Hulver's site || Our Mods Are Armed

Diaries
Everything
Who's Online?
FAQ / Help

Great scams of the academic community

By the (Mon Jun 21, 2004 at 04:58:56 AM EST) (all tags)

What are the great scams that have been perpetrated by academics and other 'pundits'? I'm not going to bother with non-sciences, or even the soft sciences, where perpetrating a scam is about as interesting as holding an obfuscated Perl competition. What about the 'harder' subhects like engineering, computer science or physics? The big one at the moment is nanotechnology.

We all know (don't we?) that the whole field is a scam. Sure, some people are building micro-motors and so on. But self-reproducing factories and even just nano-constructed cars are probably a century or two away.

A scam I've had personal involvement with is the use of molecular simulation in biochemistry and drug design. But maybe in 20 years time it'll no longer be a scam.

A great long running scam was the heyday of AI. "True AI" is just around the corner people kept saying. And amazingly, people like Kurzweil were still saying it recently. (Kurzweil and Drexler must rank as the two supreme Scam Meisters of the last few decades.)

Whatever scams are there that I should know about?

< Hah! I feel good. Na na na na na na na | BBC White season: 'Rivers of Blood' >

Login

Related Links

- nano-constructed cars
- the's Diary

Comment Controls

Great scams of the academic community | 47 comments (47 topical, 0 hidden) | Trackback

Fusion by DesiredUsername (3.00 / 0) #1 Mon Jun 21, 2004 at 05:06:42 AM EST

Both hot and cold. Zero Point Energy. Chaos theory.

---
Now accepting suggestions for a new sigline

I don't know if cold fusion was just... by the (3.00 / 0) #5 Mon Jun 21, 2004 at 05:21:38 AM EST

...an honest mistake. These things do happen in a healthy scientific community, even if Pons and Fleischmann did choose to reveal their results in a slightly unorthidox way. There must be a good historical book on the subject by now, I'll have to read it. I'm not sure there was a concerted long-term effort by many groups all over the world trying to make money out of research grants like there is for nanotechnology.

ZPE - I don't think anyone really believes in it do they?

Chaos Theory is a good one. And I'd lump togteher with that, the Chaos Theory of the 70s, Catastrophe Theory. But these are both good subjects in their own right - it was just the hype that was a scam.

--
The Definite Article

[ Parent ]

I mean the current cold fusion people by DesiredUsername (3.00 / 0) #6 Mon Jun 21, 2004 at 05:25:32 AM EST

but I guess they aren't mainstream, so it doesn't really count. Hot fusion though--the history is basically identical to that of AI ("in the next 10-20 years") and the results have been even more disappointing (that is, we are using failed AI more than failed fusion).

---
Now accepting suggestions for a new sigline

[ Parent ]

But hot fusion will work one day... by the (3.00 / 0) #29 Mon Jun 21, 2004 at 07:08:00 AM EST

I must already be a member of the cult.

--
The Definite Article

[ Parent ]

Oh, I agree, it probably will by DesiredUsername (3.00 / 0) #31 Mon Jun 21, 2004 at 07:18:54 AM EST

And so will real AI.

---
Now accepting suggestions for a new sigline

[ Parent ]

: I think I'd put money on hot fusion working... by the (3.00 / 0) #32 Mon Jun 21, 2004 at 07:31:32 AM EST

...a century before true AI.

--
The Definite Article
[ Parent ]

Scam by ucblockhead (3.00 / 0) #2 Mon Jun 21, 2004 at 05:08:49 AM EST

Is it a scam if the people perpetuating it truly believe it? I've met no end of people who really, truly think that real AI is a decade away.
---
[ucblockhead is] useless and subhuman

: They are the most annoying people! by the (3.00 / 0) #3 Mon Jun 21, 2004 at 05:14:46 AM EST

How can people be so out of touch with reality? The strange thing is that the most zealous believers are the ones who should know best as they work in the field. I can understand people being this mistaken in the first wave of AI research - people really had no clue how hard it was to do the stuff brains do and there was a 2000 year old tradition that said that you could just use simple syllogisms to deduce everything from axioms (even though there was a more philosophical tradition that said otherwise). I'm reminded of the apocalyptic predictions of groups like the Jehovah's Witnesses - but even they eventually learned to stop giving hard dates for the end of days.

