String research quality for 2007 (guess the cites)

[Martin]

In another thread the issue came up of judging the scientific QUALITY of research. Department hiring and tenure committees do that in part by looking at the CITATION COUNTS of the papers somebody's written. Lots of cites mean that other researchers in the same field consider the work significant and relevant to their own line of inquiry. Like any numerical measure, cite-counts have their limitations---sometimes one's subjective feel can be a truer guide to the real quality.

 

Be that as it may, the SPIRES system at Stanford-SLAC puts a lot of effort into keeping citation counts for everybody's papers in hep-theory and hep-phenomenology, as well as a bunch of other fields. You can easily look up any paper and see how often it has been cited, and they have "Top Forty" lists as well:D

 

Each year since 2002, string researchers have shown DECLINING INTEREST IN THEIR OWN RECENT WORK. This is quite remarkable. It seems quite unnatural and I think it could turn around. By recent, I mean "published in the past five years"---so in 2002 it would be papers published in 1998-2002. In 2006 it means published in 2002-2006. I decided to track the trend by looking, in each year, at the eight recent topcited papers: those which garnered the most citations in that year, and averaging up the counts.

 

Assuming I did the arithmetic correctly, I get:

 

year average cites for top eight

2002: 357

2003: 243

2004: 145

2005: 117

2006: 119

 

WHAT DO YOU PREDICT THIS STRING CITATION INDEX WILL BE IN 2007?

Notice the signs of a turnaround. It could rebound to the 2004 level, even higher!

 

We won't know until sometime in January 2008, when Spires makes the final tallies. Here are the two relevant lists they put out for 2006. The URLs for this year will presumably be the same, just with the year changed.

 

http://www.slac.stanford.edu/spires/topcites/2006/eprints/to_hep-th_annual.shtml

http://www.slac.stanford.edu/spires/topcites/2006/eprints/to_hep-ph_annual.shtml

 

If you take the eight most highly cited string papers from those two lists which were published in 2002-2006 and average the eight cite-counts you get 119. I will do the same thing next year to see whose guess was closest.

===================================

Here is that HEPAP report where they polled university physics departments to get their hiring plans and put it together into a composite picture

 

http://www.science.doe.gov/hep/ugpsreportfinalJuly22,2007.pdf

 

High Energy Physics Advisory Panel

is a blue ribbon standing committee set up to advise DOE and NSF on funding for high energy physics----this was a subpanel focusing on Federal grants to universities---i.e. on support that goes thru the universities instead of, for instance, to the National Labs.

 

Picture was that departmental plans, in aggregate, envisage a 20 percent cutback in number of stringtheorist on the faculty over the next 5 years (by 2012). this report came out in late July 2007.

 

See Figure 3 on page 43. (because of ten extra pages of TOC and intro this registers on computer as page 53)

[YT2095]

I`ll guess at stay more or the less the same as 06 (flat).

[BenTheMan]

Picture was that departmental plans, in aggregate, envisage a 20 percent cutback in number of stringtheorist on the faculty over the next 5 years (by 2012). this report came out in late July 2007.

 

This is because departments are hiring LHC people.

 

Martin---every post you make has the purpose of trashing string theory.

 

The reason the index that you've arbitrarily created shows a decline is because so many people are still working on AdS/CFT.

 

Again, when you create the meter stick, you can measure things as you wish.

[Martin]

This is because departments are hiring LHC people.

 

Read the data on page 43 please before you answer. There was a big planned increase in astrophysics (which has only the most tenuous connection with LHC). There was a smaller planned increase in phenomenonlogy. Something else is obviously going on besides LHC.

 

The reason the index that you've arbitrarily created shows a decline is because so many people are still working on AdS/CFT.

 

You are mistaken. Of course I included AdS/CFT in the citation counts! I also explicitly included AdS/CFT in the publication index. It is one of the keywords I used! The annual publication rate must be declining for some other reason.

As for the citation counts, you are welcome to add them up for yourself if you think I missed any stringy papers. I just take the top eight recent ones, recent meaning in past five years. Of course that is arbitrary but some finite cutoff is needed when you say recent.

