Posts by Michael H.F. Wilkinson
1875 posts • joined Tuesday 24th April 2007 14:31 GMT
Page:
According to inflation theory
it was space that moved faster than the speed of light. This is perfectly allowable according to relativity. It is hard to get your mind around, I agree.
be careful!!
Someone may have copyrighted the phrase "Tick Tock"
Now astronomers have one more reason to detest astrology
only lawyers will benefit from this.
Keyboard, eat hot, er,.... tea!
The 60ns difference equates to 18 m
I trust there has not been an 18m shift in position, as that is earthquake magnitude. The mean temperature of rock beneath the ground is really stable, simply due to its HUGE thermal capacity, and high degree of thermal insulation. This is why (wine) cellars often have high thermal stability.
Previously, the peculiar, apparently superluminal motion of jets in quasars and active galaxies could be explained elegantly by invoking special relativity. by assuming the jet is traveling at near light speed almost directly towards us. Here general relativity may well be an explanation (but it may not)
Posted in Apple cofounder Steve Jobs is dead at 56
He will be missed most by those near to him of course, but many others will miss his undoubted leadership. Though I am not an Apple user, many Apple designs set new standards, and gave others a point to aim at.
I will join you in raising a glass to his memory and achievements.
Just two questions:
Playmobil reenactment?
Do these things come with kevlar reinforcements?
Posted in Put down the Java manual
I can see why you might take offense
but I read the "as well" as meaning that it is OK to learn Java, and "real programming" also can be done in Java.
On the topic of teaching programming: Some years ago we decided to use C in the initial programming course (Imperative programming). This allowed us to ditch the object-oriented overhead of Java which was used before. We go on to teach them algorithms and data structures in C, complete with structs containing function pointers. The latter paves the way for understanding what objects actually do for you. They also learn to clean up any mess they leave behind. Object oriented programming is then done in Java. Later courses include functional programming (I think in Haskel), and parallel programming and concurrency.
This is just the main programming track of the curriculum, other tracks focus on software engineering and architecture, database theory, operating systems, compilers, etc. Many of these skills might never be used in practice, but they do give better insights, and mean you have learned to learn difficult topics.
The best reason to learn a programming language is to learn new ways of thinking about problems. It does not help to learn 15 subtly different OO languages at university, it is important you learn different programming approaches. Once you have learned a style well, it is trivial to learn another. I learned programming in Pascal, the switch to C was quite trivial; I started OO programming in C++, learning Java was quite simple.
Posted in Premier League loses footie decoder case
Ruling makes sense
These are the kinds of rulings that let ordinary people believe there are benefits to being part of the EU.
Posted in October 14 declared 'Steve Jobs Day'
Darn!
I missed taking a towel to work on May 25
I missed saying "ARRH" on September 19
I think I will give October 14 a miss as well
Pirate, because, well AAARRRRHHH!
Posted in The plane story of galactic clusters
I thought they just made planets
but they might be branching out into whole galaxies.
And there I was thinking
they might be suffering from dyslexia.
Or they were in such a hurry to make money they forget the last letter of their nam.
having said that, their behaviour causes dyspepsia.
Posted in The plane story of galactic clusters
re: dark matter
I just love it when people come by and start saying: look scientists got it wrong in the past, current science must also be wrong. This shows they do not quite get the scientific process.
"Getting things wrong" is part and parcel of science. All current theories do is model the world in such a way that a large set of observations are explained. Testing theories consists of making predictions, doing experiments and seeing where the theory gets it wrong. If it gets things wrong, we have to make a new theory which explains all the old observations AND the new. Alternatively, there may be a mistake in the new observations, and the theory survives to be tested again. Each new theory offers a better approximation of the behaviour of nature, which should be harder to prove wrong than the previous. However, even if it mimics the behaviour of nature exactly, we have no guarantee it is the real mechanism behind nature, it is just a perfectly good model.
It is natural for scientists to be cautious when a theory confirmed by thousands of experiments is contradicted by a single experiment. At the same time many physicists are unhappy with the notions of dark matter and dark energy, and are looking for alternatives. Where there is disagreement there is progress in science.
Posted in AMD Llano vs Intel Sandy Bridge
"Intel hardware can't run the OpenGL test"
Why not, is the OpenGL support incomplete, does it lack required extensions?
