Several blogs today have discussions of the arXiv trackback issue, mostly spawned by Sean Carroll’s posting on the topic.
Jacques Distler has a posting where he explains that the arXiv has instituted an ill-defined “active researcher” criterion for allowing trackbacks to blogs and that:
Peter Woit’s publication record doesn’t put him anywhere close to “active researcher status”
I’ve written an extensive comment over there about why I happen to think I am an active researcher and the evident absurdity of the idea the Jacques Distler is capable of making a rational evaluation of this question.
More discussion of this is at Chad Orzel’s Uncertain Principles, and Georg von Hippel’s Life on the Lattice.
Update: Yet more on trackbackgate from Lieven le Bruyn , Evan Goer and Capitalist Imperialist Pig. The first has the useful suggestion of using Technorati to automatically generate complete sets of links to discussions of arXiv papers on blogs. The second reminds me why it’s a bad idea to hit the “submit” button when you’re extremely pissed off.
Update: In discussion over at Jacques’s blog about what an “active researcher” is, the dicey issue has arisen that Jacques actually doesn’t seem to be one himself. One commenter has suggested that this whole issue could easily be resolved by just picking a definition of the term, noting that
Few will object to defining a minimallly active researcher as one who has posted an average of 2 papers per year to arXiv over the last 3 years.
Jacques comes no where near qualifying as “minimally active”, since he has only posted 3 papers to the arXiv during the last 3 years.
Personally I don’t think the 2 paper/year criterion is very good. It doesn’t account for length of papers, or that they may be the product of a collaboration, so the author in question is only responsible for a fraction of the paper. A more accurate measure would be based on counting pages of papers and dividing by number of authors. Under this measure, over the last four years Jacques’s research productivity looks like this:
2002: 9.3 pages
2003: 3 pages
2004: 0 pages
2005: 23.7 pages
for an average of 9 pages/year (this count is being a bit charitable, since 15 of the 2005 pages are from a “landscape” paper that may not even be science).
Jacques has made it clear that a certain author for whom this number is 14 pages/year is not “anywhere close to ‘active researcher’ status”, so I guess he is even farther away from qualifying as an “active researcher”.
Always seems surprising the way people living in glass houses like to throw stones…
Update: This particular food-fight has even been written up for the on-line component of Discover Magazine.
Update: A couple more people weigh in on this, Jim Hu and Alejandro Satz.
Update: I haven’t heard anything at all from anyone associated with the arXiv, but a couple trackbacks to one of my recent postings have appeared, so there seems to have been some sort of change of policy there.
Lee Smolin over at cosmic variance very elequontly said:
If academic freedom means anything, it must mean that the university must do nothing to impede free discussion by professionally competent experts on scientific controversies. Given that Peter Woit is a Physics Ph.D. and a faculty member at a major university, who has published papers and has a book in press on the topic, he is without doubt part of the academic community to which the principles of academic freedom apply.
End of discussion
Just a suggestion:
My initial suspicion that it was useless to argue about the “active researcher” criterion, idiotic or no, was I doubted the argument would make any difference. Hopefully I was wrong, and it looks as if maybe I was. Some folks even diametrically opposed to your position on ST clearly don’t think much of the policy, and some have even gone so far as to insinuate it’s likely a load of b.s. contrived as a pretense to do exactly what you claim: Suppress your point of view on less-than-ethical grounds.
I’d be having a Catagory 5 conniption were I in your shoes.
I’d also hope somebody threw a bucket of cold water on me before I fired off too many expressions of indignation, righteous or no, in that frame of mind. It’s something I’ve found out the hard way.
Anyway, best wishes as always, and I’m really sorry for what is obviously a terribly frustrating experience. I hope things all work out for the best in the end.
Thanks for the wise words. So far I only regret one thing I’ve fired off today, perhaps by tomorrow I’ll have realized that more were a mistake. Will keep your advice in mind the rest of the evening….
I belive that Peter’s credentials are more than adequate. For instance, he had recently made a presentation on work by Freed, Hopkins and Teleman, math AT/0206257, which later was elaborated by Atiyah and Segal in math.KT/0510674. I am sure, that majority of physicists either unaware or simply do not have skills to appreciate this important recent work. Peter’s presentation is aimed exactly at this type of audience. I found his explanations very illuminating. The problem is certainly not with his credentials but rather in a shallow level of many discussions tolerated at his blog.
ark, so you mean the blog has to be well-moderated AND the blogger should be an active researcher with decent credentials. That is known as moving the goal post. Anyway, Lubos Motl’s blog has no less shallow level of discussion, and it is permitted to trackback.
