Posts by John Brookes 54 publicly visible posts • joined 23 Aug 2007
I don't think you could claim prior art because someone said 'it would be cool if X' when someone later patents the most natural method of achieving X.
They might be even further ahead of the game and think, 'if they want X then they'll need to Y', or 'once they've got X, the natural next step is Z' and then patent Y or Z.
...iPlayer Radio now has 1mth catchup. I'm pretty sure it was 1 week before the weekend....
Both of those statements are based on the 'days left to listen' bit below each listing in the category sectionof the android app...
Can anybody with a better memory than me confirm it changed at the w/e?
No, I wouldn't count it as a supercomputer any more... but....
This could be major boon for a researcher at a small institution struggling to cram in sufficient simulations using a threaded/OpenMP'd application on a workstation, or a miniscule amount of CPU time on the department's/intitution's resource..... Or a researcher whose papers stand a high chance of rejection due to their sims' low resolution (or other over-simplification of the system under study).... Or the researcher whose research is at an impasse because there's just not enough RAM to contain his/her system at all..... Or where IO requirements are so overwhelming that having (N<6)TB of RAM disk yields 10^x speedups.... or, or, or....
Sure, it's a publicity stunt, but it could be an invaluable one for a number of poor schmoes on the coalface of science....
> The only devices I tolerate this from are for children - there is has a function, and they have an off switch.
My children have been gifted toys that appear to have an on/off switch, but on closer inspection that switch is revealed to be an on/demo switch.
'Demo mode' *can* mean, "I am going to bleep constantly until it's time for you to spend more of your hard-earned on batteries." That makes it an 'on/even more on' switch...
That is clearly evil, but not as evil as when it's 'On mode' that signifies 'incessant' and 'Demo mode' means the device randomly - or, through use of sensors, pseudo-randomly - makes its irritating array of noises. Annoying during the day, eerie through the night and (in the case of a toy police car I *told* the Mrs not to pack) utterly terrifying when driving through the backwoods on a country road in the pitch black.
They get binned as soon as the sprogs fail to notice when it's been locked in the outhouse for two days. Can't do the usual charity shop donation in good conscience....
...in Bristol, many moons ago (2001?).
I was walking through the Bearpit, when a huge guy on a (comparatively) tiny BMX came up and demanded my phone. I sighed, he put his hand in his coat pocket threateningly, I explained that I wasn't going to give him any problems, it was just a PITA.
"Actually," I continued, "do you mind if I just pull the SIM - that's not what you're after and it's got all my numbers on it..." He acquiesced, so I pulled the phone out and began the process.
"Hang on, mate... What's that?" "An Ericsson T39", I replied. "Oh, you're alright then...." he tutted and began to cycle off.
I was indignant - for it's day, it was a seriously nice bit of kit. I didn't chase after him, but I did enquire as to why he didn't want it. "Only after Nokias, geez; can't get the chargers for those..." he answered over his shoulder as he rode off.
Saved by a proprietary charger, hoorah! :-)
It isn't.
I know doctors, lawyers, builders, etc. and they all report the same.
Working in a certain profession means that you *will* (or, at least, *should*) know more than the general milieu about the subject. The problem is that most people (in whatever profession, or none) underestimate the diversity and complexity of other people's professions. People ask eg criminal lawyers for advice about their ongoing dispute with their neighbour over property boundaries (it happened to a friend: he told him to punch his neighbour in the face. "How will that help?" "It won't, but then I'll be able to tell you how you stand legally.")
There's even a joke about it:
A doctor and a lawyer are at a party. The doctor asks the lawyer, "People are always bothering me with questions about their health; should I bill them?"
"Absolutely!" replies the lawyer, "They're asking you for services, based on years of training and your standing as a medical professional. Of course they should be prepared to pay for your insight!"
Two days later, the doctor received an invoice from the lawyer.
I can't see any particular reason for neutrino beams not being radiation, but the rest of his point still stands. I guess that if it is not considered radiation, it could only be because of its vanishingly small interaction cross-section....
Sticking the experiment in a mine is required to (hopefully) keep the SNR manageable. On the surface, you would just have a really rubbish cosmic ray air shower detector.
If I buy duty-free etc, then I am not liable for tax on that purchase. Voda *were* liable for £6bn, but unilaterally decided not to pay it. The tax office then just let them off for the vast majority of it.
