Slashdot Mirror
The BBC's Distributed Climate Prediction
CongoJoe writes "The BBC has teamed up with Oxford University to conduct the world's most ambitious climate modeling experiment." From the article: "Trying to predict climate change is hard. There are lots of factors involved - air temperature, sea temperature and cloud cover all play a part - as do dozens of other variables. Therefore, there are a huge number of calculations involved ... Using a technique known as distributed computing, we're hoping to harness the power of thousands of PCs around the world. If 10,000 people sign up, we'll be faster than the world's biggest computer. And we're hoping to be even better than that."
if those 10,000 people turned their computers off, when not in use, instead?
Not that I don't think this is a good investment of spare cycles. I'm just wondering what the power savings would be, as an alternative.
Also, I notice there is no OSX client, only Windows and Linux.
I am concerned about sending my spare CPU time overseas, i mean, they say they are "climate modeling", but could they not just as easily be selling this "distributed" computer network to the black market, or Iran, or doing nuclear weapons simulations?
:)
I mean, they ARE british, and you all know what happened last time they got their hands on a bunch of CPU time- that poor kursk...
How bout the dental hygene simulator project?
sorry, just had to rip on some of my brit friends today
They are hoping for only 10,000? SETI@home has 50 times as many participants.
What I want to know is: Why haven't we seen a distributed computing application implemented in javascript and embedded on pr0n sites, or hell, even in flash ads or something. The user's machine can chug away at some small task for a little while and report back the results when the tab is closed via HTTP.
And when are we going to see a real market develop for raw computing cycles? How many $$ per teraflop-hour?
You heard it here first...
Case/PSU: Asus TA-211
M/B: Asus K8V-X SE
CPU: AMD Sempron 2800+
This climate prediction group has been going around 2 years. All that has happened is they have managed to talk the BBC into some free advertising. I tried running the software but I found that it caused localized warming especially around my CPU.
I used to have a better sig but it broke.
At least I guess its a new version of this, ran the cpdn client for a few months ages ago (must be well over a year ago actually). I eventually stopped because I have a variable speed fan, and the cpdn client kept my cpu working hard enough to spin the fan up to 5000 rpm, which is a bit noisy. Normally that only happens during heavy gaming.
Oh no... it's the future.
Why do I doubt that 10,000 garden PCs offering up idle time will compete with the current t500 leader (280.6 TFlop/s, 131,072 processors)?
Seeing as even a moderate sized cluster is say a dedicated 256 nodes, surely this isn't going to offer anything more than loose change. 1 million PCs, chipping in 5% of their time sounds useful.
jh
Weird how it doesn't show up on the page the BBC wants you to go to, though.
So no, the BBC experiment itself doesn't have an OSX client.
Hmmm, creepy.
I just saw the story now, never heard about this thing before.
I download the software, I install it, I run it, I select "new user", enter my google mail as mail to be used... and surprise: "that email is already in use and you entered a wrong password".
And no, I am 100% sure I didn't do it before, and there's nothing in my e-mail about it, and the passwords I use for "low security" and/or "disposable" things is completely different from my e-mail password, etc.
So, that tells me one of two things:
- either they don't verify the email address AT ALL (why the heck do they require one then) *AND* somebody used MY email address to enroll (wtf)
- or their software got bugged (or funnier, slashdotted ?)
Anybody else got this problem, or is it just me ?
By reading this signature you agree to not disagree with the post you just read.
Sure it is, if you're going for a really high degree of granularity. Town-by-town climate change, just where along the coastline that hurricane is going to hit, or whatever. Modeling on the larger scale is probably a lot more feasible given their resources.
But then, there are plenty of people doing larger-scale modeling. And there have been, for years. And it's not like they keep their results secret. So we can just Google them.
(Insert gratuitous "Doesn't anybody use Google anymore?")
Village idiot in some extremely smart villages.
You know, I would actually consider running this if the BBC didn't have such a reputation for being nationalistic information hoarders.
Time and again they release something cool, a new media archive for example, and trumpet it far and wide only to say in the fine print that it is only for people who live in the UK.
How do I know that they won't take the work of my processor and only allow people in the UK to view the aggregate results? "I'm sorry, but the final assembly work was paid for by our TV tax, which you don't pay, so go away."