I agree, but the issue still remains. I do not have a solution, just an observation. Most likely, there is no solution. The fact that somebody is better or worce is a fact of all our life experiences. But, to move forward with whatever democracy, perhaps, is not the best solution since Brownian motion may or may not lead away from the starting point (all depends on dimensionality of space in which it is taking place 🙂 )
I’m with Smolin!
Since the last remark surely supports my earlier statement, that is back to square one, I was thinking that may be there is at least partial solution after all. For example, everybody can suggest a paper at arxiv which is worthy of discussion provided that the person who makes such a suggestion also states very clearly why he or she thinks the paper is worth discussing. This should include some references to earlier paper(s) which given one improves/disproves. This earlier one(s) should have already some noticeable citation/downloads count thus indicating that it is of general interest to many people. Hence, it makes sence to talk about its most recent development. It would be interesting to see (for everybody) if some agreement can be reached on usefulness of any such submitted paper. Personally, I would be very surprised that such an agreement can be reached. And for this reason I was saying earlier that democracy in science is two edged sword. With all disadvantages of the pyramidal scientific system it would be totally pathetic to equate voices of all teaching institutions (recall departmental meetings at your own scool 🙂 !)
ark – it’s obvious to everyone that Peter has not only brains, but balls.
Question: How many sites have trackbacks blocked?
It’s kind of funny to see how desperately arxiv moderators are trying to throw rationale behind their plain censorship of Woit’s blog. They have even come up with an ad-hoc active research policy to support their case. What an awful shame to be perfectly aware of the fact that Peter wouldn’t be allowed to trackback even if he started to publish papers on a weekly basis!
the level of the discussion is primarily determined by those who debate. I am at least trying to inhibit the kind of “discussion” in the direction of Danny Ross Lunsford, Quantoken, Nigel Cook, and so many others. Peter, on the contrary, likes to build on this kind of “discussion” because what these crackpots are saying is very convenient for his agenda. He likes to be viewed as the hero who fights against the evil science and the evil scientists. At the same time, of course, he is trying to pretend that he is not one of them.
I can’t believe Motl’s name-calling ad homini attack by stereotyping commentators with different ideas as ‘crackpots’ is allowed here.
I have no idea how many such sites there are. Over at Jacques’s blog I asked him for a list, got no answer. Given that it took 3 months of effort to find out that there was a list and that I was on it, and another month to find out why, I suspect that getting this kind of information out of the arXiv might not be easy.
On the whole I prefer to leave Lubos’s comments alone here and not delete them, although I would delete such comments from other people. Seeing what the string theory community considers acceptable behavior by one of its leading “active researchers” and bloggers has a certain educational value.
I found the comment by Dr. Polchinski a bit confusing. I think you were right, Peter, to apologize to him for using less-than-collegial language (I’m also doing my best to be polite and appreciative these days, though I’m sure I’ll fail on occasion), but I didn’t find the basic message to be, in any substantive way, different from what could be found in the Nature column you linked to (the statement “For something more sensible about the anthropic principle, see a recent column from Nature.” was omitted.) Example:
“Why is our Universe so exquisitely tuned to host life? Using the anthropic principle to explain the world might be a tempting alternative to invoking God, but it’s not science, says Philip Ball.*”
(To be fair, Ball seems to be mostly summarizing the ideas of Smolin rather than endorsing them, but the overall message still appears to take a dim view of the Landscape, per the header I quoted above.)
Meanwhile, I’ve seen other bloggers express their own rather unanalytic reactions to the landscape ranging in tone from chilly agnosticism to sneering derision. I’m sure Dr. Polchinski singled out your statement to make an effective rhetorical point, but I’m not sure if it was entirely just to do so.
*Not an “active researcher” but a PhD in physics and an Oxford grad, for what it’s worth.
Polchinski was within his rights to pick out the worst of my anti-landscape postings and complain about it. It’s not one I’m proud of or want to defend, although its main point that this stuff is not science is a valid one, widespread in the physics community at this moment (my allies on this include many string theorists, including Lubos).
It’s interesting that Polchinski chose to write in about this in the context of the arXiv trackback issue. Distler would certainly claim that “trackbackgate” has nothing to do with my opinions about this issue, but I think Polchinski’s comment shows clearly that this is what this is really all about.
This is what happens when the rules are not clear and unambiguous. What’s an “active researcher”? Can you define it? Can you give a metric that will, for anyone alive, tell you unambiguously whether they are an “active reseacher” or not.
