Planck Publications

I just noticed that a Special Issue of Astronomy and Astrophysics which contains the early science papers from Planck has now finally appeared, swelling a considerable number of personal bibliographies just in time for the next round of grant and/or job applications!

The thing is, though, that these papers were all placed on the arXiv in January 2011, so it has taken almost 11 months for them to get officially published. Such a delay seems ridiculous to me in this digital age.  I wonder why it took A&A  so long to publish these papers? Were they all held up by refereeing delays? Are the final published versions significantly different from the arXiv version? I’ve only looked at a few, and can’t see any major changes.

Or maybe this is all normal for A&A?

If you know, please tell…

Of course the main science results from Planck won’t be out until 2013. I wonder how long they’ll take to referee?

20 Responses to “Planck Publications”

  1. “Were they all held up by refereeing delays?” – I guess if it’s a special issue with only the Planck early publications, then it takes only one or two slow refereeing processes to set back the whole lot. Also “refereeing delays” makes it sound as though the referees are to blame — perhaps it was delays in adequately revising the papers!

    • telescoper Says:

      Yes, I understand that the whole set might have been held back because of one paper – I can see why they wanted to publish them all as a set. But I wonder which one it might have been?

  2. hde226868 Says:

    I just went through all papers and checked the acceptance dates. Most were accepted in May/June, so 5 months after submission, but three were accepted much later. The latest is paper XXIV, which was only accepted on 22 September. Paper II was accepted end of August only. Add to this a couple of weeks of arguing with A&A’s language editors means that production started in mid October. Given that typesetting etc. also takes some time and that A&A will want to have a small buffer, 6-8 weeks for production is not too bad.

    Four to Five months between submission and acceptance means that most of the papers have have gone through at least one iteration: let’s assume the referee took two months, which is on the long side for A&A (for my papers it has rarely taken more than 6 weeks). Add in 2-3 weeks for revisions, and another 2-3 weeks for the collaboration to approve the revisions, and one iteration takes three months or so. Then another 2-3 weeks for the referee to approve, and getting something that was submitted in January accepted end of May seems realistic. So if two to three papers required more than one iteration, getting the articles out in December is still fairly quick. I don’t think that MNRAS or ApJ would have been able to do it faster.

    Lazy referees might also be an issue, but A&A is usually very good with getting reports in on time and, when it looks that something doesn’t work out, asking a 2nd referee.

    I think the real problem is that Planck seems to have decided that putting papers on arXiv before they’re accepted is a good thing. I tend to think that this is only justifiable if there’s clear competition on the horizon, which for Planck is not the case. I guess, politics won here (“we published in January and it was A&A’s fault that the papers only appeared so late”), while in reality the papers had never been peer reviewed before they were put on arXiv.

    • telescoper Says:

      Well done! The joys of having a blog.

      I was hoping that someone would have gone through the whole set. I agree that September- December is not an unreasonable delay, but of course one still doesn’t know what the issue was (without further investigation).

      I think papers should go on the arXiv first, as that is clearly the most important avenue for dissemination.

    • Anton Garrett Says:

      “I think the real problem is that Planck seems to have decided that putting papers on arXiv before they’re accepted is a good thing.”

      I’d hate to have it otherwise. ArXiv is scientists’ only real way of exerting pressure on publishers and learned societies that might end the present ripoff and sluggishness. In the days when I strove to publish in refereed journals, I regarded one year as pretty quick.

  3. Daniel Mortlock Says:

    I think an even more important reason for these particular papers being put on the arXiv upon submission was to make the associated data available to the community immediately, rather than waiting ~11 months for them to be published. I agree the arXiv “diff” would be an interesting exercise, but for data-presentation papers I think the notion of refereeing is far less important than it is for papers with more interpretation.

  4. telescoper Says:

    For the record I should point out that I’ve never actually published a paper in A&A so have no experience of it being either fast or slow relative to other journals.

    • hde226868 Says:

      My publications are more or less evenly distributed over ApJ, MNRAS, and A&A. I find the process with all three journals fairly comparable. A&A’s language editing tends to be anally retentive, but at least the language editing is separate from the typesetting, which is based on your final submitted TeX. When ApJ was still with University of Chicago press, I had cases with ApJ where they just changed the meaning and didn’t even remark on their changes when sending the page proofs… In terms of paper preparation, A&A’s TeX templates are the nicest and closest to proper LaTex standards, while MNRAS’ is the worst.

      In terms of refereeing I have been a referee for all three journals and I’ve been on the “receiving side” as well. I don’t see any difference in the quality of the refereeing process or in the turnaround times, and for all practical purposes consider these three journals to be equal.

      • Nice summary. Some questions:

        What determines which journal you submit a paper to?

        How often does it happen (to you or others) that a paper rejected by one journal is published by another, perhaps even without major changes?

        Is Astronomische Nachrichten on many people’s radar? If it is on yours, how would you compare it to the Big Three?

        How would you compare PASA and PASP to the Big Three?

        A&A actually typesets from one’s submitted manuscript, while MNRAS doesn’t, right? Why the difference in strategy? What does ApJ do?