There are a number of things TFA doesn't tell you, or misleads you about.
1) The Climate Prediction experiment has been going on for several years now, first as a standalone application (like Seti@home), and now as both as a BOINC (Berkeley Open Infrastructure for Network Computing) application.
2) There are multiple BOINC [distributed computing] projects with greater than 10,000 users (see here, and here.) - thus even if the BBC meets it's goal of 10k users, it will still be far from the largest.
..that people applaud something like Mersenne's Prime or SETI when spare CPU cycles are concerned,
...
but when an action to the benefit of everyone - namely checking up on the environment - is undertaken using the same technology that those same people start commenting on artificial climate change:
"what about the power usage", "this will cause more rapid climate change" etc etc.
Is it either that
a. those people believe the climate WILL change through use of more computers
1. but do not want to know about it
2. and feel that any action towards proving it further is useless
b. are in the climate-does-not-change-artificially group and will take any reason - even those undermining their own stance - to support it because....
1. they do not want to know whether they are wrong
2. feel that there is enough proof already that the environment is not changing
Could anyone explain the logic behind the reasons A1-B2 ?
Slashdot: stuff for news, nerds that matter, matter for news, stuff that nerd
Help make the slashdot team the biggest team on the planet. Watch the team progress grow. You know you need to.
This is discussed in their help section on the site. I'm getting the same thing no matter what email I enter, including ones I can guarantee have not been signed up. [I also signed up for the discussion boards on the site, but the confirmation email hasn't come yet... and it's been about 1/2 hour.]
Apparently that's just the response you get when the servers are overloaded and can't process the registration.
Just close it and try later.
this Sunday at oh dark 30 (ok, 5:30 am PST, but I like to sleep in until 10 or 11 am) on this very subject.
Global Warming - it's a really hot issue at the Beeb.
Plus, as a bonus, it has absolutely nothing to do with certain cartoon riots, and they get to avoid talking about the whole smoking ban that's also causing ill feelings.
That, plus the fact that Seattle is famous worldwide as Green central for US cities (hey, I know it's not true, we killed the monorail and all that, plus we drive a lot, but they actually believe the spin we put out).
-- Tigger warning: This post may contain tiggers! --
Maybe they're going to try and top 11C warming that they managed last time.
It should be "How I bombed the atmosphere with carbon dioxide and waited for the headlines"
Tubby or not tubby. Fat is the question
In a long term ClimatePrediction.net user let me answer a few questions:
1. the extra electricity used is insignificant in global terms. In winter the extra heat generated means you don't need to keep your central heating so high. Super-computers use lots of electricity themselves so running these same simulations on one (actually would take dozens) would not be better.
2. We are currently at just under 50,000 users, sounds a lot but there are millions of potential models to crunch - although useful science can be done with a smaller number. I would expect over 100,000 users to enrole in total. Some users are experiencing problems, they seem a very small percentage of the total judging by the posts on the help lines. Many of the problems are due to trying to run it one machines not powerful enough, climate simulations are heavy duty programs and need a beefy machine to run in a reasonable time.
3. The exeriment is an 80 year hindcast + 80 year forecast 1920-2080, even if a model does not complete the hindcast will be useful. The second BBC program is scheduled for May which is the "end" of the project, in pratice the scientists will be able to use results reported for at least a year after than. Its the science that counts.
4. The data is not being hoarded by the BBC, it is kept by the ClimatePrediction.net team, and is available to scientist throughout the world.
5. in a few days the ClimatePrediction.net servers will start dishing out TCM models to their users, that will add another 45,000 odd machines.
6. ClimatePrediction.net is in the midst of a sulphur cycle experiment, which compelements the 2x CO2 doubling and THC slowdown experients. There is at least one other experiment in beta.
7. ClimatePrediction.net has already had one major paper in Nature, as well as many others, this is the distributed project that seems to producing the best science.
Technology is the single biggest human factor that we have reason to believe affects climate; it is the primary whipping boy for the "we're killing our planet" hysterics, after all, yet none of these "studies" can even make a start at predicting what is going to be the motive and/or non-motive power technology set du jour in twenty, fifty, two hundred years. Although we do know that it almost certainly won't be oil, because there won't be enough left to use economically for such purposes. Both power transport/storage and method of generation are questions that are totally up in the air. For generation, fusion? Fission? Solar? Tidal? Trans-dimensional down-level leakage? Other? For storage and transport, hydrogen? Ultrabatteries? Other? We have absolutely no idea. What about transport efficiency? Will friction continue to be an energy sink? Maglev? Vacuum containers? Anti-bloody-gravity?