Despite their claims that their tax bill should've been zero, they'd set aside £2.2bn to pay it, so they knew - or strongly suspected - they'd lose if it went to litigation.
I can understand the tax office being wary of a fight with Voda, given the latter's willingness and ability to fight the claim, but we didn't even ask for the £2.2bn they'd set aside.
"Give me that £60 quid you owe me!"
"Or else what?"
"Er, call it a tenner, then..."
"Oh, go on. Got change of a twenty?"
....we should all convert and become Amish? No thanks - not keen on curtain beards....
Seriously, the whole "have fewer babies and you'll be better off" thing devalues anything else they say. Such a lack of awareness of the difference between correlation and causation utterly discredits their claim to be scientists.
http://xkcd.com/552/
...tell you how you may, or may not, use said device, then whinge about the cost and the security implications?
Who actually wins out of BYOD? The employee gets to buy their employer a deliberately-crippled device, the employer doesn't save any money, but does add to its count of disgruntled BOFHs, RIM circles ever closer to the drain, Apple/Samsung/HTC etc don't get any additional revenue (unless the used-for-work device becomes so crippled that the employees have to carry two phones anyway). What's it all for?
It's been irritating me for ages that there seems to be little traction on stereo headsets. Yeah, there are a few about at any given time, but half fo them are crappy in-ear things with dangly bits (not that sort :-) And I've never seen any in mobile phone shops / general electronics shops / Tescos etc
Most phones these days have A2DP, most PMPs are, in fact, mobile phones, yet most headsets are mono... Why? I genuinely don't get it. If you have a mono headset, you have to have a pair of plugins as well, if you're going to listen to music.
It's nearly as weird as being as bothered about it as I clearly am :-D
As a wise friend once said to me, "Just because you are sure, doesn't mean you're right"
It doesn't imply any such mass wipeout of tall people: that *would* be a nonsense. What it means is that over many generations, taller people would be less likely to reproduce, leading to a gradual reduction in the average size of members of the species. Obviously,
How might this come to pass? Well, given the self-evident tendency of surface area per unit volume to decrease with increasing volume, larger animals find it more difficult to remain cool if the temperature increases - not impossible, just more difficult.
That implies less activity for larger animals, since activity produces heat that - unless removed - might cause dehydration/heatstroke/.... Less activity implies a reduced ability to support a family than a smaller but otherwise identical animal (less foraging time, smaller hunting range, or ...). Smaller families for (genetically) larger members of a species leads to a gradual diminution of the average size of a species as a whole.
Doesn't sound too implausible to me.
There are three parts to this*:
1) We don't know whether gravity treats matter and anti-matter identically (eg whether they're identically affected by the Higgs field, or <insert gravitational theory here>)
2) We know that electrons and positrons have very similar masses, but not that they're identical.
3) In a direct annhililation, the tiny fractional differences in mass would be utterly swamped by uncertainties in the measurement (both experimental and fundamental). In the proposed experiment, if the lifetime (and hence beam length) of the positronium can be made sufficiently large, it may be possible to reduce the uncertainties to such an extent that the differences are measurable.
* I don't see how the proposed experiment could disentagle the effects of 1) and 2), but maybe that's why I can only call myself a physicist if I add the prefix 'failed'.
Auto-detection is _very_ hit and miss for me (despite the fact that it always has the same ip address) and the remote is for a slightly lower spec TV, so misses a couple of nice features.
The latter issue isn't so bad, but a remote is worse than useless if it may, or may not, be able to use it. You should be able to manually input an IP address, or save the configuration between sessions.
not Hobson's choice, not punch-in-the-face-or-kick-in-the-swingers type choice etc... A choice between options that have real (meaningful) differences in outcome. Esp. as it wasn't given as a modifier at all, but rather as an additional, orthogonal property of the choice.
...over muppets who advertise their own self-contradictory mumbo-jumbo. I've just read your articles. Yeesh
You do realise that many of the denizens of this place are scientists, engineers and other sorts of folk you can't abide?
Some standard steps in science according to the world at large:
1) Make a new observation. Check it.
2) Get other people to see if they can see it too
3) Postulate an explanation for what you've seen. Check it.
