Fuddland

Entries concerning the work I do towards my PhD.

My debit card expired quite a while back but I didn’t fancy the new one being mailed all the way to China, and since I don’t really need to withdraw money from the associated account [the primary reason being there’s bugger-all in it], I asked my bank to put a hold on sending me a replacement until I’m back in the country.

As this terribly exciting event is fast approaching, I contacted them and asked for the new card to be issued. While I was at it, I enquired about the possibility of changing the title on the card from Mr to Dr, being prepared to send them a copy of my PhD certificate as proof.

But apparently these things are just taken on faith! Their reply simply said they’ve changed the title and issued the card as requested. Why doesn’t everyone do it? And why stop at being a lowly doctor? Next time a new card is due, I might mention that I’ve since been elevated to a professorship, received a knighthood and married into a royal family in a little-known Eastern European nation—His Royal Highness Professor Sir David has a nice ring to it, don’t you think?

As I mentioned, on Friday I successfully defended my PhD thesis during a two-and-a-half-hour-long viva. During the lead-up to this occasion, besides being asked if I was feeling ready [no] or if I was nervous [yes], the most common query from friends and family was: what the hell is a viva?

Before my own, I was only able to give the dictionary-definition answer: a viva—short for viva voce, a Latin phrase literally translating as “by living voice” which has come to mean “by word of mouth”—is an oral examination, conducted by two examiners [one from the student’s university, one from another institute, both somewhat expert in the relevant field], in order to ascertain just how well the student understands the body of work that they have researched and submitted in their thesis. It lasts for as long as the examiners decree and focusses on whichever aspects of the student’s work that they decide are important. Those last two points are why it was hard to say precisely what occurs during the viva, because I really didn’t know exactly what I was going to be asked, and every one must by its very nature be different, but now I’ve been through it myself, I can provide a little more detail on my own experience.

Read the rest of “Doctoring”…

Ooh look, I’ve passed my PhD.

I could try and use this weekend’s unexpected loss [and subsequent full retrieval from backup] of my weblog database as some sort of backwards justification for not writing anything for nearly three weeks—if I’d written anything recently, the backup might not have been current enough to restore everything—but I fear that wouldn’t wash with anyone [least of all me], so I suppose it’ll have to be the boring old: I just haven’t felt like it. No real reason beyond that.

I started to write about the things I’d been doing over the last few weeks, but it all seemed rather pointless: I already know, and why should anyone else care? I find myself being less and less interested in writing weblog entries which are little more than reportage [such as, erm, this very entry]. But then commenting on, or having an opinion about, current affairs requires actually reading the news from time to time, something which I’ve also become rather lax about of late; I still skim the headlines of the BBC’s various RSS feeds, but rarely click through to the main article.

Instead, I’ve been lending a helping-hand to friends in a spot of bother [involving, amongst other things, a weekend of painting and decorating which was the most fun I’d had in ages]; forging new friendships; rekindling old ones; and of course, doing a bit here and there towards completing my PhD. The pressure’s really on but the end remains frustrating, tantilisingly just out of reach.

I decided recently that I needed to listen to some really authentic blues, and after a tip-off from a scene in E.R. of all things, I went right back to near the beginning of recorded music and got hold of Robert Johnson’s Complete Recordings, the fruits of just two recording sessions in which he laid down his repertoire of songs, classic blues that are still being covered by today’s artists.

The three-body problem is a classic question in mathematics, usually posed in an astronomical setting, with its solution describing the motion of three masses of various size [a star, a planet and a satellite of the planet, for example]. Unfortunately it’s generally not possible to solve the problem precisely, so one needs to use numerical methods [that is, use a big computer] to work out approximate solutions. The fact that there is not a quick and easy solution has profound repercussions on everyday life, as I recently found out.

Read the rest of “The three bodies problem”…

We’re having the server room next door to our office kitted out, after it’s been sitting empty for over a year. This task apparently involves lots of banging, crashing, shouting, singing tunelessly along to the radio, and more banging. Obviously this in no way affects me because it’s not like I need a nice quiet environment in which to work and think. I just like to spend my day composing sarcastic emails, like the one I just sent to all the lecurers who sit, insulated from the outside world, in their cosy offices on the floors above the postgraduate room.

Idea for a new “reality TV” show

Office Swap: the Applied Maths professors and the postgraduates from a Midlands University Mathematics Department swap offices for a week, during which time extensive work is being carried out in the room next to the postgraduate office.