If you can’t do this, if you make rules and definitions subjective things, then you are asking for trouble.
i know you are a particle physicist, i am curious as to what your thoughts are with a recent paper by smolin and markapolou that LQG is relevant to particle physics in this recent paper:
we know your skepticism about string theory’s ability to say anything about particle physics.
what are your thoughts about “qg and the sm” deriving the sm from lqg.
do you think it is as promisng, less promising, or more promising than string theory? of course, yours truly, lubos has his negative comments, but he is a string theorist, not a particle physicist.
The Sundance+PI paper talks about String type diagrams – hence it is just as much String theory as LQG – despite the authors’, and others, points of view.
Sorry, I just haven’t had the time to look carefully at this work. Kea’s comment about it is quite interesting. But I know little about it, so don’t want to host a thread here on this topic.
but I think Polchinski’s comment shows clearly that this is what this is really all about.
No. Polchynski maybe quotet that comment which was most enerving to him. If someone would criticise LQG it wouldn’t hurt him that much personally. Polchynski wrote that the trackback system should be shut down. This is not, what distler wants, when you read his reply (old news, he is wasting time with that controversy etc.).
For Distler it maybe the case to let you out of Arxiv. But this is due to the fact that one might imagine it is annoying when a string theorist writes “may be this is a nightmare in which the theory is today”, recognizes that research struggles and then someone with only about 10 papers who published nothing in the field, writes how ugly the work of the stringtheorist is.
I would not want such person, to build a social success to be known in the field upon my failure.
This may be the reason why somepeople won’t led your posts associated with a scientific archive.
They want not write papers and then see automatically generated links to their public demolition from someone who has almost nothing written. This is very good to understand.
I think that the arXiv should keep trackbacks, but just honestly admit that they will not allow trackbacks to Peter’s website. I think that there is a legitimate argument that Peter’s criticisms are not made in good faith, but rather seek to undermine the honest efforts of a whole comunity of scientists. Perhaps Peter thinks that string theory is not worthwhile, but that opinion in itself does not contribute to the scientific discussion, it does not show a way forward. In the meantime theorists will continue to make progress on the problems they can solve.
That being said, I do think Peter’s blog is interesting, and it is worthwhile discussing string theory’s successes and failures on the broadest terms. However, I can appreciate that the arXiv would not want to provide links to comments which only serve to criticize the research of hard-working people trying to solve difficult problems.
Justin, you made the point completely.
This might be the reason, why arxiv moderators wanted only blogs by an active researcher.
That means, they would allow even peter, if he would work hard to try out and got some results. Even if he would find an upper bound for the cosmological constant in string theory, they would accept this.
Polchinski said, that he often cites a paper which is critical on the landscape. I believe that.
The problem is not about the critics but the problem is that one, who never worked out a serious problem on this theory or has only around 20 papers in other fields published makes the failure of others into his triumph without having at least critical results archived.
For someone working hard, and is struggling this is not of any acceptable level. Even when it is true that the media hype string theorists created is indeed dangerous for theoretical physics. One can criticise that, but this even has nothing to do in arxiv.org
Justin and Benni,
I don’t know who you are, but you show no signs at all of understanding what the scientific issues are here. You appear to be students who don’t know what this is all about, but somehow feel comfortable criticizing me (very repetitively in Benni’s case). I’ll delete any further comments from you two unless you have something substantive to say.
The criticisms of Woit I keep reading here, particularly Benni’s, rub me the wrong way. Perhaps it’s a byproduct of seeing CVs filled with mediocre papers churned out by hardworking but not terribly brilliant scientists which they use to propel their careers, or going through mass evaluations solely based on counting impact-factored pubs and invited talks. I don’t know. Regardless, someone with Peter’s credentials who has taken it upon themselves to ask important questions about a field which is consuming significant public funds, questions like ‘is this actually going anywhere?’ , ‘is there a fundamental flaw in what is being done?’ etc (I paraphrase), is doing a very important service to the scientific community and the public trust on which it relies. EVERY field of science should have these kind of questions being asked ALL the time, and in most fields these answers come trivially. (Note that justification based on reduction to technology is not what I am talking about, but the fundamental epistemology that _must_ underpin science.)
His forthcoming book is an important part of this, and I hope it gets at least a fair fraction of the same readership as Kaku’s latest crowd-pleaser, but there is also a community of practicing scientists out there that could benefit from at least having the trackback there, to be followed or not. I don’t see the point in requiring someone critical of a field to do new work in and publish it in that field, thereby lending further creedence to arguments which just _might_ need to be dismissed out of hand, if everyone involved were being honest about what is going on.
What would Fritz Rohrlich say about all this? David Finkelstein? John Wheeler? These people are alive and around, but are as silent as lambs. Even Sarfatti.
