A Small Glimmer of Hope For Faster-Than-Light Neutrinos 183
sciencehabit writes "The CERN particle physics laboratory in Geneva has confirmed Wednesday's report that a loose fiber-optic cable may be behind measurements that seemed to show neutrinos outpacing the speed of light. But the lab also says another glitch could have caused the experiment to underestimate the particles' speed. The other effect concerns an oscillator that gives its readings time stamps synchronized to GPS signals. Researchers think correcting for an error in this device would actually increase the anomaly in neutrino velocity, making the particles even speedier than the earlier measurements seemed to show."
Poor Quality Assurance does not boost confidence. (Score:1, Insightful)
Sorry, but they have found two errors already. I'm not going to buy extremely outlandish claims with two known failures already throwing off the results.
Sorry, CERN, but you need to pick up the workmanship before you can be taken seriously.
Re:Poor Quality Assurance does not boost confidenc (Score:5, Insightful)
Only on Slashdot would arn armchair critic post crap about CERN's 'workmanship' late on a Friday night.
This will require time (Score:5, Insightful)
Doing measurements like this is extremely tricky, as it exceeds the usual equipment precision by a lot. I expect that confirmation either way will at least require months, possibly years. I would not be surprised if they need to recalibrate a lot of equipment and may have to build some especially for this experiment. Anyways. in the course of doing so, they will learn a lot and the improved measurement techniques developed will be available in the future. This is science at work. I do not find any fault with the researchers, just the press coverage. But the press has never understood how science works or what scientists do.
Extraordinary claims also require extraordinary proof. So the original measurement would not have been enough anyways, even if no flaws were found. I also seem to remember that they never claimed FTL neutrinos, but an effect they could not explain, leaving it open whether this was a measurement error or something not consistent with current physical theory.
Like 0.0001% faster anyway (Score:2, Insightful)
dual doom for data (Score:5, Insightful)
I think that at this point they ought to establish two different links using different technologies, for the data, in parallel, if they can. There they'll be able to say "Oh. now we're not sure which one is correct." :)
I believe Wizard Tim would say "Three links, I say three. No more and no less is the number." And something about swallows, coconuts, and neutrinos.
A good side effect of all this (Score:5, Insightful)
Regardless of the outcome, there is a good side effect of all this. All the equipment will be checked like crazy. Everything is going to be blueprinted to perfection. We might even advance the whole science of measurement. We might come up with better procedures for QA that could be transferred to other experiments. I hope influential people are taking notes and applying what they learn to other situations.
Re:Like 0.0001% faster anyway (Score:5, Insightful)
But a lot of discoveries border on the measurement error initially, otherwise the discovery would have been made earlier with even cruder instruments.
Re:Poor Quality Assurance does not boost confidenc (Score:5, Insightful)
In this case, it's win-win for science. Either we get knowledge of FTL neutrinos, or we improve our measuring techniques/instruments. Who can complain about either scenario, merely because we don't come up with the answer immediately?
Re:Like 0.0001% faster anyway (Score:4, Insightful)
A 6 sigma is not usually considered bordering on a measurement error. Additionally, this was not an insignificant speed increase. 0.001% of the speed of light is still very fast. 300000m/s give or take.
Re:Poor Quality Assurance does not boost confidenc (Score:5, Insightful)
Actually, that's just the current paradigm. It's certainly not the only paradigm.
Science begets technology which begets economic growth. If scientists wanted to make money, they wouldn't have any problems... they'd just have to start doing things differently. A lot of research is done by grants that end up being public knowledge that is published in peer-reviewed journals and it's all academia. If they wanted to privatize and get out of the academic world, it would probably be bad for society as a whole, but pretty damn good for scientists.
It's basically the idea of peace on earth, goodwill to all men, and that kind of thing. Pretty much our whole economic engine has been created by scientists. They should really be lauded as heroes for all that they do and how little they do it for.
