Slashdot Mirror

(And many thanks to all the scientists who publish on arXiv).

Wow. That may well have been the least informative popular science article I've read in a few years.

If anyone is interested in the results and the technique they used an abstract and the preprint of their results are available here

There's too much money to potentially be made with some genetic data and the associated research, so people try to guard it. Physicists still share a lot of their research -- witness this e-Print archive, which has a lot of preprints/reprints, mainly in physics, but they're adding other fields such as CS and math.

Their stats show about 2500 article submissions per month lately (it's been increasing pretty much linearly for the past 10.5 years, although I suspect that uploading a revised version also counts as a "submission" in that graph), and about 150,000 connections per day. It's been around for more than 10 years, and is still going strong. It's a great resource; I wish my particular field had something equivalent.

There's too much money to potentially be made with some genetic data and the associated research, so people try to guard it. Physicists still share a lot of their research -- witness this e-Print archive, which has a lot of preprints/reprints, mainly in physics, but they're adding other fields such as CS and math.

Their stats show about 2500 article submissions per month lately (it's been increasing pretty much linearly for the past 10.5 years, although I suspect that uploading a revised version also counts as a "submission" in that graph), and about 150,000 connections per day. It's been around for more than 10 years, and is still going strong. It's a great resource; I wish my particular field had something equivalent.

There's too much money to potentially be made with some genetic data and the associated research, so people try to guard it. Physicists still share a lot of their research -- witness this e-Print archive, which has a lot of preprints/reprints, mainly in physics, but they're adding other fields such as CS and math.

Their stats show about 2500 article submissions per month lately (it's been increasing pretty much linearly for the past 10.5 years, although I suspect that uploading a revised version also counts as a "submission" in that graph), and about 150,000 connections per day. It's been around for more than 10 years, and is still going strong. It's a great resource; I wish my particular field had something equivalent.

4 pages in your choice of formats here.

This is not what a quantum gravity researcher would call "a test of quantum gravity", insofar as it does not demonstrate that the gravitational field is quantized. What this is, is a test of the effects on quantum matter of a classical gravitational field. In other words, as the Nature article says, it shows that gravity "can have a quantum effect" on other particles. But it does not show that gravity itself is quantized.

If you have a classical potential well, such as that due to a Maxwellian electric field, or a Newtonian or general relativistic gravitational field, a matter particle in that potential well will exhibit quantization of energy, momentum, etc. As the article says, this happens when the well is confining (when you don't have enough energy to escape the well).

An example is the energy levels of an electron in the electric field of an atomic nucleus, the standard orbitals you get when you solve the ordinary Schroedinger equation. Note that this assumes a potential due to an ordinary, classical electric field.

There are atomic effects due to the quantized electromagnetic field, like the quantum electrodynamics (QED) corrections to the Lamb shift coming from vacuum pair production. They crop up when you assume that the electromagnetic field is made up of individual quanta (photons). These effects are much smaller than the dominant, lowest-order classical effect.

So, what these researchers have done is demonstrated that a classical gravitational potential well can lead to quantized observables for matter, like the electronic orbitals of an atom. This is interesting by itself, because the gravitational field is so weak that the Earth's gravitational potential well is relatively much more "shallow" than the electric potential well of an atomic nucleus, as far as the strengths of the forces are concerned.

However, they have not shown that the gravitational field is itself quantized, any more than the quantized orbitals of electrons demonstrate the electromagnetic field is quantized. So they have not provided evidence for quantum gravity (i.e., a quantized gravitational field), any more than Bohr's law for atomic energy transitions provides evidence for QED.

True tests of quantum gravity are much harder than even this difficult experiment. To read about some proposals, try this paper on Planck scale phenomenology by Amelino-Camelia. (You can also see some of his other papers.)

This is not what a quantum gravity researcher would call "a test of quantum gravity", insofar as it does not demonstrate that the gravitational field is quantized. What this is, is a test of the effects on quantum matter of a classical gravitational field. In other words, as the Nature article says, it shows that gravity "can have a quantum effect" on other particles. But it does not show that gravity itself is quantized.

