From what you say below, you also think
Dear Zoran, I do not disagree entirely with the idea of just looking up people that we know and trust wherever they post/publish their work. But this does not apply to a lot of work posted in ArXiv, and there is no guarantee of any sort. One would have to do the refereeing oneself if one wanted to quote it. that some measure for maintaining a level of control over what is posted in ArXiv is desirable. I refer to your statement “Thus we need to get overlay boards to referee and certify certain version of arxiv papers by usual procedures and without involvement of commercial journals. We should value author's work according to the theorems and not according to the journals where they appeared or did not. We should also make available independent criticism of papers in forms of discussions, reviews etc. independent from weather the paper has been published or not.” I entirely agree. You also wrote “I know of lots of published nonsense, and false statements in hi-impact journals, and massive works which are valuable and correct and are not in the journals. I am not talking only Perelman's preprints but far much more.” I also agree with your statement above. One can, however, ignore published nonsense, but what to do with false statements in high-impact journals if the author does not care to offer a correction or even post a note indicating that an error has unfortunately occurred? Perhaps with a dynamic ArXiv this state of affairs would most likely improve the situation, as it would make such actions less problematic. Indeed there is much valuable work that is not published. Regards, Marta Date: Sun, 4 Sep 2011 15:55:24 +0200 Subject: Re: categories: Re: science_publishers From: zskoda@gmail.com To: martabunge@hotmail.com Dear Marta, I have slight doubts about your reservations about the arxiv you posted on the category list. First of all, all versions are stamped with a date and the sources of all versions are non-removable from the main archive (though in some mirrors some of old versions may be non-available). Second, while I agree in the importance of refereeing I do not understand insistence on being published as a synonym. Einstein's papers are mainly published and unrefereed. The unpublished paper can be reviewed in detail and in public. Finally the arxiv can evolve and host a circle of boards which would place refereed by that and that board to some versions of arxiv paper. So you can have v1 v2 v3 and v4 refereed by journal xyz which exists only as an overlay of arxiv. There is nothing bad in having v5 which is improved and updated in author's opinion even over certified version 4. I know some people who are in academia and do not send to journals any more, just to some archive out of their conviction. One of them tells me, why would some board tell me what to read, I decide whom I trust and whom I will read. Once I know a name I can look up on line and find an article, he says. Indeed, I trust every paper with the author name Ofer Gabber much more than a paper with stamp Annals of Mathematics. Thus we need to get overlay boards to referee and certify certain version of arxiv papers by usual procedures and without involvement of commercial journals. We should value author's work according to the theorems and not according to the journals where they appeared or did not. We should also make available independent criticism of papers in forms of discussions, reviews etc. independent from weather the paper has been published or not. I know of lots of published nonsense, and false statements in hi-impact journals, and massive works which are valuable and correct and are not in the journals. I am not talking only Perelman's preprints but far much more. Sincerely, Zoran Skoda [For admin and other information see: http://www.mta.ca/~cat-dist/ ]