Banner
    Jan Hendrik Schön: World Class Physics Fraud Gets Last Laugh - A Whole Book About Himself
    By Hank Campbell | May 5th 2009 07:30 AM | 52 comments | Print | E-mail | Track Comments
    About Hank

    I'm the founder of Science 2.0® and co-author of "Science Left Behind".

    A wise man once said Darwin had the greatest idea anyone...

    View Hank's Profile
    Jan Hendrik Schön, if you have heard the name, will either fascinate or enrage you.   His ability to progress from ridiculous fibs to world-class deception as a 31-year-old physicist working at Bell Labs in New Jersey is certainly impressive.

    How did fellow scientists let him get away with possibly the worst case of physics research fraud known?  It deserves a whole book, and Eugenie Samuel Reich is here to help.   If you can't sit through a whole book like Plastic Fantastic (out next week), her short version is in Physics World.

    The world's first organic electrical laser!   The smallest ever transistor!

    It's like a late night paid advertisement, except it cost taxpayers millions of dollars - and all Jan Hendrik Schön got was the sack.

    Why did the self-correcting nature of science not bring the fraud to light sooner?   Some of it is that there are bad scientists, just like in any other occupation, and they are drawn to money and silliness, and some of it is that peer-reviewed journals were not doing their jobs.   As book author Gary Taubes, no stranger to ferreting out bad science, said in in an interview here:
    I used to joke with my friends in the physics community that if you want to cleanse your discipline of the worst scientists in it, every three or four years, you should have someone publish a bogus paper claiming to make some remarkable new discovery — infinite free energy or ESP, or something suitably cosmic like that. Then you have it published in a legitimate journal ; it shows up on the front page of the New York Times, and within two months, every bad scientist in the field will be working on it. 

    In 2000 alone, Schön published eight papers in Science and Nature, publications that claim to be the world standard for quality science, and he became known for his breakthrough of using organic dye molecules to create an electric circuit which when prompted by an electric current behaved as a transistor, leading scientists (see Taubes quote above) in a dozen labs to likewise chase some funding, wasting millions of dollars of US government research money.  He also garnered the Otto-Klung-Weberbank Prize for Physics in 2001, the Braunschweig Prize in 2001 and the Outstanding Young Investigator Award of the Materials Research Society in 2002.

    Jan Hendrik Schön

    Jan Hendrik Schön.  Credit: Bell Labs

    In 2001 his name appeared on a paper an average of every 8 days.   That's terrific production, though less so if you just make stuff up.    Not everyone bought his schtick, of course.  Professor Lydia Sohn of Princeton University noted in his work on single-molecule semiconductors that two experiments carried out at much different temperatures somehow had identical noise but when Nature editors pointed this out to Schön, he said he accidentally submitted the same graph twice.    Oh, okay then, good enough for Nature but not good enough for quality scientists and his paper describing the construction of molecular transistors sent his career crashing - two fellow physicists, Julia Hsu and Lynn Loo, attempted to patent research showing that soft lithography could be used to make softer and gentler contact with organic molecules and they used Schön's paper as validation of how cutting-edge their experimental progress was; that's when they stumbled across duplicated data and sounded the alarm. 

    In this case, the private sector and individual investigators came though for integrity where big media journals did not.   In May 2002, Bell Labs/Lucent began an investigation but, surprise, Schön said he kept no laboratory notebooks and his raw-data files had been deleted from his computer because his hard drive just wasn't big enough.   Bell Labs fired him,  Science withdrew eight papers written by him and the University of Konstanz later revoked his PhD (can they do that?   The fraud was after the PhD).  Physical Review Journals also withdrew his papers in 2002 as well and finally in 2003 Nature withdrew seven peer-reviewed papers he had written as well.

    As Reich writes, "Science was corrected in the Schön case, but not by itself – only because individual scientists made corrections. From would-be replicators in dozens of labs to many sceptics, only a couple of researchers were transformed into whistle-blowers by the unlikely pattern of [duplicated] evidence."