        MNRAS seems to be one of the few journals in astronomy (along with PASA; not sure about Astronomische Nachrichten which leaves copyright with the author. In practice, this might not be a big deal: MNRAS still requires an exclusive license, and those where the copyright is retained by the journal, or the publisher, or some professional society, or ESO (A&A) do grant the author some non-exclusive rights. Still, I’m wondering what the rationale for the differences in policy is.

        What are typical times between a) submission and receipt to the author of referee’s report and b) receipt of the final version to the journal and publication? (I’m assuming that the time between the receipt of the referee’s report and acceptance of a revised version is dominated by the time the author needs to make the revisions, at least for revisions which aren’t that large, i.e. the paper is essentially accepted pending revisions, rather than asking for a revised version to be considered essentially from the start.)

  5. I was lead author on a paper in the Herschel special issue of A&A last year; it took just over three months to go from submission to online publication. One of the papers I was a co-author on was reviewed within two days of submission. Seems very strange that they took almost a year in total for the Planck papers, especially when there were only 26 of them, compared to 152 Herschel letters.

  6. Andrew Liddle Says:

    I see some relevant stats as a MNRAS Editor. In 2010, the median time between submission and acceptance for a MNRAS main journal article was 94 days, whereas for ApJ main journal it was 134 days (stats based on a random subsampling of articles; quartiles were also quoted). For Letters they were similar, MNRAS just slightly faster with 56 days versus 66. No stats for A&A unfortunately, and this doesn’t include the time from acceptance to actual publication which I believe is pretty similar in all cases. The managing editors of ApJ, MNRAS and A&A meet regularly so a lot of their policies are in common.

    Philip: MNRAS does typeset from the authors’ files; I think they all do.

    On copyright, I think MNRAS’s aim may simply be to try and reduce having to handle permission requests which take up admin time for no particular gain to anyone. For the same reason the journal now lets a certain number of figures per issue be reproduced without permission being required (provided they are properly credited of course).

    I also believe the papers should go on the arXiv as early as possible, so the community gets the earliest benefit from the project’s results. There were definitely some improvements introduced into the Planck papers I was involved in through the referees’ comments, but not sufficiently to justify an 11 month delay in making the results available.


    • “In 2010, the median time between submission and acceptance for a MNRAS main journal article was 94 days”/I>

      Either things have sped up quite a bit since 2010 or I am a 5-sigma outlier. (Does that mean that I am probably not real?) More details tomorrow on arXiv and in a new comment on an old thread here In The Dark (extra points for guessing which one).

    • “In 2010, the median time between submission and acceptance for a MNRAS main journal article was 94 days”

      Either things have sped up quite a bit since 2010 or I am a 5-sigma outlier. (Does that mean that I am probably not real?) More details tomorrow on arXiv and in a new comment on an old thread here In The Dark (extra points for guessing which one).

    • “Philip: MNRAS does typeset from the authors’ files; I think they all do.”

      I just got this in an email with regard to MNRAS proofs:

      Please do not supply corrections in the form of an updated version of your original LaTeX file. LaTeX is only used in order to copy-edit the original paper; it is not used to generate any further versions of the paper. Instead, please supply the corrections in such a way that clearly shows (or lists) exactly what changes you would like the Production Editor to make to your proofs.

      I guess it depends on what one means by “typeset from the authors’ files”; the above confirms my suspicion that MNRAS does not actually use the LaTeX supplied by the author to produce the article in the journal, where I think A&A does. (MNRAS also accepts submissions in non-LaTeX formats.)

  7. Pierre Maxted Says:

    The other reason to post a paper on arXiv on submission is to give anyone who might be interested a chance to send comments to the author that can be used to improve the paper. I have been on both sides of this process recently and have welcomed the chance to improve other peoples papers and been grateful for people’s input to my papers. In fact, it seems a shame not always have papers open to public review for some time prior to publication – perhaps this might improve the “signal-to-noise” of the typical journal paper.

    • pierre – i absolutely agree – requiring all papers to go to the arXiv before final acceptance would hopefully improve the general quality.

      i was going to highlight the quality of some of the A&A papers i’ve seen recently (and by implication their reviews) – until i remembered all the equally useless reviews i’ve had back from MNRAS/ApJ referees… so i think there is no significant quality variation between the journals in terms of their reviewing.

      finally, i’ve been struck by the very fast accepted-to-publication time for ApJ in the last 6 months (since they introduced article IDs, rather than page numbers). it seems significantly faster than MNRAS.

  8. George Efstathiou Says:

    I co-chair the Planck Editorial Board, which oversees the
    preparation and internal refereeing of Planck papers. It
    turned out to be a major effort to coordinate the production
    of all of these papers for release on astroph at the same time
    as the public release of the Planck Early Release Compact
    Source Catalogue. Compromises had to be made and the
    papers were, inevitably, of variable quality. The papers
    which required the most extensive revisions held up the
    rest. Although the Early Release Catalogue comes with an
    extensive Explanatory Supplement, the first wave of Planck
    papers cover additional essential material such as the
    performance of the instruments/cooling system in flight
    and the data processing.

  9. Walter Blackstock Says:

    Perhaps worth mentioning is that the Planck A&A papers are accessible to the ‘scholarly dispossesed’ (ie retired). I’m not of your discipline, but enjoyed reading about the engineering and technical challenges. Technically more-or-less flawless. Not least, meeting this engineering challenge provides some answer to the current obsession with economic relevance.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: