). Equally true is the fact that human beings are contributing to the warming of the planet. What the damage curve represents is the amount of harm caused by each additional unit of global warming, expressed in terms of the cost that would be required to mitigate it. Mitigation costs include everything from relocating people to dealing with larger malarial areas, making agricultural changes, and increased building heating and cooling costs in different areas. The costs are net of any benefits that global warming provides: such as being able to grow certain crops farther north, or having a longer growing season overall.

The most benign possibility would look something like this:

The damage increases steadily with the amount of mean global temperature change. This is helpful because it allows us to predict the degree of future damage quite effectively and make reasonably good choices with regard to how much reduction in greenhouse gas (GHG) emissions we should undertake.

A worse option looks like this:

The damage increases at an increasing rate, as temperature does. This seems intuitively more likely than the first option, since bigger increases are likely to unbalance more and more complex biological and climatological systems.

An even worse option looks like this:

It is possible that climate change would involve a big jump that we wouldn’t see coming until it was too late. An example would be the much talked about possibility that the Gulf Stream, which warms Western Europe, could be disrupted. The biggest reason this is problematic is because we might believe we were in a scenario like the one in chart one, only to be proved spectacularly wrong.

The trillion dollar question, of course, is which of these approximations we should adopt as the basis for policymaking, until such a time as compelling evidence for one of the possibilities or another emerges (hopefully not by means of humanity actually following one of those curves too far). The most cautious option is to assume that the progression would be like chart 3, with the break at an unknown location. The prudent policy, then, would be to try and stabilize GHG levels at their present positions. Of course, that could involve massive reductions in possibilities for economic growth in the rich world and poverty reduction in the poor world. A tricky decision to make, in the face of such important considerations on both sides.

Personally, I don’t think any serious action will be taken until some very real evidence of the harm that can be caused by global warming has manifest itself. As for the question of what should be done, in the ideal circumstance, I am profoundly uncertain. What do other people think?

Geekiest joke ever

A group of functions are having a party: everyone dancing and having a good time, with the sole exception of a single f(x)=ex who is standing off in the corner by himself. Another function approaches him, hoping to help him enjoy himself more, and asks: “Why don’t you try to integrate?”

In response, the f(x)=ex says dejectedly: “I know it won’t make any difference.”

-100 points to Kathleen for providing the joke that effectively won the geeky joke contest that extended across the whole Scotland trip.

Newsfilter: Vanity Fair on 9/11

There is a Vanity Fair article that is getting a lot of attention right now, about the response of the North American Aerospace Defence Command (NORAD) to the September 11th hijackings. To me, it seems well worth reading, especially for the primary source material that is included. While it is obviously a particular selection of the available evidence, it does demonstrate convincingly that this was not a scenario for which a plan existed, or during which an accurate appraisal of the situation was forthcoming.

I still mean to go see United 93, if it is still playing in Oxford. Would anyone else be interested in going?

Scotland summary

Climbing towards a loch

The thirteen hour drives to and from Scotland were mostly evidence that I have developed the capacity to sleep my way through long journeys, with occasional periods for socializing and working out where we are. As such, it wasn’t really until we were camping outside Shiel Bridge and doing our first hike that I got to know the other members of our group. The most notable thing about the drive is perhaps the pub where we stopped for dinner. I had vegetarian haggis (quite good) as well as a pint of kelp ale. Most other people had the normal haggis, though some opted for the surprisingly generous portions of fish and chips that seem to be standard in Scotland.

As I said before, a majority of the trip members were physical scientists. One leader was a biochemist, one was studying quantum chromodynamics (a field that has to do with quarks and the strong nuclear force), and one is presently working in medical imaging, but is leaving in five weeks to go to opera school – quite a significant change of tack. Among the nine of us, there was only one undergraduate, which was probably on account of the trip happening during the summer. The two people on the trip with whom I spoke most were Kathleen – a South African student of political theory, doing work on transitional justice – and Mark – an engineer working with autonomous robots. I am also grateful to Bruno, one of the leaders, for his excellent advice on how to partially fix my extremely sore legs after our grueling first day.

The Five Sisters of Kintail

That first day began with the ascent of a ridge in the rain. Unfortunately, my Gore-Tex jacket is not performing as it did in the days of its youth (at least eight years ago) and it was pretty comprehensively soaked through by the time we reached the misty top. That said, with all the exertion, the water was not particularly unwelcome.

The five sisters consist of five peaks, three of which are Munros. Scotland includes 538 summits over 3000 feet, of which 283 were regarded as “separate mountains.” Those are the Munros, with certain requirements of disconnectedness from other nearby mountains. That ensures that climbing a Munro is always a fairly involved task – though Sir Hugh Munro climbed all the ones identified when he was alive, bar a single peak that isn’t climbable without fairly advanced rock climbing skills.

Every mountain we climbed (barring the small 350m hill on our day of recuperation from Friday) had a cairn of stones piled on top. The hillsides are all strewn with a kind of rock that is unfamiliar to me, and which tends to erode in geometrically distinctive ways, such as into blades and slabs. That, combined with the lack of any trees at high altitudes, contributes to the way in which the Highland mountains are both familiar and unfamiliar to someone used to the Rockies and the Coast Mountains.