--
The Definite Article
[ Parent ]

Psychology. by CheeseburgerBrown

(3.00 / 0) #4 Mon Jun 21, 2004 at 05:20:21 AM EST

Also: W --...W --...

No, I dare not say it.

(There are some sciences so soft even speaking their name aloud can break the bubble.)

_____
I am from a small, unknown country in the north called Ca-na-da. We are a simple, grease-loving people who enjoy le weekend de ski.

: hmm by Dr H0ffm4n (3.00 / 0) #37 Mon Jun 21, 2004 at 10:23:26 AM EST

Psychology is a big subject. A dodgy, unscientific start admittedly, but saying all of psychology is a scam is like saying physics is a scam because of its roots in metaphysics and alchemy etc...

[ Parent ]

I think you're missing the point by MillMan

(6.00 / 1) #7 Mon Jun 21, 2004 at 05:26:38 AM EST

Universities are for advanced research, and that means they'll be studying topics that won't lead to anything useful (where useful means profitable by your definition I think) for a number of years. 20 years is not a long time.

Sure there are plenty of cranks out there working on BS, but nanotech is not BS. People have been seduced by it because it has become the next big thing to save American industry, thus it might be overfunded as a whole, but it isn't a BS topic to research.

I agree with you on Kurzweil, though.

"Just as there are no atheists in foxholes, there are no libertarians in financial crises." -Krugman

I'm sure there's lots of good nano-related... by the (3.00 / 0) #9 Mon Jun 21, 2004 at 05:36:01 AM EST

...research, and there's lots of good research that is getting the word nano- prepended to it to get funding. But there is a lot of BS being spoken and that BS is helping to fund the subject. Drexler has been highly infludential and yet he speaks BS the whole time.

I'm all for blue sky research. As a mathematician I've whiled away many an hour doing research with no obvious application.

--
The Definite Article

[ Parent ]

ok. by MillMan

(3.00 / 0) #34 Mon Jun 21, 2004 at 07:57:23 AM EST

I was getting one of those "research is worthless unless it can return a profit in 6 months" kind of vibes from the diary. You've cleared it up.

My roommate is just starting his phd in neuro psychology, which is moving the psychology field ever closer to being a hard science. Thus it's an exciting time to be in the field, but as a "soft science" it is littered with cranks and qualitative studies that are nothing more than opinion pieces.

"Just as there are no atheists in foxholes, there are no libertarians in financial crises." -Krugman

[ Parent ]

: I read "The Astonsihing Hypothesis"... by the (3.00 / 0) #35 Mon Jun 21, 2004 at 09:06:32 AM EST

...many years ago. I wasn't sure what to make of it. I have a general rule of thumb: if it's written by a Nobel prize winner outside of their specialist field then it's 50/50 whether or not it's worthwhile. (This is a lower chance than I would give for non-Nobel prize winners because it's easier for a Nobel prize-winning crackpot to get into print.) I couldn't tell if Crick was a crackpot or not. He said lots of interesting things along the way but I vaguely remember him suggesting a low frequency pulse (20Hz or something) being the seat of consciousness.

However, there is another prize winner (though not Nobel) whose opinions on consciousness definitely are leading to some of the most crackpot research being funded today: Penrose. Countless 'researchers' are now publishing mad papers about quantum computing and microtubules and God knows what else.

I'm enjoying making these armchair pronouncements about people far more qualified than me... :-)

--
The Definite Article
[ Parent ]

Scam or hype? by 606 (3.00 / 0) #8 Mon Jun 21, 2004 at 05:30:57 AM EST

A lot of the scams mentioned here are not really scams, they're just overhyped fields. Nanotech has a future, but it's years and years away. But that's no reason we can't do the ground-floor research.

A friend of mine is going into quantum computer hardware research. It'll be ten years before anything useful appears (right now they can get like, what, six to fifteen qubits entangled at most). He's willing to wait, and to learn.

Overhyped pie-in-the-sky fields are good, though, because they get people interested in science, which means more funding and more scientists.

