A subforum to comment preprints/webpages?

[arivero]

I wonder if people could be interested in a moderated forum exclusively to upload reviews and comments of preprints, with some formal rules:

-There can not be multiple threads for the same preprint or webpage. But a family of preprints could be assigned to a single thread.
-The subject of the thread should identify, clear standard and aseptic way, the preprint that is being discussed. The first post should add all the relevant info to reach it.

 

[jeff]

This is an excellent idea. I would require threads be entitled with the title of the paper, and not some provocative title that reflects any of wishful thinking, an unwillingness or inability to correctly interpret what the paper means, or any other kind of personal or political agenda not in the best interest of the forum.

 

[arivero]

Yep, the tittle of the thread should be something as
hep-th/7103007 An interesting new theorem in algebraic geometry
or, for a website, something as
""[/URL]

Then the first lines of the first entry should extend the information in
a machine friendly format, say

Authors: Jane B. Jones, John F. Jones, and Steven Q. Smith
Report-no: University of Northern Nowhere preprint UNN-MATH-96-04
Journal-ref: Obsc. Unr. Jour. 51 (1994) 87-95

As most fools are usually men of a single article, I think that by controlling uniqueness of the titles we can control duplicity. In the case of homepages and particular websites, the limit could be at the website level.

Of course particular agendas can not be controlled during the discussion, but it should be requested any comment beyond the review of the paper to be redirected towards a thread in the corresponding forum.

 

[Greg Bernhardt]

This is what I was talking about when I brought up the article system hack I could install. People could post the article or a link to the article for discussion.

 

[Moonbear]

I think this is a really cool idea. Sort of like a global journal club! The only problem is people would be limited to accessing older articles that are available online for free or to those their respective libraries have subscriptions for, which means not everyone could participate in that discussion. Even with that limitation, it's a very attractive idea, and would elevate the discussion options. It would be nice to have a place where we can get into the nitty gritty details of articles without having to stop and explain all the basics to the lay audience here (I enjoy that part too, but it would be nice to have a place on the board where it would be understood that those discussions would be advanced discussions and more basic questions would be taken to the appropriate board). At least that's what I'd like out of it if you set up something like that. Or do I have to figure out how to work paypal and become a contributor if I want to add my two cents here? :-)

 

[arivero]

This is what I was talking about when I brought up the article system hack I could install. People could post the article or a link to the article for discussion.

Greg, I missed the comment about an "article system hack".

ALso, it could be possible to do some hack to gather data from citebase, I am looking about it.

 

[arivero]

The only problem is people would be limited to accessing older articles that are available online for free or to those their respective libraries have subscriptions for, which means not everyone could participate in that discussion.

It is not a big limitation. KEK has also a good bunch of scanned preprints, previous to ArXiV. And in some cases the author self-archival rights could be invoked.

 

[ZapperZ]

I wonder if people could be interested in a moderated forum exclusively to upload reviews and comments of preprints, with some formal rules:

-There can not be multiple threads for the same preprint or webpage. But a family of preprints could be assigned to a single thread.
-The subject of the thread should identify, clear standard and aseptic way, the preprint that is being discussed. The first post should add all the relevant info to reach it.

Having read the entire string up to this point, I have one rather obvious question: What exactly is the purpose of such a forum? Unless I missed it, I don't think you or anyone has actually presented a compelling reason for having such a thing other than just to say "let's do it because we can".

I'm not against the idea, but I would much prefer discussing peer-reviewed papers, rather than preprints. The primary reason would be to try and understand it inside-out. I hate to think we would be spending time on a preprint that might end up in obscure land. I would rather put effort into something that at least as some degree of legitimacy to it.

Not only that, if we could contact an author or authors from the published paper and have him/her/them describe or participate in such a discussion, it would add to the value (not to mention, the understanding) of the paper.

Zz.

 

[arivero]

Actually the peer-reviewed paper stack is almost so huge as the preprint ones, and impact index are not very different. Of course one could say "lets discuss only PhysRevLetters", or something so, but if one opens to any peer journal, then the ArXiV follows naturally. Also, think that a big percent of ArXiV preprints are peer reviewed articles, but thanks to the preprint format they are freely available.

The utility of the subforum could be just the one you are asking for, namely to give a view of the utility/impact/validity etc of the paper. In mathematics, I have found sometimes useful the "Mathematical Reviews" collection, where some random researcher comments about the work of another.

Ultimately, some interfaces could fuse the subforum with the ArXiV browsing, just as actually Citebase and Spires are. The thread voting tools could be used, too.

 

[ZapperZ]

Actually the peer-reviewed paper stack is almost so huge as the preprint ones, and impact index are not very different.

I'm not sure how you can get away with saying that since there are no citation index for arxiv preprints, whereas for there are for most of the peer-reviewed journals.

Of course one could say "lets discuss only PhysRevLetters", or something so, but if one opens to any peer journal, then the ArXiV follows naturally. Also, think that a big percent of ArXiV preprints are peer reviewed articles, but thanks to the preprint format they are freely available.

Then let's deal with only arxiv papers that have ALREADY been accepted for publication. I'm not asking that we find an actual, journal-formatted pdf version of a paper, only that it will appear/has appeared in a peer-reviewed manner. That's all. Most, if not all, of the arxiv papers that I cited on here have already appeared in journals. I have no qualms in using the arxiv version.