    Reich continues, "Fraud was able to stifle questions about Schön's lab technique that would otherwise have been asked, and to turn review processes at journals into opportunities for additional fabrication. Other scientists' support of the fraud was unwitting, but the decision to place so much trust in a colleague was a conscious rationalisation that continues to be defended in science to this day."

    Article: Eugenie Samuel Reich, 'The rise and fall of a physics fraudster', Physics World, May 1, 2009

    Comments

    Hfarmer
    Schon told the world what it wanted to hear.  So he was believed. He was much like one of the tailors in "The Emperors New Clothes".    "Anyone who wasn't able to see these clothes must be incompetent or a fool", as the story went.  This tailor has told us that we can make transistors small without limit (more or less).  Moore's law will stand!  
    It's really a modern spin on a really old story.  The moral.  Don't belive people just because they have authority. 

    As for the revocation of his PhD.  I can't believe that's legal.  If his thesis was a fraud wouldn't we all know that by now? 




    Science advances as much by mistakes as by plans.
    twicker
    Re: PhD revocation - could easily be legal:
    1.  There may be a university policy that, if you use your degree for fraudulent ends and, thus, bring disrepute on the university and other students, your degree will be revoked (I'd be surprised if this clause wasn't in all university policies).
    2. It's a German university; being unfamiliar with the laws there, but noting that a major university took this decision, I would not be surprised if it was actually quite legal 
    yrs,
    twicker


    Hank
    It just seems awfully arbitrary.   The Enron guys committed some fraud but their universities did not decide they did not get a college degree in the past as well.   I guess they can do whatever they want in the third world but that would never fly in America.
    twicker
    Re: arbitrary: it depends. If either Germany or the university have a public policy that you can be stripped of a doctorate for unprofessional or unethical behavior, then it would not be arbitrary; it would be standard of practice. I would see it as akin to disbarment for gross professional misconduct, given that scientists can't be "disbarred" since there's no professional examination they have to take. Revocation of your degree for blatant fraud would seem to be a reasonably-effective deterrent, and something that would serve to protect all other graduates of the university.
    I guess they can do whatever they want in the third world but that would never fly in America.
    Wow! I'm thinking you rather hate Germany, huh? Not sure I've ever heard the Bundesrepublik Deutschland referred to as "third world" ...

    Hank
    ha ha ... no, I do not hate Germany.    Schon was effectively disbarred, removed from his position and had his livelihood damaged, the same way a lawyer would.   What Harvard Law School cannot do is remove the law degree if someone is unethical later.  They cannot because in the US the student paid for a service and the school provided it - his degree is proof of completion.     The fact that a university in Germany might do something so silly as standard practice does not make it less arbitrary, unless they let German universities remove degrees for everyone who does something wrong later in their lives - that is instead saying you did not get a degree, rather your tuition purchased a license to say you have a degree from a university and they can remove that license at any time.   I am betting if that is so everyone would call Germany not just the third world, but the Dark Ages, if they are really that heavy handed.
    twicker
    Re: effectively disbarred - not necessarily; he's obviously a huckster, and, if he can fool AT&T/Lucent, he can probably fool the next guy, too.
    Re: "does something wrong later in their lives" - ah, would that he had only, say, gotten a speeding ticket. He was fired with cause; not sure that qualifies as the same as merely "doing something wrong." :)

    Re: "paid for a service" - aye, there's the rub. I don't know what his program was like, but I can tell you that, in my PhD program, we don't "pay for a service." Quite the contrary: being in a PhD program, we are paid a stipend to be at our school, and then have our tuition paid for by the school. So, there's no "fee for service" for me to later fall back on; I'm here at the pleasure of the school. This is dramatically unlike law school, or med school, or a master's program, or any of that, where I'd be the one paying them. They pay me, so I'm pretty much figuring they get to make most of the rules. Then again, I'm not planning to commit massive fraud on my future employer or on the scientific community, nor commit a felony, so I'm not real worried (even though I might just get a speeding ticket one of these days :).