The professors will be tested to their very limits as they attempt to undertake their work with the office being used as a public thoroughfare. The first task will be to come up with a reason why the corridor that runs just alongside the office cannot be used as, you know, a corridor.

The next task must be deciding which is more annoying: having to continually get up to open the door for the people who have forgotten the code, or leaving the door propped open and enduring the continual yet inexplicable beeping from the so-high-tech-it’s-never-bloody-worked alarm system just outside. The only catch is, if you do prop the door open, you have to leave a few laptops lying around unsupervised.

The afternoon brings a disco theme as the light-switches are turned on and off randomly whilst the new switches are tested, to the tune of whatever pap chart music is pumping from the radio in the next room. You may dislike it at first, but the fumes from the glue-gun or whatever it is will soon send you into a dizzying high and you just won’t care anymore.

Coming soon to a cable channel near you.

By the end of today, I shall be an expert C++ programmer.

Current level of C++ expertise: zero. That’s a fair old climb.

Just what the hell is the Laplacian of a monomial? The expression I’ve come up with doesn’t equal the expression I have in a book, and now my expression is somewhere between bewildered and bemused dismay.

If only there was a maths equivalent of RentACoder, from which—believe it or not—some of our Computer Science students have been caught requesting solutions to their coursework. Thankfully one of the coders had some morals and alerted the Department that a few of our students were dirty rotten cheats. I’m not sure what fate awaits them, but I say, if only for the sake of the future employers who might think they’re hiring someone who can actually code: kick ‘em out.

Post temporarily removed.

Well that went pretty well I thought [and, more importantly, so did a few other people, including my supervisors]. There were a couple of the inevitable sticky questions at the end, but also some suggestions as to interesting directions to go with my work.

But obviously the first direction I’m going is pubwards. I leave you with the highlights of my talk.

It’s the day of my talk. Eek. I hate doing this sort of thing. Still, it’ll all be over in about five hours, and at least I get a nice free lunch out of it. There’s actually two of us speaking today—I think I’m on second, which I prefer. I’m just slightly nervous because I know certain members of the audience are working on rival methods to solving the same sort of problems that my work concerns, and I’m not entirely convinced that theirs aren’t better. Not that I’d ever admit that publically. Except in a public weblog. Err.

I’m going to find a dark room to hide in.

I met some members of the University’s Court today. I didn’t even know the University had a Court, or even what one was, so I did a little digging.

Court is the highest body in the university system, and [in our case at least] consists of:

1. The Members of the Council;

2. The Members of the Senate;

3. The Emeritus Professors;

4. The Registrar;

5. The members of the Standing Committee of Convocation;

6. Five representatives of the Students’ Union elected by the Union;

7. One representative of each of the recognised trades unions of the University;

8. Such other representatives of the University as the Court may determine.

Every year they tour a few departments, to get a flavour of what goes on research-wise. This year was our turn, so yesterday afternoon [yesterday afternoon! As if we’re not busy enough] a couple of colleagues and I were asked to prepare some demos of our research.

I made a few movies [in lowly 2D due to the time constraints] which purported to demonstrate some of the ideas and applications of interpolation. These were promptly blown out of the water by someone else’s 3D simulation of the solar system aging a couple of billion years, but there’s obviously no jealousy in the world of mathematics. [Not since Galois anyway.]

I chatted to a very nice chap at the beginning; I cunningly kept him talking about himself for long enough that the rest of the group passed me by without a word, and he was a non-scientist so was obviously very impressed by my 2D movies [I made sure he had his back to the solar system simulation]. Once he’d left I did had to explain what I did to a couple of other people, but they seemed to get the gist, or at least [and more likely] were well-practised at nodding in the right places.

I do enjoy important people trying to seem interested in things they really don’t know anything about—it often results in things like this [paraphrased] exchange, which occurred towards the end of the afternoon:

Court member

How much more accurate is your model of the solar system than the ones the Babylonians and Egyptians were using?

Colleague

Well I don’t know—how accurate were their models?

Court member

Er. I don’t know.

Colleague

I think they were only interested in a couple of thousands years at most, whereas my model does things in millions and billions years. Also, I think they were more interested in alignments and significant astronomical events such as eclipses. My model is really trying to predict whether the solar system is stable or if the small chaotic traits we are aware of will result in total entropy, the dissolution of the system and the end of any life that might happen to still exist on Earth.