If you have a classical potential well, such as that due to a Maxwellian electric field, or a Newtonian or general relativistic gravitational field, a matter particle in that potential well will exhibit quantization of energy, momentum, etc. As the article says, this happens when the well is confining (when you don't have enough energy to escape the well).

An example is the energy levels of an electron in the electric field of an atomic nucleus, the standard orbitals you get when you solve the ordinary Schroedinger equation. Note that this assumes a potential due to an ordinary, classical electric field.

There are atomic effects due to the quantized electromagnetic field, like the quantum electrodynamics (QED) corrections to the Lamb shift coming from vacuum pair production. They crop up when you assume that the electromagnetic field is made up of individual quanta (photons). These effects are much smaller than the dominant, lowest-order classical effect.

So, what these researchers have done is demonstrated that a classical gravitational potential well can lead to quantized observables for matter, like the electronic orbitals of an atom. This is interesting by itself, because the gravitational field is so weak that the Earth's gravitational potential well is relatively much more "shallow" than the electric potential well of an atomic nucleus, as far as the strengths of the forces are concerned.

However, they have not shown that the gravitational field is itself quantized, any more than the quantized orbitals of electrons demonstrate the electromagnetic field is quantized. So they have not provided evidence for quantum gravity (i.e., a quantized gravitational field), any more than Bohr's law for atomic energy transitions provides evidence for QED.

True tests of quantum gravity are much harder than even this difficult experiment. To read about some proposals, try this paper on Planck scale phenomenology by Amelino-Camelia. (You can also see some of his other papers.)

Sadly, you are partially correct.

Going to long distance means (in this context), that the required resources increase only polynomially with the distance (opposed to an exponential increase if you do not use a quantum repeater).

Theoretical means, that there is no experimental implementation of a quantum repeater yet. Nevertheless, it would be much easier to build it than, say, a fault tolerant quantum computer.

Using these techniques, it is possible to build secure switchable user-to-user channels, which are secure even if all repeater stations or switches are under the control of an allmighty eavesdropper (Eve). As always, Eve can prevent you from communicating, but if you can communicate, you can be sure to communicate privatly.

For more details, see e. g. quant-ph/0111066.

IAmHansemann

Try this paper. It's got a figure that provides our best estimate (as of 1998, at least, and to my knowledge it hasn't yet been substantially improved on) for the value of the cosmological constant and fraction of matter, with references of course.

Actually, speaking as someone studying in the field of quantum gravity, the question of whether singularities exist is still very much up in the air, though many people hope that Planckian physics will remove them. There are even some tentative results in that direction. On the other hand, it has been suggested that quantum gravity must contain at least "mild" singularities. So it is rather premature to claim that "most scientists" (or even most quantum gravity theorists) have abandoned the idea of a singularity.

Actually, speaking as someone studying in the field of quantum gravity, the question of whether singularities exist is still very much up in the air, though many people hope that Planckian physics will remove them. There are even some tentative results in that direction. On the other hand, it has been suggested that quantum gravity must contain at least "mild" singularities. So it is rather premature to claim that "most scientists" (or even most quantum gravity theorists) have abandoned the idea of a singularity.

• VTKis a very good package for scientific visualization.
• Maxima is a Free computer algebra system, a bit like Mathematica. It can solve equations, do calculus, plot things, produce TeX output of what you've done, and lots more. Incredibly useful for long tedious bits of algebra.
• gnuplot is a versatile graphing package (2D and 3D, but maxima or VTK are IMO better for 3d stuff). As well as graphing, it can try to fit arbitrary functions to your experimental data.
• LaTeX -- it's very hard indeed to typeset equations better than LaTeX can.

• In June we got the news from the Sudbury Neutrino Oscilloscope that from the detection rates of muon-type and electron-type neutrino's coming from the Sun we should conclude that neutrinos oscillate (change type) and are therefore massive, which is in full contradiction with the SM.
• In March this year the results of the 1999 data of the muon g-2 measurement at Brookhaven National Laboratory showed that the (anomalous) magnetic moment of the muon is not described correctly by the SM. This 'magnetic moment' indicates how much the spin of a muon is affected by a magnetic field (a bit like how quickly a compass needle reacts to a new orientation of the compass). This measurement generated lots of theoretical ideas for mods of the SM and/or signs of supersymmetry and what not.
• The Standard Model is ugly.