The exception is Kurzweil's last turgid piece-of-shit book which presented pie-in-the-sky science with an "it's going to happen real soon now whether we want it to or not" attitude. It's not going to happen if we don't work on it. Duh.

Personally, I was involved in Artificially Intelligent Software Agents, which is a nice overhyped topic in the software field. You can do some really cool things with software agents-- just ask MIT. However, they only work well for very specific problem domains and certainly didn't deserve the emphasis they got around 1999 to 2000.

-----
imagine dancing banana here

Quantum Computing by the (3.00 / 0) #10 Mon Jun 21, 2004 at 05:43:09 AM EST

I don't think this will come to much except for some specialized applications. But I won't write it off as a scam because I think the issues are hard. I'd be impressed if there is a 15 bit quantum computer in existence. That's getting into the realm that I claim won't be possible for a few years. Getting the qubits entangled is easy - it's getting them entangled with each other and not their environment that's hard.

Overhyped pie-in-the-sky fields are good, though, because they get people interested in science...

Yes. I won't go writing a book blowing the whistle or anything :-)

Software agents were so hyped that I couldn't actually drill down to the content to figure out what they actually were in reality.

--
The Definite Article

[ Parent ]

: Software Agents in a nutshell by 606 (3.00 / 0) #30 Mon Jun 21, 2004 at 07:16:06 AM EST

Step one, take an expert system, neural net, fuzzy logic system, or any predictive algorithm. This is your "behaviour" logic.
Step two, take a database, "blackboard", messaging system, or any encapsulation of state. This is your "environment".
Step three, use the state of the environment and the algorithm to decide on an "action" to execute, and then do it.
Hookay, are you ready for the revolutionary bit? Put all that in an infinite loop, fork it off, and let it run on its own.
Bam! New Paradigm!
Really simple idea, and useful in many situations, but not really revolutionary.

-----
imagine dancing banana here
[ Parent ]

(Comment Deleted) by yicky yacky (3.00 / 0) #11 Mon Jun 21, 2004 at 05:47:18 AM EST

This comment has been deleted by yicky yacky

How by damballah (6.00 / 1) #13 Mon Jun 21, 2004 at 05:49:37 AM EST

is java related to the academic community?

[ Parent ]

(Comment Deleted) by yicky yacky (3.00 / 0) #14 Mon Jun 21, 2004 at 05:52:35 AM EST

This comment has been deleted by yicky yacky

[ Parent ]

but by tps12 (6.00 / 1) #15 Mon Jun 21, 2004 at 05:57:08 AM EST

Surely the scamminess of Java as platform independent is a consequence of Sun's marketing rather than a flaw inherent in the notion of a virtual machine in the realm of academic computer science?

[ Parent ]

(Comment Deleted) by yicky yacky (6.00 / 1) #16 Mon Jun 21, 2004 at 05:59:51 AM EST

This comment has been deleted by yicky yacky

[ Parent ]

Is it? by DesiredUsername (3.00 / 0) #19 Mon Jun 21, 2004 at 06:08:49 AM EST

Virtual machines have some nice features, but "platform independence" isn't really one of them. You still need to standardize the virtual machines to do that and if you are going to standardize why not do it on the *real* machines?

---
Now accepting suggestions for a new sigline

[ Parent ]

Unless by DesiredUsername (6.00 / 1) #20 Mon Jun 21, 2004 at 06:10:53 AM EST

You are saying that platform-independence is only a feature promulgated by Sun, not by the CS community. In which case IAWTP.

---
Now accepting suggestions for a new sigline

[ Parent ]

platform independence by damballah (3.00 / 0) #24 Mon Jun 21, 2004 at 06:21:32 AM EST

I thought that was the point of VMs? Otherwise, why bother with it?

[ Parent ]

: Volume! by DesiredUsername (3.00 / 0) #25 Mon Jun 21, 2004 at 06:27:44 AM EST

They are good for testing things like networks on a single machine or as a sandbox for something like a virus. Plus there's VMware. Also, the term "virtual machine" is now being used by people who simply have embedded interpreters in their programs. What used to be "global variables" are now "virtual registers".