Court member

Well moving on…

i think i’ve got that code ticking along nicely now, with forty minutes to spare. initial tests didn’t look good, until i realised i was plotting the wrong set of values — no wonder the outcome was completely wonky. as the saying goes: more haste, less curry.

[fingers crossed i don’t find an ‘out of memory\disc space’ error when i return to my desk later today tomorrow.]

good job too — horribly cold this morning, and very foggy too. i couldn’t see my building as i walked across the sports field, but sadly it loomed into view once i’d got a bit closer.

time is of the essence this morning: i need to get a little routine coded and running perfectly so i can leave it churning away whilst i join the other postgrads for our end of semester curry & beers at half past twelve. the prospect of having to leave the pub to come back and carry on working does not appeal [and probably wouldn’t result in a productive afternoon anyway].

unless you’re double-jointed and aren’t planning on letting other people read your document, here’s the recommended setting for the laser printer in my building.

all i want to do is photocopy a paper from a mathematics journal.

Read the rest of “see the hoop? jump, dvd, jump”…

i’ve broken my matlab code.

it used to work. but then i improved it. and now it doesn’t work.

i’ve decided the easiest thing to do is redefine the word ‘improved’ to mean ‘broken’. please inform everybody you know of this new definition.

i’m up to my bloodshot eyeballs in code today, immersed in a sea of matlab .m files, and it’s no fun. actually, that’s not quite true — i do quite enjoy writing the initial code, it’s the debugging i loathe. this is mostly due to the fact i am always sure the routines i write will work straight away; it’s disappointing when they don’t, and having to trawl through until i find and correct all the mistakes is just tedious.

thankfully a lot of the errors are usually down to me storing the variables as a column instead of a row vector [or vice versa], or something stupid like that, and the main problem now is not about the algorithm i’m working on, but how to convert the data generated into flash movies so that i can actually see what’s going on. [note that’s not flash movies, just movies that are quite flash, in a ‘watching primary-coloured objects rotate’, kind of crap, kind of way.]

sadly windows and matlab do not get along very well, and windows suddenly gets all selfish and refuses to relinquish control of the movie files i’m generating, so when matlab tries to overwrite them, it spouts out a ‘can’t delete file’ error. i’ve no idea how to stop windows from doing this, and besides it’s too unstable to confidentally run my code for any period of time.

however, this forces me to instead use the version of matlab on the department’s solaris based machines via remote login. now i’m sure solaris is a great operating system, and has many advantages over windows [its stability being a major one], but its graphical interface is just awful — i hate it hate it hate it — and i’m forever losing one window behind another or running out of room for all of them.

all i need is a nice, stable, usable environment to get the code to the stage where i can leave it running for hours at a time, so i can sit here watching it run and pretend that counts as real work.

i had my end-of-year progress review this morning — a good month or so after submitting my report due to the interviewers’ holidays amongst other things.

most of the emphasis was on how i’ve found shifting supervisors, and areas and styles of research. when i worked with pwl, the research was exclusively theoretical — no computing of any kind, and mostly banging one’s head against a brick wall until a theorem or proof pops out — whereas now a major part of my work involves coding algorithms; in contrast this is a fairly constant project in that there is always something tangible to be getting on with. the interviewers were contented by my assurance that the shift was one that had requested myself; their worry was i had been coerced into it.

i was particularly pleased that my inclusion [as an appendix] of a slightly modified version of my analogy for interpolation was commented on and appreciated — especially by the non-applied mathematician, since i was aiming for it to be understandable by someone outside the field.

so i’m [unofficially] through to my third and, as far as funding is concerned, final year of fudding — a sobering thought indeed, which must mean it’s time for a beer…

‘obvious’.

i don’t hate that word in everyday usage. in fact, i probably overuse it, to be honest. in maths papers though, the word ‘obvious’ gives me cold sweats — specifically, when used in this sort of context:

[from the paper i am currently trying to decypher in order to use the theory for my own purposes.]

only in maths can you spend half a day trying to figure out why something is completely, utterly, kick-self obvious.

of course, once i figure it out, i’ll not bother to write out the full reason — because it’s obvious.

everyone in the department just received an email from the secretaries with the subject ‘outing’. sadly this was not an exposé of the sexual preferences of various members of staff, but an announcement of an intended evening of drinks and a chinese meal to mark the last days of the department of mathematics and computer science at the university of leicester.

in a move fully understood by only a few members of senate, the department is to be rearranged as follows. there will now be a school of mathematics and computer science, and within that school will be two departments: the department of mathematics, and the department of computer science [or, the department of computing — the name is still under debate, being such an important point]. there is to be a head of the entire school, and two separate heads of each department.