If so, it is wholly unsuitable for me, alas.
or
Niccce fissshesssss, gollum!
@Blitterbug
I actually state that there are those who can teach themselves, but maybe I should have stressed that more. I studied astronomy, and am therefore largely self taught, when in comes to computer science. This is why I am still learning more about core computer science today.
Quite right
When I take on PhD students for projects with a lot of coding, I always look for passion, for people who do extracurricular stuff (not just in terms of coding, mind you). One problem is that 3-5 years is way too short to learn to code REALLY well. You typically need ten years (as for any real skill). A Uni can give you the theoretical foundation, you have to build a house on that by work experience (or hobby).
A nice site on this topic is
http://norvig.com/21-days.html
@J.G.Harston
We do teach computer science. There are IT courses taught by the Hanze University of Applied Sciences next door, but we, and all other traditional universities over here still teach computer science. However, very few of our graduates ever become coders. They become researchers (in academia or industry, a lot of ours go to companies like Philips Medical Systems) , or become software engineers and architects (OK, they are also involved in coding, but cost a lot more ;-) )
@Blitterbug
I actually state that there are those who can teach themselves, maybe I should have stressed that point more. In fact I studied astronomy, so much of my computer science and coding skills is self taught. This is why I am still learning new things about core computer science.
I also notice there are huge differences in IT degrees themselves. Some degrees are much more oriented to learning to use available tools and languages than to actual problem solving. We sometimes get students from these courses applying for our MSc course in computer science. These guys no way more than our students when it comes to common tools out there. Where they have huge problems in in their problem-solving skills. The best acquire theses skills, there rest flunks the course.
@CD001
Good point. We TRY to teach them, but then some just replicate instructions until they pass the exam, and then forget all about it. Some people are, as we say "resistant to education." At the same time there are many self-taught people who are excellent. Besides, there are many practical skills we do not teach (especially on the wealth of different tools out there).
The most important thing you can learn in education is learning itself. I have had no formal training in lattice theory, but I have acquired the skill set to learn it, and now contribute to the field. Many of my coding and debugging skills I learned in practice, in my first job as scientific programmer (developed a 150 kloc image processing system). That is when theory gets turned into practice.
Play "Air Traffic Controller" with added realism !!
I'll get me coat
Now that is cool
I'll doff my hat
(the Tilley today, the Barmah is too hot in this weather)
Later today, I will no doubt raise a glass as well
Re: Rather ironic
I stand corrected.
It is a good thing I don't teach language ;-)
But,
WON'T SOMEONE THINK OF THE CHILDREN!!!!!
Surprised that did not pop up in this context.
I taught programming
and have seen quite some horror's produced by various "bedroom coders". They do get things to work, but the code is often not an "oil painting". There are certainly those who can teach themselves, but there are those who do benefit from learning a more disciplined approach.
The lack of discipline can be really astounding in some. There was one guy who insisted he wanted to hand code in in C# rather than Java. I told him of course he could hand his assignment in in C#, so long as he did not mind failing the course. He found this unreasonable. I suggested my attitude reflected that of a potential employer or customer, who more often than not have some requirements on programming languages, coding style, comments. If you hand in your work in a different language, you would be in breach of contract, or get fired.
He still thought I was being very unreasonable.
Another story I like is the guy who handed in an iterative solution where the assignment explicitly stated: "implement a recursive method to compute ....." He argued this was more efficient, we said that was true, but that the assignment was to learn recursion. He said but my implementation is more efficient, we said that was true, but that the assignment was to learn recursion. He said but my implementation is more efficient, we said that was true, but that the assignment was to learn recursion. ..............................
This went on a while until we terminated this infinite loop (not by kill -9, but more humane methods)
This is not to say the tuition fees aren't outrageous. You can get a much more favourable deal in the Netherlands, and the university I work at (Groningen) is drawing more and more students from the UK. Our MSc courses are English language anyway, and our BSc courses are headed that way as well.
They would say that!
After hearing it, they just had an uncontrollable urge to ......
Oh, wait.
A lot of people seem to be jumping the gun a bit. Firefox have not said they are going to block Java, they are considering the balance of usability vs security (as they should). It could also be a shot across Oracle's bows to get them to fix the hole in Java. As far as I can see nothing has been decided yet.
