Is Distributed Computing Being Distributed Badly?

Carl Bialik from WSJ writes "Distributed computing could help researchers studying climate change or Alzheimer's, but SETI@home's search for extra-terrestrial intelligence continues to dominate. Wall Street Journal columnist Lee Gomes says that's a big waste, especially because SETI doesn't seem likely to yield results: 'This continued fascination with living-room SETI comes as professional setiologists concede that early assumptions about the search for intelligent life -- notably those popularized by astronomer Carl Sagan -- have proven naively optimistic. For instance, it's now conceded there is little chance of detecting the "leaking" transmissions of another planet -- its version of "I Love Lucy" broadcasts. Those signals are too weak to stand out from the universe's background noise.' Gomes also traces the origins of SETI@home to Berkeley computer scientist David P. Anderson, and explains that users stuck with the ET search rather than medical investigations in part because of nationalistic competition. Yet Anderson no longer runs SETI@home. 'Instead, he donates his spare computer power to a global warming project. But he doesn't presume to tell others what they ought to be doing with their CPU cycles.'"

Crunching for their profit

of course the WSJ would much rather you where crunching numbers for their drugs companies under the guise of "fighting cancer" or "protein folding" so your results can be turned into their profit (you didnt think that cure/treatment would be free like your CPU did you?)
searching for ET is not profitable so it must be bad

Re:Crunching for their profit

I'd rather contribute my cycles to a treatment that I have to pay for rather than no treatment at all.

Re:Crunching for their profit

I think his point was that since pharmas make billions of dollars in pure profit, they can afford to invest some of it in highpowered computing clusters.

Nobody is going to do the same for SETI.

I bet the pharmas could even write it off on their taxes.

Crunching for their lives

"I think his point was that since pharmas make billions of dollars in pure profit, they can afford to invest some of it in highpowered computing clusters."

I currently work for a pharmaceutical company, and in a visit to a research lab I learned just how much computing power they throw at these problems. They do have supercomputers, intranet clusters, etc. to try to solve these problems. They are so incredibly complex, however, that those are not enough.

Re:Crunching for their lives

I currently work for a pharmaceutical company, and in a visit to a research lab I learned just how much computing power they throw at these problems. They do have supercomputers, intranet clusters, etc. to try to solve these problems. They are so incredibly complex, however, that those are not enough.

Well, maybe if the pharmaceutical companies threw a little more of their obscene profits at the problem, it would be enough. The Board of Directors and the stockholders might have to cut back on Cuban cigars, though...

Re:Crunching for their lives

2 reasons.

1) A lot of the basic research is paid for with my tax dollars and tax breaks for researching. It's therefore ridiculous to be paying the most for these drugs (compared to say Canada). I also don't think that the money needs to be spent on advertising. If a drug is good and worth while, then tell the doctors and they will use it.

2) I'm one of those crazies who thinks that this kind of research should be done in spite of the costs. Because of this, I think that the majority (80-90%) of the profits should be put back into R&D. I don't buy into the argument that the only reason we have all this research is because people can get rich from it. I think if you gave most researchers a decent salary they would be more than happy to continue researching.

I wonder if we need to break medical research into 2 categories.... Life-saving medicine and cosmetic medicine. Let's find a good way to provide enough resources to get the 1st group done, and let the pharms do "whatever they want (tm)" with the 2nd group.

Re:Crunching for their profit

If these distributed projects were in fact run by the big pharma companies, I would agree with you. But as I have already stated in three earlier posts for this topic, they are in fact not, they are run by universities or non-profit organizations, they make the results publicly available and can research things that the big pharmas don't see much profit in. For instance - medicines for the HIV subtypes currently spreading in poorer nations.

Re:Crunching for their profit

I hope everyone see that they are not simply donating spare cycles that cost you nothing.

Modern computers enter a powersaving mode in the times their CPU is not busy. Enabling Folding@Home or SETI@HOME on your machine consumes these powersaving cycles and draws more power.

Leave these programs running for a month and check out the huge difference in your power bill.

Ironically, the distributed system to calculate climate change could actually contribute to it!!

Re:Crunching for their profit

of course the WSJ would much rather you where crunching numbers for their drugs companies under the guise of "fighting cancer" or "protein folding" so your results can be turned into their profit (you didnt think that cure/treatment would be free like your CPU did you?)