---
Now accepting suggestions for a new sigline
[ Parent ]

Oh, right by damballah (6.00 / 1) #17 Mon Jun 21, 2004 at 06:01:56 AM EST

You're right. Then, again I thought the topic was talking about concepts, and not implementations. If you're saying that java sucks as a virtula machine, are you saying that all virtula mchines suck? Sorry, It just seems to me the question asked about fields that are scams, rather than products.

[ Parent ]

: (Comment Deleted) by yicky yacky (5.00 / 1) #18 Mon Jun 21, 2004 at 06:03:32 AM EST

This comment has been deleted by yicky yacky

[ Parent ]

People by damballah (3.00 / 0) #12 Mon Jun 21, 2004 at 05:48:23 AM EST

saying that proofs using computers are just around the corner. One of the fileds that' hyped right is the bio* field. It's certainly new, but I don't know how innovative it trully is.

: Are you talking about program proofs? by the (3.00 / 0) #21 Mon Jun 21, 2004 at 06:11:26 AM EST

I remember that being pretty popular for a while. I think people have now learned that even if you can prove your program meets the spec, writing a correct spec is 99% of the problem (where 'correct', here, means something very strict in order for the proof to be possible). I think it took people a while to realize that.

But given G&oml;del's Theorem 'n' all that I'm not sure anyone expects completely automatic proving, except in very limited domains.

--
The Definite Article
[ Parent ]

Kurzweil has made a lot of cool toys by Ken Pompadour (3.00 / 0) #22 Mon Jun 21, 2004 at 06:14:14 AM EST

That actually do something useful. So if he's a scam from the academic perspective, I don't care.

: Not just academia by the (3.00 / 0) #23 Mon Jun 21, 2004 at 06:20:28 AM EST

But very conveniently Kurzweil has collected the BS for nanotech and The Singularity in one place.

--
The Definite Article
[ Parent ]

I was going to say by dr k (3.00 / 0) #26 Mon Jun 21, 2004 at 06:41:53 AM EST

modern linguistics, but I guess that falls into soft sciences. So I'll suggest theoretical physics and string theory.

:| :| :| :| :|

: Linguistics? Interesting. by the (3.00 / 0) #28 Mon Jun 21, 2004 at 07:06:48 AM EST

I'm not sure what to make of linguistics. I've read a few books on the subject. I read this book right to the end and found it fascinating. But when I tried to read the primary literature, some of it started looking like BS. I wasn't sure what to make of it. At the end of the day there was a lot of technical discussion related to trees - and as a mathematician with a strong computing background, basic properties of trees shouldn't be much of a challenge for me. But I was unable to make sense of some very basic tree-related definitions. (OK, I'm just being polite, what I really mean is that I think the definitions were self-contradictory.) I felt like maybe, just maybe, some of this formalism was actually fake. I can see two possible explanations: (1) it's BS or (2) the subject is still a bit touchy feely and a real linguist would extract useful content without reading the technical defintions as strictly as me.

String theory 'n' stuff is a different matter. While it doesn't have an application, and I (and even some of its practitioners) don't think it's the theory of physics (or even a theory of physics in the sense of describing anything like this universe), people do work within some fairly tight constraints. Researchers are just working out the consequences of some fairly basic axioms and seeing where they lead. But there was the Bogdanov affair so maybe things could be tightened up a bit.

--
The Definite Article
[ Parent ]

I know it's not what you're looking for... by lb008d

(3.00 / 0) #27 Mon Jun 21, 2004 at 07:04:48 AM EST

but the Sokal Social Text hoax is worth a few laughs.

: But don't forget... by the (3.00 / 0) #33 Mon Jun 21, 2004 at 07:33:32 AM EST

...the Bogdanov Affair. Note also, that Baez (the author of the above article) is one of the participants in the affair, so his write-up might be considered to be biased.

--
The Definite Article
[ Parent ]

I wonder ... by BlueOregon (5.50 / 2) #36 Mon Jun 21, 2004 at 10:06:32 AM EST

... whether "penis enlargement" counts as engineering. Is it a hard or soft science? Certainly, for some it is part of nanotechnology.