Nice stuff
We have some workloads coming up which require SERIOUS i/o bandwidth. The TMS cards look interesting indeed.
That is often true
The USPTO has a mixed history in this respect, to put it diplomatically.
Posted in Sony Vaio SB 13.3in notebook
Not a replacement for my ageing SZ
until AMD/Radeon gets its Linux drivers sorted. Pity really
So it has got the extra leg
When will it sprout the extra head, and develop into a new species which has to be named
Bufo beeblebroxii
?
Phrenology?
Why not retrophrenology? You would just need a massively parallel hammer
In the words of Deep Thought
"The Milliard Gargantubrain, A mere abacus"
(RIP Douglas Adams)
Or just use Parson's steam turbine
The Babbage engine in turbo mode
The difference in hours is explainable in core-collapse supernovae, because light gets absorbed and re-emitted many times before reaching the surface, whereas neutrino zip through the dense core matter. The currently observed effect would result in 4 years difference even in a nearby SN in the Large Magellanic Cloud.
"Someone actually gave you a thumbs up for that?"
May have hit the wrong button ;-)
One problem is the 1987 supernova
The neutrino burst from this was detected 3 hours before the first light, which was expected because the neutrinos travel from the collapsing core to the outside much faster IN THAT MEDIUM than the light. This is because the photons are constantly absorbed and re-emitted by all the nuclei and electrons in the dense stellar matter, whereas the neutrinos move like ghosts through everything. If the current difference is correct, the neutrinos should have arrived 4 years before.
Now these neutrinos may behave differently than the ones observed in OPERA, but it shows not all neutrinos move faster than light.
Re:Dude
The Dirac equation is a Taylor approximation of a relativistic Schrödinger equation. The latter is a non-relativistic equation. QM arose independent of Relativity (see Planck), and it is still not possible to combine general relativity with QM.
I gather the result was six sigma
60ns measured, vs 10ns error. They also repeated the measurement 15,000 times. I agree there is room for systematic errors in the OPERA experiment, and we should simply wait for confirmation (or the reverse) from independent experiments.
It ain't over till the fat lady signs
Nice boffinry
It would seem fitting that a game of ping-pong heralds a new type of computer.
Posted in Acer Iconia A100 7in Android tablet
Sounds interesting. I might give this one a more detailed look.
Posted in Mars trips could blind astronauts
My thoughts exactly
Does require an even bigger craft and thus even bigger rocket. It does seem to be a simple problem compared to shielding against cosmic rays and the like.
I always expect the world to be abnormal
mine is the one with the nonparametric statistics book in the pocket
There is tangible proof of a massive impact
exactly on the K-T boundary. Paleontology that, not climate science. You are right in saying that the exact mechanisms of the extinction and the climate change as a result of the asteroid impact are speculative, but there is proof of massive forest fires in the fossil record. The estimate of the asteroid size (approx 5-10km, estimated from the impact crater remains) combined with the fact that the iridium-rich layer at the K-T boundary is found all over the world is powerful evidence that something affected the whole world just at the point dinosaurs went extinct. Bit of a coincidence that.
So the result from the break up could still be out there.....
RUN FOR THE HILLS!!
Oh, wait.
GET BRUCE WILLIS INTO A SHUTTLE,
Darn, that wont work
WE'RE DOOMED, DOOMED
always a trusty fall-back
Explosion, because well ....
That is fairly automatic these days
The Earth's rotation is the big one, but even my little equatorial mount can correct quite nicely for that. I think the 2.2 m is also on an equatorial mount. Nowadays, autotrackers are available for amateur scopes, so this big kind of scope should be fine. The Earth's movement around the sun has very little effect, the parallax created in one night or a few days is smaller than the point spread function of the optics.
Actually, the hard thing is to get a wide field
No zooming required with a telescope, but a big (268 megapixel or so) camera to get this wide (OK maybe it was not the 268 megapixel OmegaCAM, but still a biggie). The other hard part is collecting enough data, as these objects are faint. Hours of exposure are needed, and postprocessing to remove artifacts introduced imperfections in the camera (dark-current, hot pixels, slight differences in pixel sensitivity, and shading by the optics), and finally dynamic-range compression to show both he brighter and fainter parts in one shot.