    yrs,
    twicker
    Hfarmer
    That's exactly where I was coming from.  As someone who has had to borrow heavily to pay for education I would be pissed to no end if one school or the other decided that I didn't deserve my degree anymore due to some subsequent action.  Either he paid for it, or a scholarship did.  He certainly did the work.*  I really don't see how this "revocation" can be anything more than symbolic. 
    As it stands now he will never work as a research scientist again.  I heard that he now has a job with a German engineering firm.  Let's hope he doesn't try that same foolery when designing a bridge.  As a matter of fact if I knew he designed a bridge I would pray every time I crossed it. 

     * If he also faked his thesis or cheated in some way that's a different story. 

    Science advances as much by mistakes as by plans.
    Schön has just won a court case in Germany, arguing that Konstanz University had no right to withdraw his PhD.

    Hank
    I agree with that decision.   As I said above, at least in the US if a doctor or lawyer is convicted of malpractice, they lose their license to practice, no one can retroactively remove their degree because it is not related.    For a university to claim they can retract his PhD when he did nothing wrong there is their being politically correct reactionaries jumping on a bandwagon.   Could they take away his high school diploma too?   His Boy Scout merit badges?   It was baffling for any university to think they can do that - I got a degree, I did not buy a license for a degree they can revoke at any time they like.
    Hfarmer
    Good. It was a bad precedent.  A degree is something you pay for the right to work for and earn.  There are very few things in life where you pay for the right to work (Various kinds of sales work I suppose).  That degree was bought and paid for and earned and his Diploma was merely a reciept. 
    Perhaps I look at it as too much like a business transaction but that's what it is on some level.  All talk about Acadamia and self improvement aside. 
    Science advances as much by mistakes as by plans.
    Amateur Astronomer
    In science of the past a new claim had to be verified independently before it was accepted as fact. In recent times a lot of scientists have kept quiet to avoid revealing that they didn’t know how to duplicate an experiment. Cost cutting and management by non scientific methods has taken out a lot of the safeguards that would have caught the frauds sooner. Down sizing has left a lot of technical companies without technical expertise. Most often I find a person who can read a catalog to me, but no one who can answer a technical question.
    I remember thinking when the Schön story first broke how gullible people are--even, evidently, the editors of the prestigous scientific journal Nature.  It's like Hontas said, it's like the story of "The Emperors New Clothes".
    As Reich writes, "Science was corrected in the Schön case, but not by itself – only because individual scientists made corrections...

    That is science correcting itself.

    Hank
    That is science correcting itself.
    Not in modern day consensus science.   When the inconsistencies were pointed out, Nature took his side because a committee had not caught the errors.
    AdamRetchless
    I'm inclined to agree with Juice. I don't think that there is any reasonable way for journals to detect fraud. I'll have to take a look at Nature's behavior.

    I think that there are two types of "correction" involved here. The first is just to make sure that the fraudulent findings don't become part of the consensus. If others are not able to reproduce and build upon a report, then it should never get cited and eventually be forgotten. In the researcher's career, this is a measure of their productivity and influence on the field, so it should influence their job prospects and ability to get grants. Publishing in a big name journal can only get you so far. I hope to look into this a little deeper and see how many citations his publications got.

    The second type of correction -- the retroactive detection of fraud -- is a bit harder. It's basically a criminal investigation, and the suspect is innocent until proven guilty. It may never be detected, and it doesn't really matter in the long run.

    Followup: A quick look at citation counts gives ambiguous results. The ScienceMagazine webpages associated with these article report very few citations (less than 5). However, the Scopus database says that many of them have over 100 citations -- but a lot of these citations were from papers published after the originals were retracted in 2002. Some of the citations are as recent an 2009.