 

Again, when you create the meter stick, you can measure things as you wish

 

I try to make objective yardsticks and use them consistently. I am not interested in fooling myself, I want to know what is really going on. On the other hand my impression of you was influenced by the fact that that you declared that the string approach "has no competition" as a quantum theory of gravity and yet were unwilling to listen to Reuter's talk (at Loops 07) or read any of his papers. You talk as if you don't KNOW anything about the competition and above all don't want to find out! This is what I would expect of people in a delusional community, not a community of physicists.

 

Note that I am not accusing you of belonging to a delusional community just pointing out the resemblance. The head in the sand thing. Your excuse is you are too busy to even glance at papers. OK so that is fine.

 

Martin---every post you make has the purpose of trashing string theory.

I want to know what is going on in science establishment of my country.

I don't want to be fooled by self-serving propaganda----or popularization garbage. I try to check facts and report the truth, as closely as I can determine it.

 

If string starts to make a comeback in the public consciousness, or in jobs and dollar, you can bet that I will report it, whether or not I like it.

 

Come on Mr. Dundee, "sometimes you eat the bear and somethings he eats you". It is better to face the facts cheerfully---isnt that the message?

 

===================================

UPDATE REPLY TO NEXT POST HERE TO SAVE SPACE

show me how to get good low energy phenomenology out of other QG approaches and I will start reading the papers.

http://arxiv.org/abs/hep-th/0610241

Gravity and the standard model with neutrino mixing

Ali H. Chamseddine, Alain Connes, Matilde Marcolli

71 pages, 7 figures

(Submitted on 23 Oct 2006)

 

"We present an effective unified theory based on noncommutative geometry for the standard model with neutrino mixing, minimally coupled to gravity. The unification is based on the symplectic unitary group in Hilbert space and on the spectral action. It yields all the detailed structure of the standard model with several predictions at unification scale. Besides the familiar predictions for the gauge couplings as for GUT theories, it predicts the Higgs scattering parameter and the sum of the squares of Yukawa couplings. From these relations one can extract predictions at low energy, giving in particular a Higgs mass around 170 GeV and a top mass compatible with present experimental value. The geometric picture that emerges is that space-time is the product of an ordinary spin manifold (for which the theory would deliver Einstein gravity) by a finite noncommutative geometry F. The discrete space F is of KO-dimension 6 modulo 8 and of metric dimension 0, and accounts for all the intricacies of the standard model with its spontaneous symmetry breaking Higgs sector."

[BenTheMan]

Of course I included AdS/CFT in the citation counts!

 

Perhaps I misunderstood---the AdS/CFT papers (Maldacena; Gubser, Klebanov, Polyakov; etc...) were written in 1998/1999, which is definitely eight-nine years ago.

 

Note that I am not accusing you of belonging to a delusional community just pointing out the resemblance. The head in the sand thing. Your excuse is you are too busy to even glance at papers. OK so that is fine.

 

Martin---a bit smug here but perhaps I deserve it. Either way, show me how to get good low energy phenomenology out of other QG approaches and I will start reading the papers. Until then, I'll stick to heterotic strings.

 

UPDATE REPLY TO NEXT POST HERE TO SAVE SPACE...

 

Martin---do you understand any of these papers? They look as if they are not written for physicists.

 

It looks kind of iffy to me. Specifically this:

 

Remark 5.1. The estimate of equation (5.10) is obtained under the assumption that the Yukawa coupling for the top quark is the dominant term and the others are negligible. However, due to the see-saw mechanism discussed in §5.3 below, one should expect that the Yukawa coupling for the tau neutrino is also large and of the same order as the one for the top quark.''

 

Clearly this is worng. These guys are saying that the tau neutrino has to be as massive as the top quark, right? So, isn't this pretty much experimentally ruled out by astrophysical bounds, etc? So let's invent a right handed neutrino, and pretend that's what Connes is talking about. This is still no good, because the limits for a right handed neutrino which give light masses to the left-handed neutrinos is somewhere around 10^12 GeV or something. So a Yukawa to the higgs is pretty useless for giving IT mass...

 

I said realistic phenomenology Martin

[Martin]

It's very interesting that you think the Chamseddine et al predictions unrealistic, already ruled out by astrophysical data! I will recopy their paper's abstract, with your comment, and give it some thought.