I drank the KoolAid by Wise Cracker (3.00 / 0) #38 Mon Jun 21, 2004 at 12:09:58 PM EST

I'm not sure where you would put the boundary of 'scam', but physics has had its share of dead-ends, belief in which persisted even after they were shown to be bogus:

phlogiston
luminiferous aether
conservation of caloric
Vulcan
19th century cosmology with its thousand and one made-up shapes of the universe, many before they even discovered other galaxies
pretty much all of Aristotle

Today we have dark matter, dark energy, and Mills' below ground-level hydrogen. And there were the pre-Godel mathematicians like Russell who insisted that all mathematics could be formed by starting with axioms and then including all the provable and rejecting all the unprovable conjectures.

Biology, we have spontaneous generation, "DDT makes birdshells thin", the overblown fear of asbestos and mercury.

In computer science, we have bunches. Remember the software way back when that combined a simple database with a screen painter and form editor being sold in an ad with the caption, "Now you can fire all your programmers!" And then there's <troll>object-oriented programming, where "you never have to rewrite code". There was a very good article in DDJ years back called "Are the emperor's new clothes object-oriented" ( which I've got if you want to read it and can't otherwise find it ). A couple years ago, people thought XML would eliminate proprietary data formats ( forgetting that it is incredibly easy to create proprietary data formats in XML ). Coming up on the radar we also have design patterns.</troll>

But AI. There are two major extremes in the hard AI camp. One ( that I believe ) is that 1) the brain is a set of nodes and connections 2) if you are modelling the nodes and connections then you are modelling the brain ( ooh, I want to be a brain model ) and 3) the computational hardware to do just that should be available around 2030-2040.

In the 60's however, they thought they could use a variety of shortcuts, and that symbolic output from a gigantic non-discrete network ( your brain ) would be identical to symbolic output from symbol processing computers; it's not, and every other shortcut has failed to be applicable to general intelligence. But that doesn't invalidate the brute force approach of total brain modeling.

Cold fusion. The announcement of cold fusion was premature, but recent work in sonoluminescence is looking interesting ( by interesting I mean we may learn more about physics because of it even though the original goal of cheap energy may not be reached ).

Nanotechnology. I think a number of things will happen here. One, the term will become so devalued that everything new in chemistry, physics, molecular biology, and vlsi will claim to be "leading the way in nanotechnology". So that in a decade some will say nanotechnology is all around us, and others will say it's nowhere to be found.

Second. You really have to specify what your goals are in this field. If you're talking about cutting micromachines down in size by a factor of 10 or 100, then I think that is entirely possible in the near future. If you're talking about self-replicating mechanosynthesizers of diamond for orbiting beanstalks, then obviously that will take a little longer. But as far as I can tell, it seems like just a matter of technology and integration. Are there specific physical principles you see being violated with nanotechnology?
--
Caesars come, and Caesars go, but Newton lives forever

When Prince Charles is worrying about... by the (3.00 / 0) #39 Mon Jun 21, 2004 at 12:24:58 PM EST

...grey goo you know that someone, somewhere, is exaggerating about what is and isn't a real. We may as well worry about AIs taking over the world - but I forget, people do. It's no surprise that the foreword to Engines of Creation was written by an AI-is-just-round-the-corner cultist (or that Drexler references Doug Lenat, the last great classical AI scam artist).

But I'm not talking about dead ends. I think that many of the people making exaggerated claims for nanotech know exactly what they are doing. It just seems too implausible to me to believe that otherwise intelligent people think that the grey goo scenario is a risk we need to consider when developing nanotech. This isn't to say it doesn't make great speculative writing or fun science fiction.

But I don't see any fundamental laws preventing nanotech, after all life is basically nanotech.

--
The Definite Article

[ Parent ]

Grey Goo by Wise Cracker (3.00 / 0) #40 Mon Jun 21, 2004 at 01:02:07 PM EST

The people who worry about grey goo are not the real scientists. They're mainly anti-technology tree huggers, effete nobodies looking to appear down with postmodern fears ( Charles ), and policy wonks ( Drexler is more policy wonk than scientist ).

But I don't see any willful deception. I see a set of things that are not physically impossible and people who make a series of confusions: not physically impossible -> physically possible with known science -> physically possible with known technology -> not hard -> easy -> hey, we could knock that off in a weekend.