    I also noticed that most of his papers had co-authors, so that is the primary group of people who failed to catch this fraud. I guess I'll have to read the book to find out what happened to them.
    Hank
    If what you say is true - journals are not the most accurate defense against shoddy work - then they are basically irrelevant.   There is literally no reason for researchers to pay for publication and readers to pay for subscriptions if an anonymous group of experts is not even looking at signals and noticing they are identical in different groups.

    The retroactive detection of fraud was done in this case but by researchers trying to get money who needed to be as accurate as possible lest their fraud be proactive - whereas journals can just say 'oops'.

    My guess, given human nature, is that since Schon got rock star status so quickly, reviewers were more concerned with asking him to add citations of their stuff rather than confirming his.

    Georg von Hippel
    Of course the whole Schoen saga would have been completely impossible if a more open access policy with regards to lab data had been standard; cf. e.g. the discussion in this blog post.
    Suppose the graphs that he had submitted were not duplicates and consequently showed different noise levels. Would he have been caught? He probably submitted numerous false claims and data and simply became lazy and complacent. Eventually other scientists would have not been able to replicate his results and he would have been exposed. It could have taken many years and set back the entire field of Physics. The fact that he worked for Bell Labs made it easy for other scientists and researchers to accept his claims without verification. How could he have thrown away a promising career by doing such a thing? Did he believe that he would never have been exposed?

    Hank
    I can't say this as fact (nor could anyone) but in knowing the personality type, I think he liked the attention on one hand and bigger BOOM!s were needed to keep it coming, and on some level he knew he could do it, he just couldn't do it yet.  So he may have felt that he was buying time and getting credit now.

    Would James Watson have been able to get a Nobel prize for DNA in today's publishing world?  Nope.
    Btw it turns out I'm not too many degrees of separation away from the guy. Don't worry about him, he seems to be making quite a living in finance right now. Makes sense.

    I received my Ph.D. in physics at an American university, and I vividly recall reading the rules in the student handbook that stipulated the degree could be retroactively revoked for up to five years at the discretion of the university (I still have nightmares about this 15 years later!). To my knowledge this is quasi-standard fine print, but only for the Ph.D., not other degrees. Although exceedingly rare, I am aware of one other casewhere this happened many years ago for blatant dry-labbing of data in the dissertation. There too, it was caught when others tried to replicate the data and could not.

    I'm reminded of an article I ready over 15 years ago (Tau Beta Pi magazine I think) that described how one of the material constants that had been published for nearly 100 years was found to be wrong. The constant was close, but off enough that as a project was developed pushing the limits of the material, failures of the material started occurring. After sending calculations to several other groups for verification, an enterprising firm decided to dig into the history of where the constant was calculated. Turned out that it was calculated by a graduate student (assigned to him by his professor) who probably had no idea how to do it, came up with something that made sense, got published by the professor (who didn't know how to figure it out either), and was accepted in the peer-reviewed journal by other professors that didn't know how to figure it out. Because it was published, it was accepted as "fact" for nearly a century.

    Hank
    I am not sure what this means.   Schon invented something that could have been possible?   Well, sure, but so did HG Wells.

    Why did his fraud remind you of an error made a hundred years ago?   Materials can't be 'constant', anyone in the $230 billion semiconductor industry knows when they buy FR4 it will be off 10%, though in their physics analysis software it is a constant 4.
    e.g.: PhD revocation - could easily be legal:

    Please don't make that law, other researchers with strange new theories, that turn out to be wrong, get there whole career destroyed. As a scientist one thing you have to be able to do, is posit a new theory, and later try to find out if its true. The idea is either truth or not, proving the that the researcher lied, deliberate supported an idea he new to be false, for is very difficult here, and the fact is most of us as human, cling on to ideas that they should know to be wrong, due to some psychology need. I passed a PhD, and right semi crazy theories in my spare time, blog about, don't have the time, money or representation to get them through peer review, one thing, but anyone could accuse me of producing deliberate lies. And If my writing turn out to false or error prone, that doesn't take away that i passed the PhD in the first place.