 

http://arxiv.org/abs/hep-th/0610241

Gravity and the standard model with neutrino mixing

Ali H. Chamseddine, Alain Connes, Matilde Marcolli

71 pages, 7 figures

(Submitted on 23 Oct 2006)

 

"We present an effective unified theory based on noncommutative geometry for the standard model with neutrino mixing, minimally coupled to gravity. The unification is based on the symplectic unitary group in Hilbert space and on the spectral action. It yields all the detailed structure of the standard model with several predictions at unification scale. Besides the familiar predictions for the gauge couplings as for GUT theories, it predicts the Higgs scattering parameter and the sum of the squares of Yukawa couplings. From these relations one can extract predictions at low energy, giving in particular a Higgs mass around 170 GeV and a top mass compatible with present experimental value. The geometric picture that emerges is that space-time is the product of an ordinary spin manifold (for which the theory would deliver Einstein gravity) by a finite noncommutative geometry F. The discrete space F is of KO-dimension 6 modulo 8 and of metric dimension 0, and accounts for all the intricacies of the standard model with its spontaneous symmetry breaking Higgs sector."

 

Your response then is to cite their note 5.1 and say:

...

Clearly this is worng. These guys are saying that the tau neutrino has to be as massive as the top quark, right? So, isn't this pretty much experimentally ruled out by astrophysical bounds, etc?..

 

One reason this is interesting is that the guy who is organizing "Loops 08" (which will be called QG²) got essentially the same results as Connes at about the same time----he and Connes brought out their preliminary papers within a week of each other back in August 2006.

He and Connes seem to be on very good terms and I suspect that Connes will be a featured speaker at "Loops 08".

Another geometer, Shahn Majid, is on the scientific advisory committee for the conference, along with Lee Smolin, Carlo Rovelli John Baez and so forth. It should be a very interesting conference. the full title is "Quantum Geometry and Quantum Gravity 2008".

 

Anyway plans for the June conference are certainly not definite but Connes could easily be giving one of the invited plenary talks.

You are saying that his prediction of tau neutrino mass is already ruled out. I'd like you to get the opinion of one of your professors about this. I would be extremely interesting if you could quote somebody reputable to the effect that Connes predictions are wrong. It would be great!

 

======================

UPDATE REPLY TO NEXT POST (#7)

Mr. Dundee, read my post. We don't have to talk so much. I showed you a paper (not saying anything about it).

You said "CLEARLY THIS IS WORNG."

I say great. this is an important paper by a major guy. So I ask you to BACK IT UP. It would be great to have confirmation that one of Connes predictions is already ruled out, as you say, by "astrophysics". So I would like you to get some quote from one of your professors---to show that some more experienced person agrees with you. Get me some reputable basis for saying Connes is wrong. I'll be glad because theories are supposed to be falsifiable. Connes geometry is a really major non-string QG approach and it is making falsifiable predictions so (unlike some other theories or almost-theories) it is really AT RISK of falsification, even by LHC! So let's not slither around this, back up what you said. Ask one of your profs about the paper and get back to us.

[BenTheMan]

Martin----

 

It is possible that there is something I'm missing. It seems that Yukawa couplings of the tau neutrino to the higgs should be forbidden, especially couplings comparable to the yukawa coupling to the top. Plus, I thought that they were giving neutrinos masses with the seesaw mechanism. Don't you think so?

 

If I'm wrong, my feelings won't be hurt if you say so. And the paper only claims to yield the structure of the standard model, not the yukawas. You seem to have claimed that this paper produced realistic phenomenology, and I found something that contradicted this claim. Perhaps you can correct me?

 

Again, maybe there's something I'm missing. I didn't study the details of the paper (nor do I really plan to), but I assumed that perhaps you understood some of the finer points.

[Martin]

I`ll guess at stay more or the less the same as 06 (flat).

 

Hi YT,

that is an astute guess. I am thinking more and more that the odds are for flat too. Rebound looking less likely.

 

Martin----

 

It is possible that there is something I'm missing. It seems that Yukawa couplings of the tau neutrino to the higgs should be forbidden, especially couplings comparable to the yukawa coupling to the top. Plus, I thought that they were giving neutrinos masses with the seesaw mechanism. Don't you think so?

 

If I'm wrong, my feelings won't be hurt if you say so. And the paper only claims to yield the structure of the standard model, not the yukawas. You seem to have claimed that this paper produced realistic phenomenology, and I found something that contradicted this claim. Perhaps you can correct me?