From the Folding@Home FAQ [stanford.edu]:

"Folding@home is run by an academic institution (specifically the Pande Group, at Stanford University's Chemistry Department), which is a nonprofit institution dedicated to science research and education. We will not sell the data or make any money off of it.

Moreover, we will make the data available for others to use. In particular, the results from Folding@home will be made available on several levels. Most importantly, analysis of the simulations will be submitted to scientific journals for publication, and these journal articles will be posted on the web page after publication. Next, after publication of these scientific articles which analyze the data, the raw data of the folding runs will be available for everyone, including other researchers, here on this web site."

Re:Crunching for their profit

There's a logical fallacy in there somewhere. Just because it's coming from the WSJ doesn't mean the point isn't valid. And I would have to agree with them, because I think there are better ways to spend extra computer cycles than searching for possible signals from outer space. I'd rather see extra cycles go towards things that have a larger impact for people on Earth: weather analysis, drug creation, protein folding, etc. But that's just me.

Re:Crunching for their profit

Odd, I can think of few things that would change life on earth more than a verifiable intelligent signal from outer space.

This story reminds me to go download SETI@home again.

The assumptions of SETI

There are so many problems with the assumptions behind SETI, it's hard to even know where to begin. But one big assumption (aside from assuming radically different alien life uses the same radio spectrum to communicate that we do, that they coexist with us in essentially the same timeframe, etc.) is the assumption that we would even RECOGNIZE an alien signal if we saw it.

Sure, we look for patterns. But a radically different intelligence might communicate in a way that seems random to us. Hell, they might have discovered or evolved whole mathematical systems that would seem chaotic or meaningless to even our brightest minds.

-Eric

Re:Crunching for their profit

As one of those pharmaceutical researchers, I assure you that it does not make one particle of difference to us (let alone the Wall Street Journal) whether or not some overclocker runs some "fighting cancer" thingy on his computer. If there were a computational problem that we cared about, we'd throw a cluster at it, not wait for a bunch of squabbling AMD and Intel fanboys to solve it.

And as for global warming, I'm no climatologist but I've got to think that turning your damn computer off is more valuable than anything you could run on it.

And yet, other researchers disagree

As you can see by their actions, rather than their words... Notably at Stanford University, Washington University, Munich University, Scripps Research Institute, Oxford University etc.

http://folding.stanford.edu/about.html [stanford.edu]
http://boinc.bakerlab.org/rosetta/rah_about.php [bakerlab.org]
http://boinc.bio.wzw.tum.de/boincsimap/project.php [wzw.tum.de]
http://predictor.scripps.edu/about_team.php [scripps.edu]
http://www.grid.org/projects/cancer/index.htm [grid.org]

So... Who are you again? Yeah, you're a guy reading Slashdot... Getting much research done?

 

Re:And yet, other researchers disagree

Note that Stanford University, Washington University and Munich University are universities, not pharmaceutical companies. (Stop me if I'm going too fast for you!) From the point of view of an academic research lab, investing some essentially free grad student programming time to possibly generate a paper might make sense. For the pharmaceutical companies the OP is spinning his paranoid fantasies about, it makes zero sense.

Honestly, it always amazes me how some people are willing to spend so much time cutting and pasting and href'ing, and none on reading.

coming from a cancer survivor

However they do it, whatever it takes. Drug companies weren't formed for altruistic reasons.

Re:coming from a cancer survivor

Drug companies weren't formed for altruistic reasons.

I suspect that for the older drug companies, altruism was a major factor in their founding. Those people who cry that our society is becoming less civilized are right, but they neglect to mention that "the fish rots from the head." Compare these quotes by Andrew Carnegie to today's "Death Tax" opponents: "Surplus wealth is a sacred trust which its possessor is bound to administer in his lifetime for the good of the community." and "The man who dies rich dies disgraced."

Individuals like Warren Buffet, George Soros, and Bill Gates (yes, boo hiss, his software sucks, but he understands that a charitable foundation is a better use for ridiculous amounts of wealth than creating the next Paris Hilton) stand out today as exceptions to the "Greed is good." attitude that dominates.