Re:Poor Quality Assurance does not boost confidenc (Score:2, Insightful)
Re:Poor Quality Assurance does not boost confidenc (Score:5, Insightful)
Science doesn't work that way. It's not "should we believe him that there was a wolf", it's "is his account plausible as a real wolf sighting, is there any wolf traces and should we expend resources to try and confirm/disprove his claim"
Yes, they found one result "too good to be true" and now they're checking that result. If you'd RTFA (outlandish, I know) you'd notice this snippet at the end:
The two effects will get a new round of tests in May, when the two labs are scheduled to make velocity measurements with short-pulsed beams designed to give readings much more precise than scientists have achieved so far.
Re:Poor Quality Assurance does not boost confidenc (Score:1, Insightful)
No moron, that is not how science works. They released these results so that their peers can independently analyze them. Many eyeballs makes all bugs shallow is the way science has worked for hundreds of years its nothing new. The media are to blame for the circus by delving into the middle of a process that they do not understand or care to.
If you do not want to read media hype, don't. Go to the source, get some data and send them an email suggesting they had a loose fiber optic cable.
Re:Any way (Score:2, Insightful)
In relativity they say there is a speed limit for particles. We always assumed that it was the speed of light. It could in effect be slightly faster then light and causality is maintained.
What if causality is a faulty assumption. Goes against our intuitions, sure, but so do lots of other things physicists have discovered.
Re:A good side effect of all this (Score:5, Insightful)
In teaching engineering, I'm told, part of the experience is learning how engineering projects failed.
Perhaps science needs to include the same. Perhaps we should be teaching why experiments got the wrong result, or why an effect was not detected when it should have been. It could be anything from equipment malfunctions to sampling and interpretation bias.
Re:Poor Quality Assurance does not boost confidenc (Score:2, Insightful)
This is how science is done though. It's not all "Theory -> Test -> Proof!" It's often just like this... messy, details, flaws, re-testing, often in cycles.
If we could get a OUNCE of this type of thinking applied to religion, God wouldn't stand a chance.
Poor quality assurance indeed.
Re:Poor Quality Assurance does not boost confidenc (Score:4, Insightful)
"This result is probably wrong" != "fertilize your lawn with motor oil"
Re:Poor Quality Assurance does not boost confidenc (Score:5, Insightful)
Re:Poor Quality Assurance does not boost confidenc (Score:5, Insightful)
No one is debating that. The original press release said that they were checking results elsewhere to ensure that they were correct. Broken / miscalibrated equipment is the bane of every scientist, and with-holding of results until after several of your friends confirm things is always a good idea (it helps prevent publications of perpetual motion machines and what not).
Re:Poor Quality Assurance does not boost confidenc (Score:5, Insightful)
First of all, they're not making "demonstrably insane claims". They published the data they collected during experiments and are looking for explanation - which might be "experimental error". Discarding anything as "demonstrably insane" before you investigate the reasons for data you got is just the other side of accepting anything you hear as true without investigation. Sadly, latter is modus operandi for modern journalism, which is why this all got blown out of proportion.
Second, you sound like you personally invested in development of FTL engine at CERN and now found out it was a fraud.
OPERA was looking for tau neutrinos and found them, AFAIK, and FTL neutrino sighting was just a strange data point they will now try and reproduce to shut this case.
All of them... (Score:5, Insightful)
How many gratuitous errors and claimed impossibilities do you generally consider acceptable in your version of science?
All of them, at least in a provisional sense. The provision being that each is promptly acknowledged as an error as soon as it becomes known that it was an error. Science progresses through the discovery of inconsistency or inaccuracy in existing explanations, and this means there will be occasional false positive. Science is not a fixed body of dogma independent of truth (that would describe most religions).
Re:Poor Quality Assurance does not boost confidenc (Score:5, Insightful)
If you read the original paper, you'll find that they did NOT claim the "too good to be true" result. It was reported that way by a lot of press agents, but the original source basically says "we think there's a systematic error at work we haven't identified. Can you see the problem?" Do not blame CERN for the way the press misrepresented them.
I worked at CERN for six months doing my M.Sc. (in 2000, on ATLAS, not OPERA which is the experiment reporting the result.) Having seen the actual systems and the level of complexity and sensitivity involved, I think assembling them with two small errors identified this quickly is actually pretty damned impressive. Most tech companies dream of CERN's quality control.