Most scientists keep bound lab notebooks with hand-written notes, and those hand-written notes are vital for understanding the data. Ten years later, the notebook is probably gone, and the data are useless. For instance, I did nuclear physics experiments ca 1987-1995. The 8-mm tapes probably still exist in New Haven and Chicago, and there is probably still electronic documentation saying that parameters 101-150 are the energies recorded by detectors 1-50. What's missing is my notebook, where I have things written down like "3 am, realized target had partially melted, all data since 1:30 am should be thrown out."

This kind of thing is an issue regardless of whether data are preserved in electronic or dead-tree form. That's why people years later really have to depend on the published papers, not the original data.

For actual papers, the permanence of the formats is not as big an issue. The formats used by the popular preprint servers (e.g. LaTeX, PS, and PDF used by arxiv.org) are going to be decipherable until the fall of western civilization.

There is a nice front end for the math articles in the arXiv. This FAQ has info about contributing math preprints to this well-run electronic preprint resource.

There is a nice front end for the math articles in the arXiv. This FAQ has info about contributing math preprints to this well-run electronic preprint resource.

Loads of papers, refereed and non-refereed are available at arXiv.org . The site is mainly for physics, math and related 'hard-core science'.

Many people submit their papers to arXiv immediately after getting it accepted to a refereed journal. When I try to keep up to date, I do not use paper versions that come out months after they have been published at arXiv. I look at the relevant sections of arXiv. If something is not on arXiv, it is not news.

Physicists have actually started bypassing the reviewing/printing system by putting up arXiv.org long ago. Mathematicians have then followed, and other scientists are starting doing it now. Daniel Bernstein has some very useful advice for authors at his web page at http://cr.yp.to/bib.html.

You know, I think scientists are starting to wake up. For one thing, the equally prestigious magazine Nature had a short note recently about Dmitry's case, which was clearly sympathetic towards him.

Also, you have 27758 scientists signing the Open Letter of the Public Library of Science, and you've got physicists publishing pretty much all their material as pre-prints.

I don't think the open systems that science requires to function can co-exist with the closed systems wanted by the entertainment industry. If an open system exists, it can always be used to circumvent a closed system.

Now, it is easy to demonize "hackers" but it is harder to demonize scientists. Therefore, I think the first real battle will be over scientific publishing, and I want to be there when it happens.

Now, I don't think it will be a battle between scientists and artists, though the entertainment industry may try to portray it as such. The openness established by scientists and scientific publishing will be good for the whole of society, stimulate cultural diversity, and art will flourish along with science.

HTML might be fine if all you want to do is send a picture with an explanation of what it is. But even HTML isn't necessarily modifiable the way you want. For instance, equations and line art have to be converted to hard-to-edit bitmaps. Personally, I find HTML in e-mail a big nuisance; for instance, it can have web bugs in it. And I suspect the reason a lot of people in the bidness world send Word attachments is that they want it to look snazzier -- HTML generally looks pretty awful by comparison, e.g. many browsers don't support justified text.

The real issue is that HTML, PDF, and LaTeX are all open formats, and don't have the problem Word has with earlier versions of the software being unable to read files written by later versions.

There exists an entire sub-culture who believe in the Great Conspiracy and those who try to refute QM are no different, tilting at windmills while the rest of us get on with our lives and work with with the best available evidence. Ultimately, the anti-Non-Locality folks have to believe in a grand conspiracy of nature to make their ad hoc models even come close to reproducing the non-local correlations observed in quite a few experiments (using different systems, methodologies). We have had experiments which closed the locality loophole (Gisin, Geneva), experiments which closed the detection loopholes (Wineland, Boulder), and there are proposals to close both at the same time.