Why isn’t Roger Penrose making decisions about trackbacks, rather than Distler?
What a crowd. I’m ashamed to be a part of this, however indirectly.
(Did you know that “Distler” means “chinless” in Halb-Plattdeutsch? And “Motl” means “bumbler”?)
‘Why isn’t Roger Penrose making decisions about trackbacks, rather than Distler?’
I don’t think Penrose would been keen to encourage string theorists who rant and rave about all conceivable alternative programs being crackpot: http://arxiv.org/tb-display/hep-th/0603022
Dear Peter Woit,
Allow me to give a brief announcement and a comment. My blog has now the trackback feature by HaloScan, so now I theoretically can send trackbacks to the arxiv (at least, according to Jacques Distler, I could). I have just sent a trackback to the paper Quantum gravity and the standard model, by Bilson-Thompson, Markopoulou and Smolin [http://arxiv.org/abs/hep-th/0603022]. I believe that it will take some time to appear my trackback link at the arxiv site, though.
I have offered some suggestions concerning the arxiv trackback policy, my comments can be found among the various others at Jacques Distler´s blog, so I will not repeat them here. According to my understanding of what would be a reasonable criteria for accepting trackbacks, you would qualify. The status of “active researcher” cannot be established by the number of papers alone, in my view.
I must confess, in any case, that I have mixed opinions on the truthful importance of trackbacks. Conventionally, if a researcher wants to criticize a given paper, he would normally write a comment to that paper. Or communicate with the author(s), if that is humanly possible. But now, with the internet, things have changed of course and perhaps trackbacks will start having their weight of importance on the self-regulatory nature of the scientific research (or, at least, the way it is supposed to work).
I wish you all the best of luck and hope you gain your rights to the arxiv trackback system.
Thanks, I’ll post some more comments about the “active researcher” issue in a moment. It’s kind of a ridiculous issue, since every blogger involved in this discussion is obviously an active researcher, and some of these blogs (like yours) are quite good. This whole thing would not be especially controversial if it were not for Distler’s desperate attempts to find some kind of reason for not allowing trackbacks to my blog.
Several physicists object to the inclusion of trackbacks to their papers on various reasonable grounds. It seems to me that this can and should easily be addressed by allowing authors to choose not to allow trackbacks to their papers. This question goes way back to more than 10 years ago when I remember Ginsparg discussing a plan to allow linked commentary to arXiv papers. His plan then was to allow authors to delete commentary they didn’t approve of. When asked “what about authors who delete comments explaining why their work is wrong?”, he said something like “if no comments are there, the work is either wrong or not interesting, not clear that you should care which”.
Updated the text of this posting to include some analysis of the curious phenomenon of stone-throwing glass-house-dwellers.
Dear Peter Woit,
I have just checked, and my trackback to hep-th/0603022 can be seen now together with Lubos Motl´s trackback.
Regarding your considerations based on number of pages and collaborators (authors), well, I do not exactly agree that this could be a much improved measure of “active researcher”, compared to other purely quantitative ones. See that Einstein´s paper on E=mc^2 was a very short one. Also, what page format are you considering? Sometimes papers are submitted in 2-column format. On the other hand, I agree that the number of authors is an interesting consideration to be made.
The issue of evaluating a researcher´s credentials or “status” is an old one. How do we evaluate things like quality, impact, relevance? People today just want to measure your work by blind numbers. But it is quite obvious that quantity does not necessarily imples in quality and the other way around is true as well.
I have gone through this several times before in my career and it is depressing. The system engulfs us. I never got interested in numbers, but in quality. But people are not interested in this philosophy or, if they are, cannot properly measure it. I am happy that I still can do my research despite the fact that I am not really in the best of worlds for such an activity.
There’s obviously no good numerical criterion to evaluate whether someone is performing worthwhile research, the one in my posting was chosen purely to make a point. One could make all sorts of different points using different metrics. For the record, I certainly think Jacques qualifies as an active researcher, by any sensible definition of the term. The problem is with his insistence on the concept of some other, non-dictionary definition of “active researcher”. He clearly was choosing this so as to exclude me, didn’t seem to think much about whether it might exclude him too.
Dear Peter Woit,
Ok, I understand your point. Thanks for elaborating. Sometimes I excced myself when this issue on “how to properly evaluate one´s research work” comes up. I´ll stop here, thanks.