And then there are issues like volcanic activity, solar activity, the Earth's magnetic field is weakening...
Predicting climate without knowing what all these major inputs to the system are going to do is like trying to predict the weather without wind, humidity and temperature information. A random number generator would likely do as well.
The whole thing is an exercise in naval gazing and cynical grant-acquisition.
Of course, it does make a great distraction from the real issues of the day such as multi-national erosion of civil rights, the United States' current attempt to inflict an empire of democracy upon nations that are operating under other systems, starvation in large portions of Africa...
How... convenient.
I've fallen off your lawn, and I can't get up.
Suppose the Republican Party graciously donates the spare cycles of all the computers at its headquarters, and the results magically turn out to mesh with their official line that there's no such thing as global warming?
Step into a huge movement. Don't Tread In Me.
thanks for the link!
-- Tigger warning: This post may contain tiggers! --
I am concerned about sending my spare CPU time overseas, i mean, they say they are "climate modeling", but could they not just as easily be selling this "distributed" computer network to the black market, or Iran, or doing nuclear weapons simulations?
You could always donate your CPU cycles to help Fold Proteins at the Baker Lab here at the University of Washington.
Then you'd know your CPU cycles were doing good American scientific research.
And Prof. Baker is a neat guy, very slashdot.
-- Tigger warning: This post may contain tiggers! --
Sorry, here's the Baker Lab link, for the Biochemical Structure prediction effort, folding proteins with your home computer. They also have another Beta for a different project as well.
-- Tigger warning: This post may contain tiggers! --
While you have a good point, in that it might be used for restrictive medical patents, as a Bioinformatician I've actually used some of their genetic libraries for pattern matching as part of open scientific research funded by NIH, NIAID, and now NIA.
So, since the BBC is just the front for a great scientific college, I'd be ok with that.
You could always help out the Baker Labs here at the University of Washington, folding proteins with your spare CPU cycles. The protein folding is for various projects worldwide, the vast majority of which are all public research, and hence not used for restrictive patents.
-- Tigger warning: This post may contain tiggers! --
It was the "Baker Labs" thing I was referring to; I went on a look-around on the website, but nowhere could I find addressed who was going to own the results, or in what form they were going to be made available. And "Public research" is of course not a guarantee that the results will actually be avaiable on a royalty-free or similar basis.
Trust the Computer. The Computer is your friend.
I agree, there is no guarantee.
I can, however, as someone who worked on the same floor as the Baker Labs - they're in J wing, I was in K-wing, and David Baker's been at many Biochem seminars that I've been to, attest that the vast and overwhelming (something like 90+ percent) of Protein Structures developed by Folding Predictions are for other open research, usually funded by NIH, NIAID, and various equivalent groups in places like the UK, where the research is published in Science or Nature or Cell and they're not used for restrictive patents.
Is that helpful?
I'm now over by Lake Union, but my girlfriend works one floor down from the Baker Labs at the UW, so I can attest it's pretty above-board.
Will, formerly of Biochemistry, now of Medical Genetics (Dept of Medicine)
-- Tigger warning: This post may contain tiggers! --
Anyone who thinks you can create a predictive simulation without critical data that is KNOWN to modify the situation needs their head examined.
I've fallen off your lawn, and I can't get up.
I'm now over by Lake Union, but my girlfriend works one floor down from the Baker Labs at the UW, so I can attest it's pretty above-board.
Don't get me wrong; I'm sure it is above board, and from all I see it's good people running the place.
But this is rather like the issue of software licensing. You can have a great group of people doing wonderful work, and you know they will share the work right back with the community. But people do leave or get replaced, companies get bought up and so on and so forth. We've seen people's contributions disappear into commercial black holes before when the work was done under the "trust us" model (the CDDB database is perhaps the most widespread example). The lab may be a wonderful place, but they do not - for example - have control over UW:s policy on research results, and if they determine that all results must be fully commercialized there is probably not a whole lot the lab can do. I'm really not comfortable with things like this not being spelled out in writing.