As new experiments are done, improving on the previous ones, the results merely confirm the predictions of QM, no sign of deviation from what we should expect. Not to mention that QM is the basis of electronics, which seems to work to a remarkably high degree, for a supposedly `wrong' theory. I see a parallel to the `Creation Science' debacle, where anti-evolutionists play their `argument from incredulity' card and try to poke holes into a straw man of their devising. Those rejecting the established results of QM are our latter-day Flat Earthers. I admire their ability to perform logical and factual contortions to support their untenable position.

As for the funding situation, the amount of money spent on Quantum Computation is mere loose change, just enough to keep mainly theoreticians lean and mean, with a few small scale experiments running (we're not talking SSC here). A 10 year time-line for the development of a fully functioning quantum computer is a stretch though, the technical challenges are very formidable, and it may follow the Fusion Power story. Still, the fundamental research in Quantum Information Theory is still very worthwhile in its own right, whether or not we can build a scalable quantum computer. The outlook for quantum crypto is much better, commercialised systems may come out within a 10 year horizon.

Sure, there is a lot of hype, and I cringe everytime I read a popular account of QC for all the misunderstandings promulgated, but it is a very serious research programme. Through considerations of quantum information, we may find a way of unifying the two great pillars of modern physics, GR and QM.

There exists an entire sub-culture who believe in the Great Conspiracy and those who try to refute QM are no different, tilting at windmills while the rest of us get on with our lives and work with with the best available evidence. Ultimately, the anti-Non-Locality folks have to believe in a grand conspiracy of nature to make their ad hoc models even come close to reproducing the non-local correlations observed in quite a few experiments (using different systems, methodologies). We have had experiments which closed the locality loophole (Gisin, Geneva), experiments which closed the detection loopholes (Wineland, Boulder), and there are proposals to close both at the same time.

As new experiments are done, improving on the previous ones, the results merely confirm the predictions of QM, no sign of deviation from what we should expect. Not to mention that QM is the basis of electronics, which seems to work to a remarkably high degree, for a supposedly `wrong' theory. I see a parallel to the `Creation Science' debacle, where anti-evolutionists play their `argument from incredulity' card and try to poke holes into a straw man of their devising. Those rejecting the established results of QM are our latter-day Flat Earthers. I admire their ability to perform logical and factual contortions to support their untenable position.

As for the funding situation, the amount of money spent on Quantum Computation is mere loose change, just enough to keep mainly theoreticians lean and mean, with a few small scale experiments running (we're not talking SSC here). A 10 year time-line for the development of a fully functioning quantum computer is a stretch though, the technical challenges are very formidable, and it may follow the Fusion Power story. Still, the fundamental research in Quantum Information Theory is still very worthwhile in its own right, whether or not we can build a scalable quantum computer. The outlook for quantum crypto is much better, commercialised systems may come out within a 10 year horizon.

Sure, there is a lot of hype, and I cringe everytime I read a popular account of QC for all the misunderstandings promulgated, but it is a very serious research programme. Through considerations of quantum information, we may find a way of unifying the two great pillars of modern physics, GR and QM.

On http://xxx.lanl.gov/RobotsBeware.html:
Click here to initiate automated "seek-and-destroy" against your site.
Uh, what exactly is that supposed to do???

Specifically, as to van Flandern, here's the comment I made in response to the earlier guy who brought up the same web page:

van Flandern is a crank who doesn't understand gravitational radiation, despite careful correction. (Though he has done some decent work in other areas.) This issue has been beat to death on Usenet. See, for instance, the FAQ, Carlip's correction of van Flandern, Hillman's archive of a Usenet discussion with van Flandern, etc.

Does anyone have a link to an article talking about how in tracking V'ger and the other solar-system-exploring satellite thingies, they have discovered that they seem to be moving slightly slower than they're supposed to be?

van Flandern is a crank who doesn't understand gravitational radiation, despite careful correction. (Though he has done some decent work in other areas.) This issue has been beat to death on Usenet. See, for instance, the FAQ, Carlip's correction of van Flandern, Hillman's archive of a Usenet discussion with van Flandern, etc.

Quick, someone mirror the PDF Version before the whole site dies under our collective weight.