Re:Crunching for their profit

That's not a fair criticism, for two reasons:

1) If a company starts manufacturing a product so expensive that they cannot make a profit on it, they will soon cease to exist, as will the benefitial product they hoped to give to the world. So what would you have them do? Commit organizational suicide so they can manufacture medicine to cure a few people, and lose any chance of contributing to the engineering of a better, more accessible solution for the world?

2) Companies are not soul-less collections of worker drones, however much karma it may provide to claim that they are. Most of us work for companies. Most of us are not horrible, soul-less drones. Individuals within companies make decisions, and those individuals usually do the best they can to make the right decisions based on several important angles, like:
a) what is best for the people that make up this company (see #1 above)
b) what is best for the community we serve
c) what is best for the people who invested in our success

More than one

Does Carl realize that it's possible to crunch more than one project at a time with BOINC?

Right now I'm attached SETI, Einstein, Rosetta & LHC. It works on one for a bit and then will switch to another for a bit. And so what if SETI@home will never find anything, it's a cool looking screen saver!

Re:More than one

Actually I see the problem as being two-fold. All those free computer cycles are not that free. Modern CPUs consume more electricity to do more work and someone has to pay the electricity bills. Busy CPUs need more cooling and fans that run at full throttle for a year do wear out and fail (and you risk burning some important component, even if the PC is designed to shut down when it detects overheating). That's simply because desktop PCs are desktop PCs and not workstations and the assumption is that the fans will have to run at full throttle for maybe half an hour at a time. The real costs are not easy to work out, but it might, just might be more efficient to donate the money to charity.

The other problem is deciding which project deserves most attention. I think it's well beyond me to judge whether computer time is better spent running climate change simulations or protein folding for some medical research. Hence if someone wishes to donate computer time it will be useful if all one had to do is to download a BOINC like client that will then run whatever the server sends it. Of course you'd need a reputable institution with a sensible scientific board running the server...

Re:More than one

I think you're underestimating the quality of a desktop PC. I ran SETI and climateprediction.net for about 4 years straight on a dual G4 PowerMac. Ran like a champ. 100% CPU for months straight. Never had a problem. They can take abuse.

Re:More than one

I upgraded from the old SETI@Home client to BOINC when it became available - but the BOINC client required too much effort on my part and was getting in my way.

I know you what you're gonna say, I guess could have configured it better, RTFM, yadda yadda, but that's the point really isn't?

I'm donating my CPU cycles to some altruistic cause, I don't want to have to RTFM. I just want to install and forget. For this reason I miss the old SETI client, and have, as a result, now stopped contributing.

I simply can't be bothered.

Well excuse me

for using my computer to do what I want to do with it.

Re:Well excuse me

And that is why I won't do the ones for the drug companies. My grandfather was denied a chance at surviving cancer in the 60's, but the big drug companies went to the FDA against the doctor who had a good success rate for curing colon/stomach cancer because one of the chemicals used was not FDA approved. The big drug companies are not looking for cures, they are looking for drugs to sell.

Re:Well excuse me

Personally, I think protein folding is lame because I know that the IP generated is going to be locked up for the next 70 years.

Since people posting FUD gets modded up like crazy here I guess I have to repost this:

From the Folding@home FAQ [stanford.edu]

"Folding@home is run by an academic institution (specifically the Pande Group, at Stanford University's Chemistry Department), which is a nonprofit institution dedicated to science research and education. We will not sell the data or make any money off of it.

Moreover, we will make the data available for others to use. In particular, the results from Folding@home will be made available on several levels. Most importantly, analysis of the simulations will be submitted to scientific journals for publication, and these journal articles will be posted on the web page after publication. Next, after publication of these scientific articles which analyze the data, the raw data of the folding runs will be available for everyone, including other researchers, here on this web site."

For instance, you can read the 37 papers generated so far here [stanford.edu].

Re:Well excuse me

Regardless of what they do with the data, or where they release it, most likely some company will go through all the data/findings, and figure out a way to use the information. At that point it will be patented, and locked up for years to come. All the time we put in, running our computers at 100%, will be wasted, on drugs that we can't afford.

By objections to this:

a) You don't know for sure that will/can happen.
b) If the steps taken to take this research and create a anti-cancer drug from it were obvious, it would be an outcry if the US pantent office gave the patent to a single company.
c)Even if this worst-case scenario did happen, the cycles donated would not be wasted. You would have helped advance human scientific research, and the medicines created would still be saving peoples' lives.