Sorry for the typo [“exceed”]. I´m calming down now. 🙂
“There’s obviously no good numerical criterion to evaluate whether someone is performing worthwhile research”
There is a numerical formula to qualify as an endorser at the arXivs. Since, on Distler’s blog, you advocated the endorsment system on the arXivs, why use the same formula:
“The number of papers depends on the particular subject area, but has been set so that any active scientist who’s been working in her field for a few years should be able to endorse…”
I wasn’t really advocating an “endorser” system, just noting that that is the one the arXiv already uses, and there was no good argument for why they weren’t using it in this case (other than the fact that links to my blog would presumably be allowed under the “endorser” system).
As I wrote over there, I also don’t have any problem with an “active researcher” criterion, just with such a criterion that excludes active researchers.
That is the criterion they seem to be using to qualify as an endorser. Maybe it is (or should be) the same criterion to qualify to send trackbacks.
Actually, does anyone know what the current criterion for number and timing of postings to hep-th is to qualify as an endorser?
The discussion of the endorser mechanism assumed that the role of endorsers would be to endorse blogs just as they now endorse arXiv postings.
All I’m saying is that they use the same phrase “active researcher” to describe both. Maybe they mean the same criterion in both cases.
Distler was pretty explicit in refusing to give a precise definition of “active researcher”. I would think that if he had the arXiv endorser criterion in mind (whatever it is, exactly) he would have mentioned it.
They don’t say precisely what that formula is either.
But, if that’s what it is, would that be an acceptable benchmark?
Hard to say, not knowing what the formula is. But this is all besides the point. One just wants to distinguish blogs that have some interesting content from blogs run by crackpots or people who know nothing. It’s pretty obvious to me which ones have interesting content, and I think most trained physicists would agree on this. If all of the people running blogs with interesting content were arXiv endorsers, then that would be fine. I’d suspect there are a significant number of interesting blogs run by non-endorsers, so that would be a problem.
The real problem to be solved is to be able to locate all blogs that discuss a particular arxiv pre-print. This should be possible without track-back, if there is a standard way to label references to the pre-print that do not overlap with anything else that the search engines might index.
If arxiv is trying to screen for high quality, that is not possible when so many physicists host blogs, on which anyone can post comments.
“If all of the people running blogs with interesting content were arXiv endorsers, then that would be fine.”
“Interesting content” does not equal “should be linked-to from the arXivs”.
That seems to be the whole nub of the dispute.
I have a reasonably high standard for “interesting content”, thus do think it is pretty much equivalent to “should be linked-to from the arXiv”. I think the problem is more that different people will differ about what is “interesting content”.
To me “interesting content” is something I’d find worth taking the time to read if I was interested in a certain arXiv paper. I don’t see the downside here to resolving the question of what different people find interesting by being as inclusive as possible. The main danger with being overly inclusive is that you’ll end up with too long a list of trackbacks to each paper, making it hard to find those you personally are interested in. This doesn’t seem anywhere near to being a problem these days, given the small number of blogs linking to arXiv papers. If it becomes a problem in the future, it could be addressed then, at which point one would have a much better idea of what criteria might work to get rid of less interesting trackbacks.
The other problem is the one of authors objecting to all or certain trackbacks to their work. In particular, an author might reasonably object to links to a source containing comments they consider offensive or incorrect about their work. I think it would be a good idea for them to have the ability to have such links removed, or to not allow any trackbacks if they so chose.
So if all string theorists decided they didn’t want trackback links from Peter Woit appearing next to their abstracts (and were allowed to remove them), how would that differ from the current situation?
“I have a reasonably high standard for “interesting content”, thus do think it is pretty much equivalent to ‘should be linked-to from the arXiv’.”
Apparently, many arXiv authors disagree.
One way it would differ from the current situation is that links to discussions here of non-string theory papers (yes, this does happen…) would be there. It’s also true that quite a few string theorists read this blog and tell me that they find what I have to say worth listening to, even though (or especially because) it challenges what they are doing. I’d be quite happy if the situation were that some trackbacks were there to string theory papers where the author either didn’t care, or better yet, was interested in engaging in discussion of the paper on this blog, but not there to ones where the authors were offended by them. I’m really not trying to piss people off, rather trying to engage in serious scientific discussion.
Yes, certain arXiv authors do disagree about whether what I’m writing here is interesting content of a reasonably high standard, and that’s what this is all about. I’m happy to debate this with anyone, but not amused if they refuse to debate this and instead engage in ad hominem attacks on me, as “not an active researcher” or some such.
Why am I reminded of “We have always been at war with Eastasia”?
The answer to me seems clear: get rid of trackbacks. Any filtering system will eventually come to trouble, and you can always just use your favorite search engine if you’re interested in discussion of some paper.