If no one manages to do so, here's his Abstract:


Abstract
The detection of apparent anomalous forces in the vicinity of high-Tc supercon-
ductors under non equilibrium conditions has stimulated an experimental research in
which the operating parameters of the experiment have been pushed to values higher
than those employed in previous attempts. The results conrm the existence of an
unexpected physical interaction. An apparatus has been constructed and tested in
which the superconductor is subjected to peak currents in excess of 10 4 A, surface
potentials in excess of 1 MV, trapped magnetic eld up to 1 T, and temperature
down to 40 K. In order to produce the required currents a high voltage discharge
technique has been employed. Discharges originating from a superconducting ceramic
electrode are accompanied by the emission of radiation which propagates in a focused
beam without noticeable attenuation through dierent materials and exerts a short
repulsive force on small movable objects along the propagation axis. Within the
measurement error (5 to 7 %) the impulse is proportional to the mass of the objects
and independent on their composition. It therefore resembles a gravitational impulse.
The observed phenomenon appears to be absolutely new and unprecedented in the
literature. It cannot be understood in the framework of general relativity. A theory
is proposed which combines a quantum gravity approach with anomalous vacuum
uctuations.


I think that last sentence says it all. I don't buy it.

1. Very long term storage. Solved to the extent that any long term storage of electronic media is.
2. Immutability. The arXiv does not allow anyone, the author(s) included, to modify a submitted article. The author(s) may submit revisions in which case the old version remains and withdraw a paper in which case it is not withdrawn from the archive but marked as withdrawn by the author(s). Of course, this does not guarantee full immutability as the maintainers of the archive could possibly make modifications. However, the archive is mirrored and if it's considered a problem one could probably think up some electronic signature scheme.
3. Currently this archive coexists with ordinary journal publication. This presents an interesting problem as when an article is published the author usually (for the journals I am aware of at least) signs away the rights, yet the arXiv will not remove the submitted paper. I have really no idea of the legal ramifications of this (what will happen if the publisher demands that the paper be retracted from the arXiv?). So far this problem has not arisen (as far as I know).

This was started by a lone eccentric^H^H^H^H visionary genius named Paul Ginsparg, a physicist employed at Los Alamos National Laboratory (yes, tax dollars at work!), in 1991. The site is nearly self-maintaining and serves up hundreds of thousands of e-prints a year to physicists and mathematicians worldwide. For those who are interested, this report gives Dr. Ginsparg's view of the archive circa 1996.

The New York Times had a nice article about the archive on May 1 of this year (now only accessible via pay at the "Premium Archive"... how's that for irony?) and how it has levelled the scientific playing field for researchers in less-developed countries who cannot afford premium journal subscriptions.

The most important point, though, is that this free e-print service coexists with the high-subscription journals (Phys Rev; Science; Nature; ApJ) that serve these communities. Young researchers bucking for tenure submit to the prestigious journal... and also to arXiv.org. That way their research gets both respect and the broadest possible distribution.

This sort of compromise is going to be harder to achieve, I think, in fields where publishing is an even bigger-money business (MoBio comes to mind). But it does demonstrate that full-scale war is not the only alternative to all-out capitulation to commercial interests.

-Renard

This was started by a lone eccentric^H^H^H^H visionary genius named Paul Ginsparg, a physicist employed at Los Alamos National Laboratory (yes, tax dollars at work!), in 1991. The site is nearly self-maintaining and serves up hundreds of thousands of e-prints a year to physicists and mathematicians worldwide. For those who are interested, this report gives Dr. Ginsparg's view of the archive circa 1996.

The New York Times had a nice article about the archive on May 1 of this year (now only accessible via pay at the "Premium Archive"... how's that for irony?) and how it has levelled the scientific playing field for researchers in less-developed countries who cannot afford premium journal subscriptions.

The most important point, though, is that this free e-print service coexists with the high-subscription journals (Phys Rev; Science; Nature; ApJ) that serve these communities. Young researchers bucking for tenure submit to the prestigious journal... and also to arXiv.org. That way their research gets both respect and the broadest possible distribution.