    "get *their (not there) whole career destroyed." Also do you really have a degree? Because your comma placement is... atrocious. And you really failed to formulate your sentences and thoughts properly...

    There are non-native speakers in this thread. So that might be the case here. It might also just be the internet. When posting in blog comments, people often don't care about there (sic!) writing. Also, even if it is an educated native speaker, I can attest as a teacher of grad school in the U.S. in the humanities that basic standards of writing are very iffy even among graduate students. I have people getting MA degrees in my classes who write as if they don't read very much or very often - broken grammar, randomized punctuation, very awkward word choice, etc. I have seen much, much worse than what is being posted here.

    Was there any validity in some of Schons previous work that gained him notoriety and the luxury of the benefit of the doubt with in the scientific community? I mean I remember hearing about this guy being put up on a good like standard for a period of time. It's not comforting to think that the brightest minds of our generation could be fooled so easily. I ,not being a scientist, have heard from people that it is extremely difficult to get published when even producing a valid exhibit of work. A few years ago I heard about a scientist that claimed he discovered a way to produce cold fusion. He was immediately placed under a microscope and when his results could not be replicated he was discredited.
    Now that Schon has done something similar to the results of the cold fusion experiments of the 80's I'm sure any claims about being able to use microbes for transistors will be placed under intense scrutiny.
    Did his actions set the research of silicon transistors back and if so is there currently any realistic solution being researched for moor's law?

    Hank
    His previous work was valid but suddenly he became ridiculously prolific and was going Usain Bolt on the rest of the world - but the things he claimed to do were not 'impossible unless we invent magic physics' stuff, like cold fusion and they didn't impact modern semiconductors because that is a size issue - approaching 25 nM we have a real train wreck coming and only doing an end-run around physics (on the quantum scale) will solve it.

    It may be that computers as we know them today reach a commodity-style dead end.   There are no 'new' transformer designs in about 70 years, for example, because transformers are terrifically efficient, and washing machines may get incrementally different features but you only buy one when your old one breaks because this year's model will not be better.

    There may instead be a computer on a completely different branch of the tree that is only related to today's PCs the way ours are related to transformers.
    There already is solid replicated empirical proof of small scale ESP in peer-reviewed literature.
    http://www.youtube.com/watch?v=qw_O9Qiwqew

    The University of Konstanz is trying to have the decision appealed. The regulations on the doctoral degree at Konstanz, where Schoen got his degree, are indeed nebulous. It states that the doctoral title can later be revoked, "in accordance with the law". It does not clarify that futher. The formulation is similar at my university (Tuebingen).

    I found the relevant law online. It reads:

    An academic degree from a German state university can be revoked...

    a) if it is later determined that it was earned by deception or if important prerequisites for the granting of the degree were erroneously assumed to have been fulfilled (wenn sich nachträglich herausstellt, daß er durch Täuschung erworben worden ist, oder wenn wesentliche Voraussetzungen für die Verleihung irrigerweise als gegeben angenommen worden sind,)

    b) if it is later determined that the degree holder was not worthy of having been granted an academic title (wenn sich nachträglich herausstellt, daß der Inhaber der Verleihung eines akademischen Grades unwürdig war,)

    c) if the degree holder demonstrates by later behavior that he is unworthy of holding an academic title (wenn sich der Inhaber durch sein späteres Verhalten der Führung eines akademischen Grades unwürdig erwiesen hat.)

    Both b) and c) are obviously controversial and open to broad interpretation. But I think c) could reasonably be said to apply here. Schoen's later exploits directly relate to his academic life and effect the whole scientific community.

    Hank
    Both b) and c) are obviously controversial and open to broad interpretation. But I think c) could reasonably be said to apply here. Schoen's later exploits directly relate to his academic life and effect the whole scientific community.
    If schools in Germany are free I suppose that's fine.   However, if a student pays $100K for a degree and then later gets caught dog fighting or whatever else makes him 'unworthy' by a fluctuating cultural standard, he should get his money refunded.    Schon didn't do anything fraudulent to earn his degree so the university was being fashionable and piling on him.   This is why they lost the court case.
    I agree with you supposition that a university has no right to take a degree away from someone who had done nothing fradulent to achieve it. The degree is proof of successful completion of the education/training.