Trust the Computer. The Computer is your friend.
Don't get me wrong; I'm sure it is above board, and from all I see it's good people running the place.
...).
But this is rather like the issue of software licensing. You can have a great group of people doing wonderful work, and you know they will share the work right back with the community.
True, but in the end we come back to the scientific dilemma of sharing code (open source) and information (public research science).
Anyone can file a patent, but it's a lot harder to do so if there's prior art - and when you help an open project like what the Baker Lab does, there's a much greater chance that that will create prior art, and hence not lead to restrictive patents.
David's a nice guy. He runs a good lab, and they do good work that a lot of other good scientists use, but I can't guarantee anything since he's not me.
But, on the other hand, you could just send him an email or phone him, he's in the online directory of staff, students, and faculty. Just look up David Baker and send him an email or click on the entry and get his full mailing address, UW mail stop, and phone/fax - I'm sure he'd be glad to answer questions directly (naturally, since he's the PI of the Lab, and has a teaching load of a Professor as well,
Maybe he can clarify it further.
-- Tigger warning: This post may contain tiggers! --
Anyone can file a patent, but it's a lot harder to do so if there's prior art - and when you help an open project like what the Baker Lab does, there's a much greater chance that that will create prior art, and hence not lead to restrictive patents.
David's a nice guy. He runs a good lab, and they do good work that a lot of other good scientists use, but I can't guarantee anything since he's not me.
I think we may be talking past each other a bit here. Again, I have no doubts about him or his lab; that's not what worries me. In fact, it looks like just the kind of place I'd want to work in. To put it this way, even if you were he - even if I were he - I would still not trust this fully. He (you|me) is not in control of what his superiors may get into their heads to do if the project turns up something that is economically hugely valuable.
The best way to dispel any such nagging doubts is to spell it out, in a binding document, what the project may or may not do with the results of peoples' contributions (and I'm not necessarily asking for some hugely liberal, "anyone can do anything" kind of license eiter; just that whatever will be allowed is spelled out).
There's plenty of worthy projects out there, and choosing between them, the lack of a formal statement is just the kind of thing that would make me look for something else. Just like for software, in fact.
Trust the Computer. The Computer is your friend.
"The whole thing is an exercise in naval gazing and cynical grant-acquisition."
You may have faired better had you bothered to find out how the model currently caters for the things that you are complianing about and what it is they are hoping to achive with the experiment.
Calling internationally respected scientists cynical money grubbers based on your own false and myopic assumptions is definitely flamebait and I applaud the moderator for their good judgment.
And did you exchange a walk on part in the war for a lead role in a cage? - Pink Floyd.
So, Sparky, you think the BBC and its co-modelers know what technological changes are coming. What energy sources will be in use. When this will happen. When the next change comes. Etc. And you are assuming all of this is plugged into the model, apparently. You think I should check to see if the modelers have "properly" predicted the future of all energy technologies and incorporated that information before I dare put forward any critique of the lordly and all-knowing, future-predicting, BBC's master plan for ten thousand plus computers.
You're actually right on one thing. I should never have said anything. There's rarely any point in bringing sensible commentary to a discussion that is dominated by hysteria and ignorance — as you clearly demonstrate.
I've fallen off your lawn, and I can't get up.
The only hysterical (both meanings) one is you. For a start the BBC is pushing the experiment not designing it.
The UK MET office models make various assumptions about the effect of future tech on CO2 output, particulates, volcanic activity and a gazillion other things that you would not understand. The models make predictions on both past and future climate using a massive number of senarios with different inputs for the variables you speak about. The models themselves are not that difficult to research, there is a wealth of peer reviewed material on the web that looks at the MET models in particular, not to mention the same techniques are used in many other areas of research that have nothing to do with climate.
From the very start you poo-poo the whole thing without having a clue what it is about and then complain at being modded down. Either come up with an informed critisim or put up with the justifyable moderation, you can't have both.
And did you exchange a walk on part in the war for a lead role in a cage? - Pink Floyd.
I think we may be talking past each other a bit here. Again, I have no doubts about him or his lab; that's not what worries me. In fact, it looks like just the kind of place I'd want to work in. To put it this way, even if you were he - even if I were he - I would still not trust this fully. He (you|me) is not in control of what his superiors may get into their heads to do if the project turns up something that is economically hugely valuable.