This sort of compromise is going to be harder to achieve, I think, in fields where publishing is an even bigger-money business (MoBio comes to mind). But it does demonstrate that full-scale war is not the only alternative to all-out capitulation to commercial interests.

-Renard

And you're right: no, we're not going to see a quark-gluon plasma emerge. You need very high temperatures to reach a phase transition. Confined quarks are the low energy state.

And you're right: no, we're not going to see a quark-gluon plasma emerge. You need very high temperatures to reach a phase transition. Confined quarks are the low energy state.

There are a lot of explanations going around this thread about how core collapse supernovae occur. Some good, some terrible, none quite right. Rather than correct what's been said, I'll instead point y'all to a few of the sources of real info out there.

A good place to start is the NASA Observatorium page on Stellar Evolution and Death

A friend in the business maintains a page full of links to SN pages. Many of these are links to research groups, but there are also links to general education and image catalogs.

BTW, in case you don't believe that I know of what I speak, follow this link

There was the paper (abstract here, paper here (PDF)) mentioned in the Slashdot article here about the resilience of the 'net; crash 99% of the nodes at random and it'll still run. Which isn't bad.

Problem is of course when you crash the <1% of nodes that actually do the major routing.

Routing's getting hairier and hairier; it should really get fun once IPv6 kicks off and everyone and their dog have a squillion IP addresses each.

Yup. Though if the main site has blocked you, you should still be able to get in through a mirror site - we operate one at:
http://aps.arxiv.org/.

Send me email if you can't get in there.

today's usage for arXiv.org (not including mirrors)

Paul Ginsparg maintains the xxx.lanl.gov LANL e-print arXiv. Ginsparg has a position paper on his (biased, of course) opinion the situation; there is also an update.

Paul Ginsparg maintains the xxx.lanl.gov LANL e-print arXiv. Ginsparg has a position paper on his (biased, of course) opinion the situation; there is also an update.

Paul Ginsparg maintains the xxx.lanl.gov LANL e-print arXiv. Ginsparg has a position paper on his (biased, of course) opinion the situation; there is also an update.

In the current experiment by Phillips and co-workers which is only a demonstration of the ideas and not a useful device, they can only stop about half the light pulse and it decays away inside the atomic cloud fairly quickly, so only a small fraction of the incident photons actually come out.

1. Most papers have multiple authors. Once the work is ready to be written up, one author writes the actual text, and then iterates with the coauthors for a while.
2. They start giving talks about their work, and getting more feedback that way.
3. They post the paper to a preprint server such as arxiv.org.
4. They submit it to a journal.
5. The journal solicits anonymous peer reviews from people it knows are active in the field.
6. The peer reviewers typically ask for some (major or minor) changes, and then recommend publication (unless the whole thing was a botch).

The reason #4-6 used to be required, not optional, is that there was no such thing as zero-cost electronic distribution.

BTW, cold fusion was messed up because Pons and Fleischman never even made the relevant info public. They bypassed the whole publication process and just gave press releases. Other scientists who wanted to try to reproduce their results were reduced to stuff like looking at photos and trying to estimate the size of an electrode by comparing it with the diameter of Pons' wrist!

The Assayer - free-information book reviews

Take for instance Mitra's paper astro-ph/9904162. Look where he claims that the radial geodesic of a massive particle becomes null at the horizon. That's trivially false; any initially-timelike geodesic always remains timelike, and in fact proving that is what Misner et al describe as "the easiest exericise" in their entire text. You merely have to realize that geodesics parallel-propagate their own tangent vectors, and parallel tansport preserves the lengths (and hence timelike/spacelike nature) of 4-vectors. Since the remainder of Mitra's claims in that paper (namely, that black hole interiors don't exist, etc.) are based on this incorrect assertion, that entire paper is worthless.

If you examine Mitra's "proof", you find that instead of proving that the geodesics of massive infalling particles become null at the horizon (as he claimed), what he actually proved was that the horizon itself is a null surface. That's well-known and doesn't "contravene the premises of GTR" nor imply that the event horizon is a true singularity, as Mitra claims. In fact, it's quite amusing that Mitra insists that the horizon is a "true singularity", when the curvature is perfectly well-bahaved in the very Kruskal-Szekeres coordinates he's using.