4) Get other people to see if your postulated explanation works with their related, or independent observations
5a) Try and make a prediction using your explanation that differs from prevailing explanations - see if it happens
or if 5a not possible
5b) Use your explanation and existing data to attempt to accurately post-dict real events
6) Repeat 5 until dead, retired or bored of the topic
If you fail at any point, go back to 1)
Scientific method according to Louis Savain:
1) Have a particularly vivid dream
2) Convince yourself that it is both right and obvious
3) Have a quick google to see if you can find any off-the-cuff remarks from long-dead philosophers, natural philosophers or scientists that don't entirely contradict your dream
4) Publish (NB self-editing for blatant contradictions is not only optional, but to be discouraged)
5) Slag off anyone who contradicts you
If you fail at step 2, you're not trying hard enough, if you fail at 3, just skip it.
From Nobel's will:
“The said interest shall be divided into five equal parts, which shall be apportioned as follows: /- - -/ one part to the person who shall have made the most important discovery or invention within the field of physics ...”
"... as judged by El Reg commentard TkH11", it fails to continue.....
The importance of discoveries and inventions is judged by the prize committees. That's why they exist.
...an easy test.
How can I say that?
1) Go to top500.org
2) go to a list - any of them
3) go to the statistics for that list
4) view the stats by interconnect/interconnect family
Basically, you can set a reasonably accurate lower bound on HPL efficiency for your particular arch/interconnect/topo pretty easily. Run it, see what you get, compare and contrast....
Firstly, when HPL was designed, it wasn't pointless. It was pretty much the only portable benchmark that tested the capability of contemporary supercomputers to run an important class of problems.
Secondly, it's still not pointless, unless CPU speed has suddenly become irrelevent.
Thirdly, it's an excellent sanity check when throwing a cluster together:
- HPL efficiency a bit low? Hmm, let's run some interconnect tests
- Good check for excessive single-bit errors? Run HPL cluster-wide, as big as will fit on each system and see what gets reported.
Fourthly, EVERYONE knows the list is primarily used for business/marketing/dick-waving purposes. So? "Is vendor X's response to our RFQ any good?" "Yes, but they haven't built a machine on this scale since 2009." or "Yes, and I can see from the Top500 that they've got many machines in our sector at the scale we're looking." or.....
Fifthly, Jack Dongarra (et al) are perfectly aware that HPL isn't the be-all and end-all of benchmarking. That's why HPCC exists (and JD is the chair of its pretty stellar committee). Oh yes, and HPL is part of that improved benchmarking suite.
Sixthly, all benchmarks that aren't your application are - to a greater, or lesser degree - pointless. So let's not benchmark anything, eh?
Still, I'm sure your achievements dwarf those of Prof Dongarra....
They're comparing to the host-to-goodness, full-fat, speed-limit-o-the-universe in vacuo speed....
A better comparison would be with the apparently super-luminal velocities of the plumes of active galactic nuclei 'observed' a while back... Further (non-trivial) research showed that actually, all was right with the cosmos and it was illusory.
Hmm, I could either pull up a reference for this, or down tools and head to the pub.... Guess which is likelier
@Mikel - You're right in your area of interest, but the point of the article was that how appropriate blades are for you is application- and datacentre-specific. If you're hammering the cores and memory constantly, small deltas can mean significant spot heating, leading to throttling, leading to crappy performance - if one of 1000+ cores is throttling during a tightly-coupled parallel job, you're wasting half your machine. Which might only last half as long - or, equivalently, require twice the maintenance -because you're pushing the envelope. Datacentre (thermal) environment is therefore a crucial, expensive and non-trivial consideration.
@danolds,the cube - you're both right: water-cooled rear doors CAN do a great job, but you DO have to be damn sure that the fans can pull (at the very least) sufficient air through. If they can't, they're worse than useless due to internal recirculation.
Oh, and speaking of doors: for maximal-density, heavily-hammered blades, front doors considered silly.
Not necessarily. Instead of meaning:
a) They will only buy things that they are convinced will not improve performance
they might mean:
b) They will only buy things that they are not convinced will improve performance
or (and this is my personal favourite):
c) They definitely won't buy things that they are convinced will improve performance.
As you say, got to love triple negatives :-))
Anyway, it's half six
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