Zz.

 

[arivero]

A compromise solution: if the subforum were finally created, we could simply to ask for publication date in the subject line. So in this way both possibilities can be managed, and the subjet let's you to filter the desired content.


Moderation should check that publication is true, as well as wipe away avoid postings not referring to the papers. (Greg, for moderation, I -and some other volunteers, surely- could help on it if you are overloaded).

 

[ZapperZ]

A compromise solution: if the subforum were finally created, we could simply to ask for publication date in the subject line. So in this way both possibilities can be managed, and the subjet let's you to filter the desired content.

I could be wrong, but if my guess is correct, you are still thinking of including preprints that have not been accepted for publication. May I ask why? Including such things should commensurate with the purpose of having the subforum of which, unless I've missed it, no one has actually explained. If all we want to do is study the content of the paper inside-out, then may I ask why you or anyone would want to spend such effort on something that hasn't been refereed by the experts in the field that the paper is in? I mean, it's not as if we are short of peer-reviewed paper to select from - there's a gazillion of them out there. So I don't understand why we need to select from ones that have not gone through the rigorious peer-review process.

Moderation should check that publication is true, as well as wipe away avoid postings not referring to the papers. (Greg, for moderation, I -and some other volunteers, surely- could help on it if you are overloaded).

You mean someone here can become a surrogate referee to judge if that publication is "true"? As someone who is a referee for several journals, even *I* sometime can't judge if a paper is true. That isn't what refereeing is supposed to be. (How do you determine an experimental result is true short of repeating it yourself?) We could save ourself a lot of grief and let the journals do this part of the process by sticking only to the published papers.

Zz.

 

[arivero]

I mean literally, "publication is true". The fact of publication. The referee should go to the databases and check the accuracy of the reference, that is all.

As for the fact of preprints not published, I think that some physics happens there. In conference announcements or posters, for instance. It makes sense to comment on it, and I see more benefits that loses in this.

I do not see a "journal club" as a study seminar, but more as an awareness seminar. At least I remember to go to a lot of seminars I was not interested on studying, but I was interested on knowing about. In this focus, preprints are included more naturally. Note that a lot of seminars in our centers refer also to unpublished work. The key is simply to avoid redundancies.

 

[ZapperZ]

I mean literally, "publication is true". The fact of publication. The referee should go to the databases and check the accuracy of the reference, that is all.

As for the fact of preprints not published, I think that some physics happens there. In conference announcements or posters, for instance. It makes sense to comment on it, and I see more benefits that loses in this.

I do not see a "journal club" as a study seminar, but more as an awareness seminar. At least I remember to go to a lot of seminars I was not interested on studying, but I was interested on knowing about. In this focus, preprints are included more naturally. Note that a lot of seminars in our centers refer also to unpublished work. The key is simply to avoid redundancies.

We may have language problem, because I could not quite comprehend half of the things you said here.

You want to have the "referee" on here, in this Physics Forum, to double check on JUST the references alone? And this would determine if the publication is "true"? I don't understand!

Is there any physics that "happens" only in the preprints but not in published papers? My argument is that if any preprint is reporting something that is worth something, it WILL end up in a peer-review journal. If a preprint has lingered in limbo without being published, I certainly would look at it with suspicion. Wouldn't you?

If your primary reason for starting this was to inform on cutting-edge, news-as-it-develops type of physics, then why even start a forum to comment on preprints? Why not just subscribe to all the latest physics news updates, etc? And I do attend a lot of seminars and colloquiums where the presenters were presenting new results and work that have not been published. However, unless I have a vested interest in any of them, or if I personally feel that these things are clearly going to appear in a publication later on, I tend to wait till those things finally do appear in print before I put much emphasis on them. We must be VERY careful in putting too much brouhaha on something that are very new, especially if it is highly controversial, contentious, and unverfied. The Quacks hovering around sites such as this will latch on to the flimsiests "evidence" without understanding what they are (re: Theory Development section). I hate to think we are giving them more fuel to feed their ignorance.

Zz.

 

[arivero]

It is a language problem. I mean: to check if the fact that the article has been actually published is true.

As for the preprint/published point, I was thinking on the spirit of seminars. For an instance, I have been lately very worried about the 68 GeV charged scalar reported a couple years ago, and that finally was forgotten due to additional statistics, so the preprints remained unpublished. I had wished to heard about how relevant it was, and how the statistic model depends on the physical model. On other hand I am surprised about how many papers are published I had considered irrelevant. So I was not willing to take the publication factor as the main relevant point. Instead, I consider valuable that someone has taken the work to read and comment the paper. As I said first, a simple request "lets read this paper" should not be acceptable start for a thread.

I have been noticing the Quaks, yep. Except for one or two mathematically instructed, they hide when you turn on the light.

My primary reason? I think it is awareness, yes, but not necessarily cutting edge. I agree that a factor stressing the importance of a work is that someone has taken the duty of actually doing all the referee fighting job. But this factor is subject very heavily to non scientific reasons (career path, etc).

Anyway, ok, I can agree to the "published" prerrequisite. But keeping clear that it is not enough cause to start a thread: a review of the paper should be the only valid start.


--------
[EDITED] hmm perhaps I did not say explicitly this before. But it is the thing I understand for "uploading a review or a comment".

Even a more strict requeriment could be to request the initial reviewer to proof that he is not the author of the article.