    However, it could be argued that a Phd has no regulating authority. As a lawyer, the law school confered the degree proving that I have been trained as a lawyer but the "State Bar" is the authority that regulates my practice. If I screw up they take away my licence to practice but the law degree remains (the irony being that the degree is useless without the ability to pracitce and make a living).

    With a Phd I can use that anywhere. If I comit fraud or do shoddy work...I could take my Phd and go somewhere else. Arguably, it would rely depend on "word of mouth" - reputation. However, if nobody cares (or my next employer doesn't care), I guess they are free to hire whomever they want.

    Lastly, as to the univeristy "piling on" - sure, there might be some of that. However, I think you miss the bigger point. Which is that since science has no "governing authority" that regulates practice - so then it's up to the "business" to regulate itself. That means the "scientific industry" has to police itself by ruining reputations. Which is what happened.

    However, I think the german university wasn't so concerned about the ex post facto fraud but rather the imparting of the reputational taint from the bad scientist onto them.

    They are afraid that people will ask... "where did he learn to practice such bad science?" and if the answer is "university x" - enrollment will drop at "univeristy x" as people are more reluctant to hire students from that "cow college" and thereby hurting the school's ability to MAKE MONEY.

    Hence, what better way to get ahead of the fallout and criticism by takinging that jerk's degree away and claiming..."he's no son of ours." We only produce "quailty" scientists here!

    Simply put, they could give two (2) shi-ites about Schon but they DO, however, give two (2) shi-ites about themselves.

    There are so many papers published in peer review journals where results cannot be reproduced for one or another reason. For example, in the PDB database alone there are hundreds of structures coming with no raw data where it is pretty obvious that many of these involve components obtained by "manual placement" instead of by actual electronic densities. Examples many, the most massive case got caught, and what? No change.

    The more and more competitive race for (quality) papers can only destroy the ultimate purpose of science. If this case did not significantly change anything, I cannot see why we should believe any results presented in the published papers. It is a 50%-50% die throw.

    @Dimi, that is because not the quality of the papers matter, but their number. It happens
    everywhere in science, even in mathematics. To me there is a clear analogy with the academy
    painters from the end of 19th century, just before the explosion of creativity of the Salon des Refuses,
    see "Boring mathematics, artistes pompiers and impressionists"
    http://arxiv.org/abs/1011.3465

    "@Dimi, that is because not the quality of the papers matter, but their number."

    I kind of disagree. Unless I do not understand what you refer to by "that is".
    At any event, if quantity matters more than quality in science, then science is meaningless.

    I know where he is!

    I think the most disturbing thing is how many of the comments here are concerned with the legality of the degree revokation, rather than with the real problems, i.e. the massive fraud in scientific publications, especially in leading journals such as Nature and Science. It is not just fabricated results, it is incomplete explanations, ambiguous language, etc., etc. A physics paper in the sixties was written so that other researchers can understand and continue the work. Nowadays physics papers are written to make the claim on something and confuse the potential competition as much as possible. With science becoming a business, that seems, regrettably, inevitable.

    Science inevitably involves a large degree of trust. I had my accounts repeatedly subjected to random audits by outside accounting firms hired by granting agencies. Nobody ever audits lab notebooks or raw data (unless the data were specifically prepared for regulatory agencies and gathered under Good Laboratory Practices guidelines). Furthermore, there is an understandable bias towards doing novel research. It is difficult to get funding for a study that is a mere repetition of previously published work. It is also difficult (although not impossible) to publish its results.

    Having said that, the system seems to be working fairly well. Scientific fraud is uncommon relative to, say, accounting fraud.