i think this is primarily to do with budgeting issues — in particular, how the budget is allocated to each discipline — and having distinct departments makes this more transparent. however, there is a further divide between pure and applied mathematicians, in both the physical sense — the applied group are soon moving to a different building — and in terms of their budgetary needs. the computer science group is essentially very theoretical and, it seems to me, very close in terms of their needs to the pure maths group. so i’m sensing that in the near future we’ll have another split to form an official ‘applied maths group’ and ‘pure maths group’, each having their own allocated budget to do with as they will — with possible collusion between groups and departments — and thus each requiring their own head to oversea each group.

but within each group, there are definitely subgroups, some of which would like lots of money for big computers, others would prefer to spend the cash on inviting distinguished visitors over for extended stays. i can envisage this spiralling down until each member of staff has a personal piggy bank on their desk, which they shake out onto the table whenever they fancy splashing out on a new pc or pencil sharpener.

however, before all that happens, it seems the question giving the powers-that-be the most trouble is: how to share the three secretaries between the two departments. personally i’m in favour of getting another one and having two each, since they’re the only ones who ever get anything remotely administrative right first time.

currently wrestling with this conundrum:

• open the windows in the office, and get deafened by the noise of the building work going on outside;
• keep the windows shut and rapidly overheat from the combination of nice sunny day, ten computers & monitors in enclosed space and no air conditioning or fans.

at the moment i’m leaning towards secret option three: go and sit in the park and read a book.

you might be thinking that my recent lack of posts is because i’ve jumped on the blogging-hiatus bandwagon, or that i was in a huff because of a couple of nasty comments, but in reality i’ve been busy polishing off a king-size twix my second-year ph.d. progress report.

most of the trouble sprang from having two supervisors whose views on various sections of my report never quite tallied — if you just have one supervisor and they tell you to change something in a certain way, you can go off and do that without much bother [provided you see their point and agree to change your work]. with two supervisors, i had to take on board each of their suggestions and somehow make alterations which made both of them happy.

the suggestions themselves have been slowly iterating from the useful [e.g. ‘you should say more here’; ‘this isn’t clear’; ‘this is completely wrong’] down to tiny pedantic points [‘write “computationally expensive” rather than “expensive, computationally”’; ‘say “increasing the dimension” not “upping the dimension”’; ‘this is completely wrong’].

my own feeling is that, since the report isn’t an official thesis or academic paper, but is rather an internal document to satisfy the department i’m getting on okay, the language shouldn’t be as formal as it could be — a few less ‘hences’ and ‘thuses’, and not a hint of a ‘whenceforth’; i thought this report should sound like it was written by me, not any old mathematician.

however, after over a year of writing virtually every day in this blog, the hardest part was starting each sentence with a capital letter and not ending a proof with ‘swidt?’.

there’s an article in the june 2003 edition of mathematics today, the bimonthly magazine of the institute of mathematics and its applications, which looks like it could be very interesting — it’s entitled ‘how the human brain is endowed for mathematical reasoning: are numbers a product of evolution of brain on earth or have they an existence of their own?’

Read the rest of “technicalities”…

Read the rest of “warning”…

Read the rest of “things my supervisor and i have argued about”…

Read the rest of “code woes”…

Read the rest of “reportage”…

Read the rest of “quick heads up”…

Read the rest of “irrationality, transcendentality, and perfection”…

Read the rest of “kidding myself”…

Read the rest of “after the horse has bolted”…

Read the rest of “the a ect maths has on nicky printers”…

Read the rest of “other things to think about”…

Read the rest of “art dealing as a metaphor for maths”…

Read the rest of “a node by any other name”…

Read the rest of “eternal optimism”…

Read the rest of “backtracking”…

Read the rest of “talk talk”…

Read the rest of “going from good to worse”…

Read the rest of “mind not on the job”…

Read the rest of “tinker, tailor, mathematician, bollocks”…

Read the rest of “bug eyed”…

Read the rest of “a break from the norm”…

Read the rest of “thinking about thinking”…

Read the rest of “the tortoise and the (s)hare(d network drive)”…

Read the rest of “this is what it’s all about”…

Read the rest of “road to nowhere”…

Read the rest of “flattery gets you everywhere”…

Read the rest of “seeking justification”…

Read the rest of “getting busy”…

Read the rest of “tributes to pwl”…