Global warming project?

Fighting fire with fire?

Global Warming

"Instead, he donates his spare computer power to a global warming project"
Does this attempt to determine how much global warming is being caused by donating CPU cycles.

Re:Global Warming

"Instead, he donates his spare computer power to a global warming project"
Does this attempt to determine how much global warming is being caused by donating CPU cycles.

I think that issue is answered pretty well in the FAQ [climateprediction.net]. When it comes to the real experiments being run by that particular project and their results, you can start here [climateprediction.net].

Wastes of time

You know what's a waste of time? Gardening. You spend all this time and energy just to raise a few tomatoes that could have been bought at the store for cheap.

People should stop gardening and focus their time and energy on solving global warming, but I don't presume to tell anyone what they should be doing with their time.

Re:High prices don't cause inflation.

Inflation can only go through the roof if the government print lots of extra money. Money's a commodity like anything else, supply and demand. . . . If you live in the US, you have a boatload of inflation coming your way in the next few years.

While I agree with the principle here, don't forget that with legalized banking fraud ("fractional reserve banking") they don't have to actually print more money to increase the effective amount of currency; they can simply lower the mandantory reserve ratio. Should the LBF system ever fail, the FDIC will be forced to step in and print massive amounts of paper currency to back all those accounts.

Conversely, credit contraction (higher interest rate & reserve ratio) has a deflationary effect equivalent to that of taking paper currency out of circulation. While I do believe that the long-term trend is toward inflation, indications are that we may be approaching a credit contraction phase, and thus short-term deflation. It may or may not manage to balance out the overall inflationary trend, but it's something to watch for none the less.

Global warming

Instead, he donates his spare computer power to a global warming project.

Funny....I think that all the Slashdot gaming rigs out there are contributing quite a bit to global warming, but you don't hear us bragging about it... ^_^

The REAL cause of global warming!

Instead, he donates his spare computer power to a global warming project.

Well, yeah. Running your computer at 100% CPU use is a great way to contribute to global warming. ;)

Just like donations to charities

I don't like the way that some animal charities get more money than children's charities. Obviously the people making donations disagree. The point is the donor decides, if someone is giving something away then they decide.

Re:Just like donations to charities

"If you pick up a starving dog and make him prosperous, he will not bite you. This is the principal difference between a dog and a man."
-- Mark Twain

Still true today.

Inaccurate

If we as a race are any indication, and we're all we have to go by, it's safe to assume the opposite. The more advanced we've become, the less valuable human life has become.

As others have said, Bullshit with a capital "B".

• War - While wars go on today, they are less acceptable to most. 100 years ago war was considered an integral part of diplomacy. Today it is consider a failure of diplomacy
  
• Human Rights - 100 years ago it was an alien concept. Even with the Bill of Rights in the US, and the Magna Carta in the UK, there was always the presumption that "others" (be they of a different religion, ethnicity, or nationality) had less rights than "us." A universal set of rights that applied to everyone was not a mainstream idea.
  
• Slavery As others have said, it isn't legal in too many places these days (is it anywhere), and its practice is fringe and utterly unacceptable. 300 years ago the opposite was true, and 600 years ago it was nearly ubiquitious
  
• Women's Rights Women were property a century ago, with no right to vote in most places, and no right to choose. Instead they were property of their husbands (and unable to own property of their own in many places), and their bodies became chattal of the state and church for nine months the moment they got pregnant. While there are those that seek to revert to such a state, even in right-leaning America 70% of the population opposes such a move, and in more enlightened countries the notion is even less acceptable.
  

I could go on (the acceptability of massive civilian casualties during the first two wars, up to and including the nuking of Hiroshima and Nagasaki, vs. the unacceptability of even modest collatoral damage today, etc. etc.), but you get the idea. Human life has seldom if ever been prized so highly as it is today.

For the love of God, the level of surveillance that the anglosphere tolerates is unfathomable by the standards of 1,000 years ago.