 

Again, maybe there's something I'm missing. I didn't study the details of the paper (nor do I really plan to), but I assumed that perhaps you understood some of the finer points.

 

Ben, please read my post, I just want you to back up your flat statement that Connes prediction is wrong---already ruled out on astrophysics grounds. I'd say ask a prof about it and get back to us.

[Martin]

Well hey! We have three forecasts, and the results won't be known until after the end of the year.

 

Infinitus is the most optimistic, then me, then YT (who I now think may have it).

 

Infinitus says the average cites will be 160, which would really be quite an improvement over the way it's been in the last couple of years.

I say it might rebound to 140. YT says flat----that is around 120 (like the past two years).

======================

Citations measure how interesting and helpful the other theorists find your paper, because it is the number of times they cite it as a reference in THEIR papers. So it measures how the string community values its own papers. If they cite mostly papers from the 1990s and do not cite RECENT work then it indicates that the recent (past five years) work is less important, less valuable. And vice versa.

 

What we are predicting here is you take the TOP EIGHT RECENT PAPERS in 2007----that are most cited in 2007 among those published 2003-2007 (past five years). And you average how many times each of the eight were cited in 2007. So it is current citations of recent work.

 

the idea is to get a sense of how the situation is going at present, and what the trends are, or the flatness (maybe the trend is no trend) or whatever.

 

The way we will find out who is right (Infinitus, YT, or me) is at the end of the year we will look at these links but for 2007 instead of 2006.

These links are for 2006---if you put them in browser you will see where it says .../2006/... and after yearend when the data is all tallied you can put in 2007 where it says the year.

 

http://www.slac.stanford.edu/spires/topcites/2006/eprints/to_hep-th_annual.shtml

http://www.slac.stanford.edu/spires/topcites/2006/eprints/to_hep-ph_annual.shtml

 

these lists will show the topcited recent stringy papers, and we just average up.

[Martin]

I'll try to guess which eight recent (publication 2003+) string papers will be the most cited in 2007. Whichever they turn out to be, we will average their 2007 cites and see whose prediction came closest. But for now we can't know for sure which those eight will be. Here's what i think will be the lineup:

 

http://arxiv.org/hep-th/0301240

http://arxiv.org/hep-th/0302219

http://arxiv.org/hep-th/0212208

http://arxiv.org/hep-th/0308055

http://arxiv.org/hep-th/0312171

http://arxiv.org/hep-th/0509003

http://arxiv.org/hep-th/0409174

http://arxiv.org/hep-th/0405146

[Martin]

...

Citations measure how interesting and helpful the other theorists find your paper, because it is the number of times they cite it as a reference in THEIR papers. So it measures how the string community values its own papers. If they cite mostly papers from the 1990s and do not cite RECENT work then it indicates that the recent (past five years) work is less important, less valuable. And vice versa.

 

What we are predicting here is you take the TOP EIGHT RECENT PAPERS in 2007----that are most cited in 2007 among those published 2003-2007 (past five years). And you average how many times each of the eight were cited in 2007. So it is current citations of recent work.

...

 

http://www.slac.stanford.edu/spires/topcites/2006/eprints/to_hep-th_annual.shtml

http://www.slac.stanford.edu/spires/topcites/2006/eprints/to_hep-ph_annual.shtml

 

these lists will show the topcited recent stringy papers, and we just average up.

 

This is a really important index to track. Citation counts really are a standard measure of quality of research----how much later research uses it, depends on it, and refers to it.

 

So tenure committees and hiring committees and funding agencies watch the citation count measure. We should watch it too, if we want a realistic idea of what is happening in the field.

 

I see two more people have joined the poll and put in their predictions. We now have FIVE guesses about what this year's citation average will be.

 

What I'm doing, simply to get a definite number is I take the EIGHT MOST CITED RECENT STRING PAPERS in any given year, recent in this case being published 2003-2007, and add up their cites during that year, and divide by eight. It's simple and obvious but it works.

 

We can do the experiment of computing that for the eight papers I picked as likely candidates to be the eight most cited recent string papers in 2007 (I listed them in the previous post) assuming their cites this year will be the same as last, and seeing what it comes out to.

 

Here is what we are trying to predict:

average cites for top eight recent string papers

2002: 357

2003: 243

2004: 145

2005: 117

2006: 119

2007: ?