The rest of Mitra's papers have nonsense like this all throughout (and are based on the incorrect work in the above paper), but this one is the one that I bothered to study in detail.

In fact, I think Slashdot is at its worst when it gets onto topics that aren't part of its core unixy mission, i.e. when the editors aren't choosy enough. Some of the dopiest discussion happens when Slashdotters start expounding on subjects they know nothing about, or when the Slashdot editors make bad choices of articles on physics, astronomy, etc. (My own pet peeve is all the physics posts that don't link to the original paper, which is almost always available on arxiv.org.)

The traditional response to people who don't like getting their Slashdot submissions rejected is that anyone who doesn't like Slashdot's editorial policies can always start their own Slash clone. This is a valid response, but I think it's interesting that Glasscode seems to make more of an effort to encourage people to start up new sections on a variety of topics. It'll be interesting to see if the quality of discussion on all these sections turns out to be good.

The Assayer - free-information book reviews

The original paper is here.
The Assayer - free-information book reviews

See the HyperTeX FAQ for details.

I like LaTeX's ability to separate semantic structure from layout logic, but any language that will allow style sheets can do the same thing, including HTML4. Also, TeX has a Turing-complete macro language, which I tend to dislike in a document description language. So while I like some aspects of this idea, I can't altogether support it.

--

First, our work deals with resilience of many kinds of networks, not just the Internet, and may be a better model of some of the others than it is of the Internet. As several people have pointed out, preserving connectivity in the Internet is kind of pointless if the performance is appalling, as it would be if you destroyed most of it. (We did actually make this point in an earlier version of the PRL paper, but it got cut out to make way for other things.)

The whole paper is available here, and if you look at it you'll see that we also talk about applications to things like the power grid, and contact networks which result in disease transmission. There are also many other important networks which one could apply it to. Distribution networks, like UPS or regular mail; transportation networks, like airline routes or highways; food webs; neural nets. One of the most interesting, perhaps, is not the Internet, but the Web, which is also a network (of pages linked together by links). This has also been shown to have a scale-free degree distribution (A. Broder et al., Computer Networks 33, 309 (2000)) and so should be highly robust in the sense that you can still surf from one place to another even if most of the pages in the Web disappeared overnight.

To answer the question about non-random failures, yes, things are very different there, and this is also discussed in the article. If you attack only the most highly connected domains/web sites/people/airline hubs, etc. you can destroy the operation of the network in no time at all. These networks are very vulnerable to directed attack. This could be very bad, but it some cases it is good. For instance, it means that a focused attack on a disease-causing contact network, for example through vaccination of the people with the highest number of contacts (the so-called "core group"), could prevent an epidemic with comparitively little effort. (Of course, identifying the core group may not be trivial.)

The guy who made the point about the Australian fiber dying last week is onto the right idea. This is the same effect - if you take out the right connection (or wrong, depending on your point of view), then you can do a lot of damage. It's only if you take out a random one that the network is robust.

All the best,
Mark.

Brendan Koerner was previously a Senior Editor at U.S. News & World Report, where he wrote more than 100 feature articles including six cover stories, since 1996 on subjects ranging from higher education to urban renewal to religion to business and technology. His stories have included "A Lust for Profits," a cover-story, study of the Internet pornography industry, and "Can Hackers be Stopped," a special report on information security.

As a Markle Fellow at the New America Foundation, Mr. Koerner will use the lens of the digital underground to examine some of the most critical issues affecting the future of information technology and the new economy--including the commercialization of the Internet; the struggle over copyright and patent law and the preservation of the information commons; the ethics of free speech on the Internet; and the increasing concerns over cyber-security. In his writing, Mr. Koerner will seek to meld traditional reportage with well-informed commentary that will be solutions oriented. In particular, his work will seek to suggest workable compromises and creative policy alternatives that might allow the wired establishment and its hacker critics to develop a less adversarial relationship.