Hardly. The surveillance was done by a different entity 1000 years ago, namely the Catholic church. Its mechanism was low-tech...guilt and mentally batter your subjects into such a perpetual state of guilt and then encourage them to go the "confession" and receive absolution. Everyone reported their sins to the local priest, and often discussed their "concerns" with said priests likewise. Even kings had their confessors...which gave the church an immense level of day-to-day surveillance of an entire continent during the middle ages that is still unrivaled even today.

Even 50, 20, 10 years ago (hell, today for that matter), if you think government serveillance of your life in the big city is bad (and it is IMHO very bad, and very dangerous), it is nothing to what your family and neighbors make a point of knowing about you when you live in a small community. Talk about "Big Brother", try adding "Big Aunt", "Big Sister", "Big Cousin", "Big Mother", "Big Father, "Big Neighbor", "Big Gossip Down the Street", etc. to that.

So your arguments are false on their face, and as for reasons not to venture into space, spurious and irrelevant at best. Space brings with it problems and solutions, just as the discovery of America did, and every other migration and advance of the species has over the millennia. If and when we do meet another sentient species, that too will bring with it challenges ... and the stimulus for growth that will push our species into addressing and developing further refinements in ethics, diplomacy, and the wisdom to use military force (or not) as needed. As with any challenge, we will either rise to the occasion or fail.

However, if we cower in our little corner and forsake progress because we fear it, then failure (as in the end of the species in the nearer term) is no longer merely a possibility...it becomes a certainty, and along with it our certain extinction, the next time the planet experiences one of its many recurring major disas

Re:It is a waste...

It may be a lost cause, but it's *their* lost cause.
They support seti for the same reason that they support Soccer/Baseball teams that never have a hope of winning anything, it's more sentimental than logical.

Journalist's opinion is better (not)

This is merely an opinion piece. It's easy to take the pragmatic road and dontate personal computing cycles to cancer research or something as equally earth based - citing return of results arguments.

I postulate that the returns for finding out if there is intelligent life in outer space has greater implications for the world's population. Not immediate concerns mind you (unless something extraordinary happens), but the practical usage will eventually seep out of the acedemic and scientific circles and benefit the population in ways that we cannot possibly imagine.

The opinion the journalist writes is the simple (IMO shallow) doubts of doing science for it's own sake.

Besides, this whole opinion is practically moot. There are MORE than enough extra computing cycles out there. People can choose to which project they wish to donate too. Slow news day perhaps.

-FlynnMP3

People are using their things wrong

It's a waste that people use their cars to go see a movie when they could be delivering food to the homeless shelter.

It's a waste that people are storing ice cream in the fridge when they could be storing donated blood plasma.

grid.org - Cancer research, etc.

Personally, I always felt SETI was not very philanthropic - more like an amusing experiment in grid computing.

I have been running grid.org [grid.org] for many years. They focus on medical research. They provide great features for managing all your computers that run the grid projects. You can even choose which research to participate in. And, to satiate a geek's lust for power, they have rankings for your aggregate compute time.

It's not their fault!

I'm sure there are lots of people who would support using thier idle PCs for alzheimers research. They just forgot.

Reasons to be cheerful...

Number 1 - The Seti Experiment was not a waste. We now know that there are no signals of the kind we were hoping for in the areas we looked at. This is a finding. It is not a failure. Do not underestimate the importance of negative results in science.

Number 2 - Seti was the seed-corn for the whole concept of doing scientific computing as a distributed calculation. It was directly responsible for the development of BOINC, which is a very valuable tool for all the scientific community.

There wouldn't be other projects without SETI@home

Imagine where distributed computing would be today without the high-profile trailblazer, SETI@home. I remember reading several articles in 1999 about the project in the New York Times among other places. SETI@home's unique goal and approach attracted members, and with those, more attention for its size. Over 5.2 million have participated in the project, and in all likelihood 5 million of those were new to distributed computing.

While SETI@homes's managed to retain nearly a million members, the claim that it steals participants from other projects is absurd. Most of those other projects would face far greater obstacles to acceptance by having to woo new participants not already familiar with DC. Probably the originators of those other projects would not have even heard of DC themselves, or at least would have started several years later without a clear success story to look up to.

I gave up on SETI

... a long time ago, at the exact moment when I recognized that radio broadcast, even assuming other life forms discover it, is just a quick stepping stone toward more efficient/direct means of distribution, like wires or fiber. Or drums. Or pherimones. Or telepathy.