True, but actually his lab provides Protein Folding prediction software to other research projects, so he has even less ability to stop people from using it for nefarious (patent) purposes. Anyone can submit their data to the lab and get predictions from it, and that's the beauty of it, it's open.
It's your call if you wish to participate. I can only tell you what percentage of submitted works are for public research (non-patentable, since they become public), there's no real control over who uses it.
At some point you have to trust that science will be used for good by most scientists - sure, there are unethical or profit-driven scientists, but they're a minority in my experience.
-- Tigger warning: This post may contain tiggers! --
Finally, covering a "range" of factors means that the output of the simulation covers a *range* of outputs, but it still does not mean that the simulation predicts what technologies will be in use. "A range" does not equal actuality. It's just a subset of possibilities, not a one of which can be assured to come to pass and where the entire set is no more likely to contain the actual set of circumsstances than anything made up by an 11-year old Trekkie.
The bottom line is that you can posit as complex a simulation as you like, but if you leave out major influences, you're just hand-waving. And there can be no question that major influences are being left out.
Why? Because there is a huge range of new possibilities and paths opened up by every new energy technology, every new measurement technology, every new transportation technology, every new waste disposal technology, every new recycling technology, every new genetic advance, every additional level of understanding gained in chemistry and physics, the sudden appearance of AI could change *everything* from who is working to how work is done... the list here is literally endless and any of it will knock these models not just askew, but completely out of any possible realm of applicability. And not just because of the individual changes, but because these change interact. They do so in ways that cannot be simulated or predicted; For instance an unanticipated advance in recycling energy may affect power production in a major way. Or not. Likewise energy storage efficiency. An unanticipated advance in genetics may change how livestock passes gas. An unanticipated change in public sentiment towards genetic manipulation may cause such an advance to have no effect.
Now, I fully expect you to come back with another "YOU'RE AN IDIOT" but if by chance you elect to engage your brain instead of your innate sense of disdain when someone says something you aren't happy with, I'll be happy to address any real points you might have buried inside your preconceptions.
Otherwise, have a nice day, and enjoy your delusions.
I've fallen off your lawn, and I can't get up.
The fairy godmother is unreliable and inherently unpredictable.
/. genius, quit bragging about your brain and invent something usefull to put in the models.
From a civil engineering point of view you can only make rational predictions using non-existant infrastructure. Making predictions with non-existant technology is called science fiction, it's more comforting and definitely more readable, but it's still fiction.
BTW: The point you are trying to make is philosophical one, it is neither scientific or pragamtic in any meaningfull way. Science is based on the assumption (faith) the Universe is predictable. Pragmatisim says that "the followers of perfection are the enemies of good". The philosphy you describe is blind optimisim, but hey, if you really are a misunderstood
And did you exchange a walk on part in the war for a lead role in a cage? - Pink Floyd.
Yes, that's exactly my point. The predictions made by these long-term climate simulations are almost certainly nonsense (barring luck) because they are presuming a particular infrastructure, and that presumption is, just as you say, science fiction.
Your outlook, the presumption that since you don't know something (the actual state of technology during the simulation), you can ignore it and use what you presume to be the case... that is utter nonsense.
Even small changes affect complex systems. Weather prediction has shown us that very, very clearly. We can't even predict temperature even 24 hours in advance without a significant margin of error. Climate is a far more complex system.
If you leave major effects out of any long term simulation of climate, you might as well be having a beer instead for all the good your simulation will do for you.
I'll leave you with that thought.
I've fallen off your lawn, and I can't get up.
Pardon me, my mistake. I thought, since you were attacking my knowledge of simulation with such gusto, that you actually knew how simulations worked. Now that you have shown you don't, I won't worry about your opinion any longer.
You have a good day!
I've fallen off your lawn, and I can't get up.
I truly admire the way you took the point I was making and using critical thinking, scientific method, and relevant examples, just tore my position apart. No question, a beautiful job. You are such a stud. The "simulate with trash data then treat the results as gospel" crowd is so lucky to have you on their side, sir.
Um, I find your ideas fascinating and would like to subscribe to your newsletter.
I've fallen off your lawn, and I can't get up.