Just a quick mention of a few things before I go home, where hopefully I will finally have a phone line and therefore be able to blog at leisure when I'm not at work:

Rod Brooks (in some ways the father of the sort of AI that I work with) made HPLB a stop on his whirlwind UK book-promoting tour, and gave us 2 talks. A big open lecture (with a surprising number of non-HP people present) on things that I mostly knew about already because we study them, and a smaller seminar the next day about Project Oxygen which was HP-only because HP are a partner in the project. This second thing was very cool - it's a combination of lots of interesting human-computer interface ideas that promise to make using computer systems far more natural and intuitive, and are much better developed than I would have expected. The state of the art is several years ahead of where I thought it lay in that field, which means it won't be long before we have networks that recognise the user by face, know where all the users in the building are, and generally act like sort of omnipotent secretaries rather than cantankerous machines that we have to adapt ourselves to. They're going to be monstrously expensive for some time, because they need so many cameras & microphones and so much computing power, but it's still pretty exciting (for a geek like me in any case).

If you use Hotmail I strongly recommend checking your privacy settings, because Microsoft have gone and slyly introduced some new options, which by default are checked, allowing them to share your email address with partner companies even if when you signed up you chose for this not to happen.

The Economist, which much as I love it does tend to be an extremely pessimistic magazine, has published two fairly optimistic articles this week: one about Carter's visit and a gradual move towards a more open society in Cuba, and one about East Timor's impending independence.

Geoff Sampson issued a statement today (within the department - I don't know if it's meant to be circulated or not, so I'll err on the side of caution and just paraphrase it broadly) admitting that the publication online of his controversial article was a big pragmatic mistake, strongly watering down the offensive part of the article, denying being a racist himself, but (and I think with some justification) defending his right to hold views that others disagree with violently. His actual denial of racism was not entirely convincing (he argues that he's spent years studying Chinese language & culture, but that proves nothing - one can be a racist and favour some groups other than one's own), but most of his statement makes sense as well as saying things he needed to say, so I think the fuss will calm down somewhat, and I get the impression his job is safe. The whole debate has at times been rather silly, with words and attitudes being put into his mouth without justification, and a general assumption that he is guilty of various bad things by association, just because he expressed one opinion in one article. Once I have email at home I'll read the whole debate through at leisure and make up my own mind about this properly, but I think I'm finding myself siding with the people who argue that he shouldn't lose his job over holding and wanting to discuss an unpopular opinion. I definitely think it's right that he's been made to dissociate these views from the University.

posted @ 1:15 PM -

There's an interesting discussion about Geoff Sampson on MetaFilter. I haven't had the time to read it all, but I'm printing it off to take home with me (still no home phone. boo!), and what I have read seems well enough reasoned to be worth drawing peoples' attention to. The current state of affairs on campus is that the Students' Union have issued a statement condemning it, but also making lots of links to the wider political picture of how racism in politics does seem to be on the rise (these links to the bigger picture tend to turn people off, but they are at least semi-valid), and they are trying to organise a picket on Monday to draw attention to the issue. There's been no more official action and some strands of the email debate have been quite interesting, but I'm still waiting to see how this ends.

posted @ 2:57 PM -

jdbgmgr.exe

I've noticed that a large number of people have found this page over the last few days from searches relating to this hoax virus alert, so the hoax must be propagating well, and it seems worth clarifying a bit of information about it:

1. It is a hoax - there's no need to delete that file


2. The file is (as far as I can tell) used by the Windows Java Debugger, which means that most people won't really miss it, but it's still better not to delete system files unless you are absolutely certain they are never used


3. In general an email virus alert is almost certainly a hoax, especially if:
   • it claims to have come from Microsoft and/or IBM (they are not anti-virus companies, so they don't tend to issue such warnings)
   • it makes terrifying claims for how destructive the virus is (think back to middle school biology - a virus that is too destructive doesn't spread because it needs live hosts - likewise most computer virii damage but don't destroy)
   • it claims that no anti-virus software can detect and clean it, but you can with a few simple actions (duh! the collective effort of lots of top computer scientists can't do what a naïve user can...)
   • it says that a file merely being present is a sign that you have the virus - virus writers normally make more effort to hide their masterpieces until they activate


4. None of these things guarantee that the warning is a hoax, so if you are suspicious you should always check with a reputable source such as Symantec. If you don't see anything about the alert you have just received you can also send it to Symantec for an expert opinion - if you are the first person to report either a hoax or a real virus you will be doing a favour to millions of others


5. Hoaxes can come from eminently reputably sources, because some IT managers and tech support personnel are severely underqualified, and others are just mischievous, so even if you think there's no reason to be suspicious it's still worth checking with someone reputable in the anti-virus industry before following the instructions or passing the message on

Oh and one final thing, remember that the chances are the person who sent you the hoax had the best of intentions, so even if they really ought to know better it's not fair to have a go at them because of an innocent mistake.

posted @ 6:11 AM -

Update on Geoff Sampson (the Sussex professor who wrote an article supporting racialism): the article has been removed (possibly temporarily while the storm blows over) from his website, and his University homepage no longer links to his writing about current affairs. So the University have done at least part of what I had hoped they would do.

Meanwhile the furore over this (or at least what I can follow of it while being 200 miles away) is getting a bit silly, with the same few people emailing repeatedly, and basically repeating the message that they don't like what he said in different ways. One person even decided (I paraphrase a little, but I'm not being too unkind to the original writer) that he must be a homophobe because racists generally are. Let's keep this rational and only punish him for things he has done, not things he might be expected to do.

One very sensible point that came out of the email discussion is that his views on racialism are inconsistent with his views on language; a field in which he argues against biological specification of abilities, and for culture as a means of learning.

posted @ 5:42 AM -

Sussex professor expresses support for racism

That's not a typo. There's a small storm brewing (probably about to break out into a big one) back at Sussex Uni about a professor who has written an article called There's Nothing Wrong With Racism (Except the Name). So far the University have defended him on grounds of academic freedom, but there are some fairly serious issues here. First of all it will make it hard for ethnic minority students to work with him, and if I got a bad mark from him now I would be tempted to assume (something I normally refuse resolutely to do) that it related to my being a Jew and/or an immigrant. Rather more worryingly to the University is the fact that this is getting some media attention now, and he keeps using his status as an academic to reinforce his point. Except that he's not expert in anything to do with the subject matter of the article: he's a computational linguist.

I will be very disappointed if by the weekend he hasn't been at least made to dissociate his professional work from this article. I'm not actually sure whether he should be censured further than that, but I won't be surprised if he has been presented with a choice between retraction or dismissal, because this is about to become very embarrassing for the University....

posted @ 1:34 PM -

another small milestone

Today each of the Sussex students on internships at HP had to give a little talk to the department explaining what we were doing, what relevant things we had found in the literature, and how we planned on approaching the respective problems we had to solve. I think it went fairly well, based both on my feeling in there and feedback from my supervisor. It's also a significant milestone, because one of the things I was a bit nervous about with the culture here is their habit of making quite frequent presentations to each other about what they are reading about / working on. It's a very good habit, because academics are often quite isolated from each other, and this breaks that down somewhat, but I needed to give a first talk and see that they don't bite in order to stop worrying about it.

posted @ 1:17 PM -

We were supposed to have a phone line installed in our house last week but it still isn't working, hence the sporadic updating here. I am, once again, not impressed with BT.

posted @ 1:10 PM -