 

This shows a decline in the citations quality of string research since 2002, and the question would be

Is it going to bounce back? Or stay flat? Or will it sink further like down towards 100?

 

This is a really key indicator of what is happening in the field so I want to get a handle on it and try to predict.

If I take just those 8 that I listed in the previous post, well I don't know yet what cites they got in 2007 but I can estimate by listing what cites they got in 2006, which was 238, 109, 93, 88, 85, 82, 72, 69. If they get in 2007 just the same as what they got in 2006 then the average will be 104.5.

 

Whoah! That is down in the 100 area, which nobody even guessed yet! Could be interesting, have to wait and see.

 

==============

 

I see two more people have joined the poll and put in their predictions. We now have FIVE guesses about what this year's citation average will be.

Infinitus says average cites will be 160, which would be a remarkable improvement!

I say 140, a modest rebound.

YT, bascule and ecoli, and are more conservative, they say 120 which is essentially no change.

 

===============

EDIT to reply to next post. "What happens if you include Maldacena's AdS/CFT paper?" I did include it, of course.

We are gauging the quality of RECENT (past five years) research. That paper was recent in 2002 since it was

published in 1998. So it is included in the 2002 figure. It has not been recent research any time since then.

You already mentioned it and I already told you it was included. What purpose is served by asking again?

[BenTheMan]

What happens if you include Maldacena's AdS/CFT paper?

[Farsight]

String Theory is a dead duck now Ben. The show's over. Time to move on. There's a new kid in town.

 

Martin, here's that paper I referred to re the fine structure constant.

 

http://www.relativityplus.info/

 

See page 27. Yes, it's all back of a fag packet stuff, but it offers a "reasonable" concept. It's a start.

 

There may be some geometrical relationship between the twist associated with charge, and the turn associated with mass that sheds light on the fine structure constant α = e²/2ε0hc. This has a value of circa 1/137, and is the ratio of the energy required to push two electrons together from infinity to some given distance, as compared to the energy of a photon with a wavelength 2π times that distance. As a simple geometrical illustration, if we cut the moebius doughnut and uncurl it to form a cylinder, it would exhibit a 180o barberpole twist. If we then imagine the cylinder to be thin-walled, such that we could slice it down its length and unroll it to spread it flat, this twist is transformed into a diagonal line across a rectangle of width π and length 2π. Sine 0.5 is 30o, which is one twelfth of 360o. Let’s say that a twelfth of the 511KeV electron mass/energy is reaching out as charge. When we push two electrons together, each is coupling with a twelfth of its mass, so we have to multiply a twelfth by a twelfth to get a combined value of 1/144. This is of course not accurate, but we only need to make the twist angle 30.75687º to arrive at 1/11.7047th of 360o, and squaring this gives us the familiar 1/137.

[Martin]

Last time I checked there were five predictions, now there are seven. I'll review the situation.

 

Here is what we are trying to predict:

average cites for top eight recent string papers in 2007

2002: 357

2003: 243

2004: 145

2005: 117

2006: 119

2007: ?

 

 

Is this quality measure going to bounce back? Or stay flat? Or will it sink further like down towards 100?

 

A prediction game like this is primarily a check on how well folks are in touch with reality. You see who has a reasonably accurate mental picture of what's happening, or is just plain lucky. Here are our seven guesses.

 

Fattyjwoods says in 2007 the average cites per paper (in the top eight) will be 180. That would almost be like back in the string glory days, before the 2003 Landscape trouble took effect. Who knows? Maybe.

Infinitus says average cites will be 160, not quite so good as Fatty's figure but still an amazing improvement.

I say 140, a modest rebound.

YT, bascule and ecoli, and are more conservative, they say 120 which is essentially no change.

Phil says 100.

 

==================

EDIT to reply to next post: You have been in the past, with remarkable consistency. If I remember right we were having prediction polls back in 2005 and you were then, and in 2006 too. But this could be the time that breaks your winning streak. I think it'll be nearer to 120.

[ydoaPs]

I'm always right.

[bascule]

A prediction game like this is primarily a check on how well folks are in touch with reality. You see who has a reasonably accurate mental picture of what's happening

 

Let me just take this opportunity to reassure you that my mental picture of what's happening in the world of modern physics is nowhere close to reasonably accurate