Walking along the ridgeline – ascending and descending – we climbed all five peaks in sequence. Astonishingly, Roman (a mathematician on the trip) and Andrew (the chemist leader) ran the final portion so as to be able to recover the minibus from the starting point. The rest of us were happy to take a bit more time to rest and take photos.

After descending from the fifth peak, we found ourselves first in properly pathless terrain. Because of the aforementioned lack of trees, there really aren’t many paths in the Highlands. There are just passable areas and less passable areas. First a steep and vegetation covered hill, then a bog proved closer to the latter.

That evening, and at all times thereafter up to the present, my legs were definitely complaining about going from months of puttering about in the flatness of Oxford to suddenly climbing a series of thousand metre mountains. I think my combination of appreciation for having done such and impressive hike and the general sense that something a bit more pedestrian should follow was universal throughout the group.

Walking along the Loch

After a nearly sleepless night (on account of sore muscles and evil omnipresent midges), I was glad that our second day was more sedate. We walked for about fifteen kilometres along the coastline of the loch, before climbing a relatively small hill. The vegetation was quite interesting. Some was very familiar: foxglove, many of the ferns, and the general mossiness of the forest. The trees, however, were not kinds I had commonly seen.

After eating lunch on a pebble beach and climbing the hill, we found our way back to the minibus. Several group members convinced the ferry drivers operating between the mainland and the Isle of Skye to let all nine of us and our vehicle across for a discount. The ferry was a very odd contraption, with a deck that pivoted around a central point so that cars could be loaded from the side. It was only big enough to hold two vehicles the size of our minibus, and the driver seemed inclined to bring it into the landing at speed.

We did not spend a great deal of time on the Isle of Skye, mostly just having a look at the terrain before driving across to the bridge that connects it to the mainland. There, while following the loch back to camp, we stopped at an ancient looking castle connected to the mainland by a stone bridge. It was actually a twentieth century reconstruction and closed because a wedding was happening there. Four members of the group had varying levels of success as wedding crashers: three being turned back on the bridge and one making it as far as the island, to return in his own time.

Our meals were always quite similar: porridge in the morning; sandwiches, a packet of crips, two chocolate bars, and a piece of fruit for the hikes; and some kind of pasta and vegetable dish for dinners, with meat added for the 8/9 of the group who don’t share my dietary choice. Of all those things, I would hazard a guess that the chocolate bars were enjoyed most. They really are just the thing for when you are sitting, tired, on the shoulder of a mountain about 200m from the peak.

One last Munro

On Sunday, we had our last day of hiking. Firstly, we went up our fourth Munro under the best weather conditions so far. There was a blue sky, and you could see the shadows of the clouds moving along the mountainsides as they were carried along by the wind.

After that ascent, we had a choice between a hike that would involve two more Munros and rival Friday for length and difficulty, or the ascent of a single peak and the following of that ridgeline back to the minibus. Wisely, I think, we chose the latter and reached a pub for dinner just as it began to really rain.

That night, along with Mark, Kathleen, and Helen, I went for a walk in the drizzle along the loch. There, we found a pub and ensured that we would not leave Scotland without having had at least a dram of scotch. That may have partially explained why the last night was the only one where I slept quite soundly.

General impressions

Between the scenery, the exercise, and the company this was a really good trip. I really enjoy having diverse conversations with people in different fields, especially the sciences. Likewise, there is enormous satisfaction in a mountaintop view. It is both an affirmation of human agency – the ability to overcome so much gravity – and a humbling reminder of how briefly we live in the face of geology and millions of tons of rock.

Treason

Compass rose in Scotland

In the spirit of short entries, I have a confession to make: I am not a social scientist. Even worse, I don’t believe in social ‘science.’ Science is about things where you can access physical reality closely enough that you can be decisively proved wrong. Science is about improving our ability to act usefully in the world. Adding a bunch of regressions to your study on civil wars does not accomplish that.

PS. Political theory is about a million times more interesting than international relations theory.

[Update: 2 August] This entry doesn’t quite say what I mean, especially as regards the definition of science. It will need to be revisited when the ideas are clearer in my mind.

Apology for cursory treatment

Coming home to 900 blog posts, all laid out in BlogLines, is impossible. My apologies to you all, but your hard-earned thought committed to webservers have mostly been dismissed at a glance. It is a very concrete demonstration of the limitations of all human beings, and the hopelessness of capturing any significant share of human knowledge over the course of our lives.

Scotland 2006 photos: third batch

Taken over the course of all three days of hiking, these photos show a few more aspects of Scotland.

Scottish peak in cloud

Weather in Scotland changes rapidly and dramatically, especially at altitude. I was literally holding on to my hat on a lot of these peaks and ridgelines.

Castle by the Loch

There was a wedding happening in this castle which several members of the group were able to briefly crash before being escorted out by men in kilts.

Bruno and the slope

You get a better sense of the scales involve when there are people in the photos.

Ridgeline with blue sky

Brief periods of proper blue sky were much appreciated. Along this ridge, I shot a good portion of my roll of Velvia.

Group photo

Not a Monroe, this was the last peak we climbed before going home. By this point, my legs were so sore they became difficult to move if I stopped for more than five seconds.