    Bell Labs/Lucent only started an investigation after pretty much being forced to by a plethora of researchers at other institutions. And they've never done anything to the AT&T folks who Schön worked with/under. I'd not pat them on the back too hardily.

    The irony is that Schön may well have been on to something with his carbon-based technology. Science is often based on pushing boundaries. Schön's blunder was to make exaggerated claims based on insufficient results and an assumption that later work will somehow "fill in the blanks". The idea behind Carbon technology may be sound and should not be dismissed as a total fraud, just because the data was in this instance fraudulent. It is just that we have yet to find a way to actually make it work. Moore's Law is afterall based on an assumption too, i.e. that there is no alternative to silicon. I am convinced that there is. We just have to find it.

    Hank
    Maybe. But if I am a researcher and I write a study claiming an FTL engine and get thrown out for it, 50 years from now if someone does it I am not vindicated.  He got away with it at the time using the same reasoning you just used; it had an air of truthiness and he may be on to something in the future so peer reviewers at journals let it slide. Luckily, people outside billion dollar media called it out.
    Halliday

    Anonymous:

    A correction:  Moore's "Law" is not "afterall based on an assumption too, i.e. that there is no alternative to silicon."  It is not even truly a "law", as in a "law of Nature" sense.  It was an observation, of a technological trend.  One that, admittedly, has held up quite well for quite some time.

    While the observation was made from primarily silicon based technology, it was most certainly not based upon any "assumption" "that there is no alternative to silicon".  Even in Moore's time, there were "alternatives" to silicon (though most were relegated to other special purposes, primarily due to their relative expense), and even such "alternatives" were, and have been, experiencing rather similar "laws".

    No, whether one "assumes" "that there is no alternative to silicon", or similar semiconductor technology, or not, this has no bearing on Moore's Law, but only on the question of the continuation of this "law".  At some point, even if one goes to "carbon-based technology", or even organic molecular based technology, "Moore's Law" cannot continue, without significant modification, in perpetuity.

    David

    Hank
     At some point, even if one goes to "carbon-based technology", or even organic molecular based technology, "Moore's Law" cannot continue, without significant modification, in perpetuity.
    It's been known for a decade, almost to the year, when Moore's Law would end in current form, because we know where electrons stop moving.  At the 25nm process either we have something new - and the replacement is not close - or computers basically become home appliances, like a washing machine, and people buy new ones when they break and that is based on marketing, but no real improvements are made.

    No one is going to want the heat and size that would ramp up again to have more computing power.  We started with building-sized computers in the hands of a few and ironically could end with them as well.
    Halliday
    Hank:

    Unfortunately, even now, much of the lab level work, as far as I've been able to see, on "gate" technology is still focused upon the semi-classical solid state regime.  It is quantum mechanics that sets the limits of which you speak, yet that same quantum mechanics holds out the potential for alternate "gate" technology, even without going to quantum computing.  The very same phenomena that cause "electrons [to] stop moving" are the very same phenomena that can provide faster "gates" with zero leakage (as "classical" "gates"), once understood and properly applied.

    This isn't even what I was thinking of when I held out the possibility of a "significantly modified" form of Moore's Law.  This is still for classical computing.

    I don't know, maybe we will be pleasantly surprised with a new form of solid state "gate" technology making its way out of the lab and into microelectronics within a decade or two.  Unfortunately, it looks more like we will be faced with some orthogonal meanderings between now and then.  Maybe graphene, maybe diamond, maybe ...

