4
$\begingroup$

While procrastinating over mathematical papers on Arxiv, I came across odd dates of the papers.

That is,

Ricci flow with surgery on three-manifolds, Grisha Perelman, February 1, 2008

The entropy formula for the Ricci flow and its geometric applications, Grisha Perelman, February 1, 2008

Who made the edit to the dates? Or, is there a different administrative task being done by moderators from time to time?

Thanks. I will delete this post if it is seriously off-topic.

  • 0
    I'm not sure if this is on-topic, but I am unable to confirm what you're seeing: [entropy-formula](http://arxiv.org/abs/math/0211159), [surgery](http://arxiv.org/abs/math/0303109). Could you please provide a link?2011-06-10
  • 0
    For clarification @Theo: the OP is talking about the dates written in the papers. The arxiv version have not been edited and are [v1] Mon, 10 Mar 2003 16:44:35 GMT and [v1] Mon, 11 Nov 2002 16:11:49 GMT, respectively.2011-06-10
  • 0
    @Rasmus, yes that was what I meant. One of the papers stated Jan 1, 2011. But now, upon rechecking, I found it to be 2008. Something is happening or has happened.2011-06-10
  • 0
    In the source code the date is not specified at all (no `\today` or anything). I don't know what `\maketitle` does in this case. That the compilation dates were 10 Mar 2003 and 11 Nov 2002 is confirmed by the vertical arxiv signature on the first sides of the papers.2011-06-10
  • 0
    The paper hasn't been edited. It appears you're reading the date the .tex file was last compiled.2011-06-10
  • 0
    @Ryan, that must really be the reason. There must have been a warning somewhere of this issue at least in the FAQ.2011-06-10
  • 0
    My doubts are swept now. Should I delete the question?2011-06-10
  • 0
    In general, there is no reason to delete a question just because it has been answered. On the contrary, you want to keep the answer available for the public.2011-06-10
  • 0
    @Rasmus, my question being off-topic as it is, I thought of deleting it for its inappropriateness for Math.SE.2011-06-10
  • 0
    But why did you post it if you consider it off-topic?2011-06-10
  • 0
    @Rasmus, because it is highly probable to get an answer on Math.SE and MO with in 11 minutes( Scott Aronson's 11 min principle).2011-06-11
  • 0
    @Rasmus, what was worth, actually, was your answer. Thank you!2011-06-13

1 Answers 1

8

The paper was uploaded as TeX to the ArXiV, and did not specify a date. Therefore the date appearing (Feb 1, 2008) is when the source was converted to PDF (or PS or whatever format you are viewing).

If you download the source and compile it yourself, you should see today's date.

  • 0
    (This incidentally is one reason why digital preservation is difficult; ArXiV correctly maintained the source, but didn't understand that the PDF could not stably be regenerated from the source, and so must also be archived as "original material" not simply a derived format.)2011-06-10
  • 0
    Is not specifying a date the same as specifying `date=\today`? Or: why did tex put a date even though it was not specified?2011-06-10
  • 0
    thanks. That has to be the reason. But I have not compiled anything. But I am pretty sure that I saw it even in 2008 and it used to agree with the date on the vertical signature.2011-06-10
  • 0
    @Chulumba: ArXiV is a federation of servers. Perhaps one of them deleted the PDFs to make room, or perhaps no one asked it for the PDF until 2008 (perhaps because it didn't exist until 2008). I see Feb 1, 2008 when I download form the ArXiV, apparently some people are seeing other dates.2011-06-10
  • 0
    @Rasmus: yes, the article document class just provides \today as a default. In my long illustrious publishing career, the editor has always (twice) provided all the dates for me.2011-06-10