There are many things that were thought to be scams that proved to be quite possible:

Babbage's difference engine. People thought it was absurd that a machine could calculate, and indeed his never did, but a hundred years later they were changing the world
Home networking terminals. In the seventies there were all these people saying that one day we'd all be hooked up to some teletype where we could get stock quotes, read library materials, and strengthen our republic by becoming better informed voters. The idea then died for twenty years. Today, we do all those things, but mainly use it for chatting. So if you had told people thirty years ago, "Hey, we have to build this internet so one day everyone can chat about their relationships and download porn", I think people would not have seen that as a legitimate motive for action.

I think that current goals nanotechnologists have are likely bogus, but that the tools they produce will lead to products we will not understand how we lived without. The interesting thing to me is the idea of self-replication ( not just in nanotech, although it seems easier there ), and just a little bit of that could change absolutely everything.

So yes, there's hype, and right now it's ignorant hype, but in twenty years I think it will seem like there wasn't enough hype. I mean now, why didn't people get off their butts and deploy the bandwidth of today twenty years ago? Can you imagine having grown up with the Internet? Twenty years from now, will our kids demand to know why we didn't get off our butts and make simple self-replicating systems happen?
--
Caesars come, and Caesars go, but Newton lives forever

[ Parent ]

Go on, I'll bite by Dr H0ffm4n (3.00 / 0) #43 Tue Jun 22, 2004 at 01:09:30 AM EST

Why couldn't you form mathematics as you say Russell insisted?

[ Parent ]

That wasn't the troll part by Wise Cracker (3.00 / 0) #44 Tue Jun 22, 2004 at 01:59:33 PM EST

Godel's theorem was considered a body blow to the logical positivists in general and Russell in particular. He was in the middle of writing a multi-volume encyclopedia on mathematics. He gave it up when the theorem was published. The idea that you can start with elements of mathematics and then construct conjectures which can neither be proven nor disproven put a sort of boundary on the knowable and was in direct conflict with the positivists' belief in potentially unbounded rationality. A quote from Wikipedia's Godel entry:

These theorems ended a hundred years of attempts to establish a definitive set of axioms to put the whole of mathematics on an axiomatic basis such as in the Principia Mathematica and Hilbert's formalism

--
Caesars come, and Caesars go, but Newton lives forever

[ Parent ]

Russell was spent by 1930 by Dr H0ffm4n (3.00 / 0) #45 Tue Jun 22, 2004 at 08:21:14 PM EST

Just because he gave up does not mean that you couldn't form mathematics starting with a set of axioms (say PM) and accepting only provable formulas as universally valid. In fact there are very few meaningful or useful mathematical theorems known to be unprovable from PM, even fewer from ZFC.

[ Parent ]

You know Goodstein's Theorem don't you? by the (3.00 / 0) #46 Wed Jun 23, 2004 at 03:15:32 AM EST

It's kinda disturbing that you can't prove it in PM, it doesn't seem like anything that's particularly weird like the halting problem.

--
The Definite Article

[ Parent ]

: yes by Dr H0ffm4n (3.00 / 0) #47 Wed Jun 23, 2004 at 03:36:00 AM EST

But it's not exactly useful. It's still pretty pathological too. No arithmetical theorem that would be useful if valid has ever been shown to be independent of PM, much less ZFC which can prove Goodstein's.

[ Parent ]

OT: Aren't you in the animation business? by Wise Cracker (3.00 / 0) #41 Mon Jun 21, 2004 at 02:42:40 PM EST

Is there some kind of canonical list of animation shows and the techniques they use? Like I could never figure out if "Jay Jay the Jet Plane" was CGI, or miniatures, or some combination.
--
Caesars come, and Caesars go, but Newton lives forever

: Never heard of that show by the (6.00 / 1) #42 Mon Jun 21, 2004 at 04:17:18 PM EST

But googling on '"jay jay the jet plane" render' reveals a bit about the CG involved.

--
The Definite Article
[ Parent ]

Great scams of the academic community | 47 comments (47 topical, 0 hidden) | Trackback