    David
    Hank
    The lag shows a fundamental weakness in how the bulk of basic research is done today.  If today's academic climate needed to create a semiconductor, we would be told it takes 50 years and costs a trillion dollars.  The private sector will do it instead but there is no rush yet, it is currently a $350 billion industry with old tech.  In my previous life we chuckled at the notion that universities were doing basic research - instead, they were doing papers re-solving EM field solver problems we had solved years earlier. We did hire new PhDs, but not post-docs. The mindset is all wrong for academia to help with any pressing issue - this is also the problem with energy, etc. The solutions are truly academic and inherently unworkable coming out of academia but in things like big physics, astronomy, etc., it works fine.
    Gerhard Adam
    "Moore's Law" cannot continue, without significant modification...
    Actually Moore's Law can't continue.  Certainly some people will argue for all manner of variations [usually involving parallelism], but it should be intuitively obvious that such techniques will always have fundamental limits that are dependent on how much a problem can be broken down, and the overhead of coordinating activities between parallelized components.

    http://www.science20.com/gerhard_adam/speed_computing

    http://www.science20.com/gerhard_adam/blog/computing_speed_limit
    Mundus vult decipi
    Halliday
    Gerhard:

    Since Moore's Law is focused entirely upon the lowest levels of circuit integration, it has little if anything to do with parallelism, or other macroscopic "speed" issues.

    Since all matter we are able to handle is made of discrete units, Moore's Law cannot continue in its present form (size/density of "gates", and "speed" of individual "gates").  However, I can conceive of a modified form of Moore's Law that could, possibly, be applicable to quantum computing that may be somewhat immune to this, for some time to come.  That is why I heald out the possibility that it may continue, for a while, in some rather modified form.

    The problem, that is related, though even further off than what Hank was alluding to, is that we seem to be too far from the needed "transition", right now.

    David
    Gerhard Adam
    Since Moore's Law is focused entirely upon the lowest levels of circuit integration, it has little if anything to do with parallelism, or other macroscopic "speed" issues.
    I understand.  My only point was that people that want to disagree invariably bring up parallelism and its various permutations as a counter-argument.  The links I provided indicate the specific limits at the quantum level which represents the "speed limit".  It doesn't particularly matter how many orders of magnitude might still be achievable, the point is that there is a finite limit, despite many assertions about unlimited growth.
    Mundus vult decipi
    What Moore's law fails to account for is the supernatural power of God. Scientists at BYU have recently shown that through prayer and the intercession of the Holy Ghost, properly equipped and spiritually armed researchers can actually fit several dozen transisters onto the surface of a single atomic nucleus.

    The discovery was made when a code was found hidden within the text of the Book of Mormon. There is no such code in the English translation. But when the original gold plates were discovered in 2008 (and kept hidden from the public), the middle Egyptian hieroglyphs were converted into binary code and arranged in large grids, clear instructions for numerous technological and scientific discoveries were immediately evident - unlike the Bible Code scam of a few years ago, where single words could be dug up if enough searches were done.

    Mormon science will soon make all your work obsolete. Just remember when the news breaks in a few weeks - you heard it here first.

    Mormons understand that scientists working with traditional methods will feel threatened by these developments. They are prepared to operate through normal scientific channels to publish their research and move up through the academic ranks and gradually permeate science and engineering with their ideas. But just in case any dishonest counter-action is being considered, you should be aware that in the mountains of western Utah, the new technology offered by the Book of Mormon is being used to build a huge tachyon-energy cannon that will be able to defend Mormon interests in this matter. Consdier that before taking any rash action.

    Hank
    he new technology offered by the Book of Mormon is being used to build a huge tachyon-energy cannon that will be able to defend Mormon interests in this matter.
    Good one. You had us going for a second. 'Tachyon' is the new Prius - any time you hear it invoked, you know someone is having a laugh. At least you didn't say it was wormhole powered.
    I've never been a great believer in Moore's Law, mainly because there is more (pardon the pun) going on in the world than just silicon chips. Unlimited growth is of course a myth (well, unlimited anything is a myth for that matter). But to say that the world will stagnate and be plunged into economic depression just because silicon chip technology has reached its zenith is about as dumb as suggesting that people wil stop gambling on horses just because we can't breed new ones that run faster.

    properly equipped and spiritually armed researchers can actually fit several dozen transisters onto the surface of a single atomic nucleus.  
    The transisters aren't the problem, it's the resisters and the capaciters and the inducters and the connecters.