It's happening right now for ourselves. The entire hi-power broadcast radio phenomenon on this planet will have begun and essentially ended within about a single lifetime, maybe two. We've no data to indicate that radio would remain a prefered means of communication anywere in the universe for any race that understands technology *that* well.

SETI has always barked up the wrong tree. Not because there are no intelligent races out there -- and I really do suspect there are -- but because if they *are* intelligent in a way that we would even recognize then they've moved on to other forms of communication, or settled into a fine state of just dealing with everyday as it comes and not worring about events in their version of Iraq.

Optical SETI versus Radio

Radio SETI is really a waste of time. Optical SETI is the logical choice because;

1) Visible light-emitting devices are smaller and lighter than microwave or radio-emitting devices.
2) Visible light-emitting devices produce higher bandwidths and can consequently send information much faster.
3) Interference from natural sources of microwaves is more common than from visible sources.
4) Naturally occurring nanosecond pulses of light are mostly likely nonexistent, although there are all kinds of radio signals that could be similar to intentional SETI transmissions. Thus Optical SETI does not require grid computing to find signals.
5) Exact frequencies of light are not required, as nanosecond unfiltered light pulses would still outshine the planet's star by over 30 times.

Optical SETI detection out to 100 light-years is doable today, with a bit more work optical SETI out to 1,000 light-years is possible.

Optical SETI paper [princeton.edu]

Helping people with Alzheimer's

I used to work in computational chemistry for Glaxo.* (regexp because I can't track the company's name changes since I left). After two years there I became strongly convinced that computers do not find cures for diseases - or even give you much understanding of illnesses. Molecular modeling is so far from being able to model in vivo molecules that it's practically worthless. One time I joked that people might was yell try to design molecules using yarrow stalks and the I Ching and then a few weeks later we took delivery of a piece of software that was practically the same thing - it generated random molecules (using a ball and stick model) that fitted within a defined volume of space. This stuff is nonsense. It tells you little about how molecules will behave in a real person.

One time my manager showed me some statistics for drug discovery. Drugs need to go through various rounds of testing: it might start with assays with just receptors, move up through animal tests to full blown clinical trials. He showed me two interesting facts: firstly, the correlation between success at one stage and success at the next stage was low. This meant that the correlation between the earliest stages and the final in vivo drug activity was tiny. Secondly, the best drugs were often outliers in the sense that you could often discern some kind of pattern allowing you to predict drug activity for a class of molecule, but that the good drugs fell way outside this pattern. Because activity levels predicted from simulation are so poorly correlated with the first stage of drug trials, and we already know that trials at this stage are poorly correlated with actual drug usefulness, simulations are just as much a waste of resources as SETI.

It seems to me that molecular modeling is actually one of those hard 'macho' (but ultimately pointless) projects that gets funding because to criticize it makes you seem anti-drug, anti-therapy and ant-human-progress.

(I'm not saying people shouldn't try to model molecules. This is a great blue-sky goal. But people who are trying to find drugs or therapies shouldn't be wasting their time with such techniques.)

It's not CPU power American medicine lacks...

It's political and financial willpower to do the right thing.

If there was a way to make as much money on a one-shot cancer cure as on pills to control stomach acid, we would have it now. Antibiotics are easy to develop, the test procedures have been refined by years of experience, they've been mass-produced for a hundred years now, yet no new antibiotics have gone on the market in the last 20 years. Does anyone really think science has run out of substances that kill bacteria? No, the problem is that there's no money on cures or prevention, people take them once and then recover (or don't get sick in the first place). There's far more money to be made in selling Americans with health insurance $3 purple pills to treat heartburn or baldness or enlarged prostates or to let old farts have sex until they're ninety than in saving hundreds of millions in Africa from certain death by AIDS.

If the drug companies that stand to benefit from current medical research want donated CPU cycles, then they should start acting like they really intend to develop and market (at affordable prices) a cancer cure or a vaccine for AIDS or some other miracle cure rather than yet another heavily advertised long-term treatment to help baby boomers keep pretending they aren't getting old. If they want to keep on milking the old folks' prescription drug benefits for all they're worth, they can use some of those profits to pay for supercomputer time.

Re:Useful CPU cycles use

Bah! Anyone running on a P4 is contributing to a global warming project.