Sunday, April 27, 2008

Sacrilege of Modern architecture's high priest

I saw in today's Sunday Times that Norman Foster is going off to live in an 18th century chateau in Switzerland. Hey, I admire his taste. But, why do the rest of us have to carry on trying to like (or pretend to like) all the ugly glass-and-steel modern stuff that he and others designed, when he turns his own back on it? (Aside: when I looked at Foster's Wikipedia entry, linked-to above, the news article that I also link to, was already linked-to from the Wikipedia entry! I admire Wikipedia's efficiency.)

Friday, April 25, 2008

Doctoral Training Centre proposal

I am helping to prepare a bid for a "Doctoral Training Centre" (DTC) in response to this call for proposals. We need to make a 3-page outline to submit by the 6th of May, then some of these outlines lead to invitations to submit more detailed proposals.

It seems that Liverpool is preparing about 5 of these bids. According to one of the people who is helping to coordinate the bids, it is likely/plausible that this is going to be the new model for how PhD studentships are allocated by the research councils, as opposed to the "doctoral training account" ones I mentioned in my previous post here. While my participation in this proposal suggests a certain level of endorsement of the general concept, I can still consider the advantages and disadvantages in a fairly impartial-looking way...

DTCs can be advertised effectively; a DTC may typically take in about 50 PhD students over 5 years, so it's a big centre. You can spend money on proper advertising, and better yet, you know in advance that you will have the funds for the students, so you can make a proper advertising campaign. (Note the complaint I made in my previous post.)

But, because they're big centres, taking in a large part of limited funds, DTCs lead to concentration of activity in a small number of institutions. Good news if you're one of the "winners", I guess, but the losers face the prospect of having to get out of the business of PhD supervision altogether, at least if the prognosis mentioned above is accurate.

Can DTCs as envisaged, be faulted for reducing the choice available for prospective research students? (Thus, currently, prospective students can submit their own research proposal, hopefully after working on it with an academic who could supervise.) I don't think that's a problem --- I reckon that prospective research students cannot identify a research project in much detail; it only makes sense to expect them to know their general research area of interest, e.g. "algorithms and computational complexity". (Experto crede — I am PhD admissions tutor!) On the other hand, I think they can be criticised for reducing academic freedom, or at any rate, flexibility in choosing a research topic for one's PhD student.

Finally, whatever our success at this effort, it is somewhat interesting to get to meet up with people at the University from outside my own department, and get a view of the bigger picture.

Thursday, April 24, 2008

PhD studentships

Finally, we get our allocation of Doctoral Training Account studentships from EPSRC. I have posted the following advert to various lists and encouraged colleagues to do similarly. I am unhappy about the delays to the allocation of these studentships; I have no idea what caused it, e.g. maybe some university entered into a dispute with EPSRC about their total allocation? That is sheer speculation, but it's the sort of thing that may have happened. Basically, the studentships get allocated by EPSRC to universities, based on research grant income, and then universities allocate them to departments.


Applications are invited for PhD positions at the Department of Computer Science of the University of Liverpool, UK. The Department has funds to support three PhD studentships to begin on or around September 2008, and to run for 3 to 4 years. Further details are at:



http://www.csc.liv.ac.uk/~pwg/2008-studentships.htm



The Department has particular strengths in Algorithms and Complexity, Logic in computation, and multi-agent systems, so we particularly encourage applicants with an interest in any of these areas.

Note that two of these posts are EPSRC Doctoral Training Account studentships, and are only available to students who are resident in the UK; the above web page has details on eligibility. The other studentship (funded by the Department) is available to any candidate who is a citizen of an EU country.

Tuesday, April 22, 2008

COMSOC-2008

I am helping to organise COMSOC-2008, the second International Workshop on Computational Social Choice. Specifically, I am co-chair of the program committee. Fundamentally, it's an informal meeting; there is a proceedings but it does not purport to publish papers; it just provides a forum for presenting work in the area. One of my better ideas regarding this meeting was to solicit ideas from the invited speakers about additional events, such as an open problem session, which should, at this point, go ahead, provided there is sufficient interests from participants. Next thing on my list is to send out an updated Call For Papers to various mailing lists and other places... Paper submission deadline is June 3rd.

Tuesday, April 15, 2008

EC'08 accepted papers

In the time-honoured manner, I hereby commemorate my work on the program committee of ACM-EC 2008 by linking to the list of accepted papers. (Although, since it was a big program committee, my contribution was correspondingly small.)

Wednesday, April 09, 2008

Who rates the league tables themselves?

This article at the BBC News website (Tables "affect university policy") reports that universities' policies are strongly influenced by their effect on the various league tables that are used to compare universities. That fact will not be surprising to most academics. The article is about a New report compiled by HEFCE, Counting what is measured or measuring what counts? that raises the concern that league tables may have a bad effect on universities themselves, and may fail to measure the quaities that they purport to measure.

Since they compiled by the popular press, often in haste, and without the attention to detail and rigour that we take for granted in the acedemic world, we should not be surprised by this finding. The one thing we can sure of is, however, that they will undoubtedly continue to exist in their present form, and will continue to be used extensively by, for example, prospecitive students, simply because that's all the comparative analysis they've got.