New Publication at the Open Journal of Astrophysics!

It’s two in two days because we have published another new paper at The Open Journal of Astrophysics. The title is A Beginner’s Guide to working with Astronomical Data. Here is a grab of the overlay:

You can find the arXiv version of the paper here.

The author is Markus Pössel of the Haus der Astronomie at the Max Planck Institute for Astronomy in Heidelberg (Germany). This is a long paper – 71 pages with over a hundred figures – that gives a comprehensive introduction to the various kinds of astronomical data and techniques for working with such data. I think this paper will attract a lot of interest from many different kinds of people but it will be particularly interesting to students doing undergraduate projects involving astronomical data (and their supervisors).

Another point worth noting is that there’s a small addition to the overlay for this paper, which will apply to all future papers (and retrospectively once we have worked through the back catalogue) and that is in the bottom left of the image above. It shows that the article is published with the latest form of Creative Commons License (CC-BY-4.0). It has always been our policy to publish under a CC-BY licence but Scholastica have very helpfully set up a new facility to make this explicit on each page. This is part of our efforts to ensure that we are compliant with Plan S which makes CC-BY licenses mandatory.

UPDATE: the CC-BY-4.0 license has now been applied retrospectively to all our publications.

53 Responses to “New Publication at the Open Journal of Astrophysics!”

  1. “Here is a grab of the overlay:”

    Who chooses the overlays?

  2. “It shows that the article is published with the latest form of Creative Commons License (CC-BY-4.0). It has always been our policy to publish under a CC-BY licence but Scholastica have very helpfully set up a new facility to make this explicit on each page. This is part of our efforts to ensure that we are compliant with Plan S which makes CC-BY licenses mandatory.”

    I’ve wondered about this before, but now that I have published something using the CC-BY license I’ve become more interested in it.

    Why does Plan S require this license? It seems to me that even the most restrictive CC license would be sufficient for open-access publishing.

    OK, if showing a figure from a paper in a talk or lecture is a derivative work, then surely CC-NC would be sufficient? It seems a bit strange to me for an open-access publisher to license the published works in a manner which allows commercial use.

    I would also like to understand the logic in the colour scheme on the Wikipedia creative-commons-license page.

    Of course, Plan S open access includes gold open access, where the publisher make a profit, so of course they want commercial use. But the OJA was essentially founded as a way of avoiding overblown profits, so it seems strange that the OJA wants to allow commercial use. Or is this just a case of being Plan-S compliant being more important than what one would really like to have in terms of policy?

    • telescoper Says:

      I understand there was quite an argument about the form of license required, but if you look at Scholastica’s implementation plan you will see that it assumes CC-BY 4.0 so that’s what we’ve done. I’ve got no objection at all to companies using scientific results for commercial purposes, as long as the science itself is in the public domain.

      • Phillip Helbig Says:

        “I’ve got no objection at all to companies using scientific results for commercial purposes”

        I wasn’t thinking so much about the results of the papers but rather the papers themselves. Not that it is likely to happen; I just wondered why essentially the least restrictive license was chosen. (But who knows? Maybe Jeff Coons will print out everything in the OJA, add some porn, and sell it for a million.)

        “s long as the science itself is in the public domain”

        I understand what you mean, but legal eagles will point out that “public domain” has a very specific legal meaning, at least in some countries, according to which even most open-access papers aren’t in it: https://www.templetons.com/brad/copymyths.html

      • Phillip Helbig Says:

        At the Scholastica page, it says “The journal or platform must enable authors to publish under a CC BY 4.0 license (alternatively CC BY-SA 4.0 or CC0).” The former alternative is somewhat more restrictive, the latter less restrictive. There would be no problem with arXiv since arXiv supports all of these and more. (One can also license the work itself with a CC license and grant arXiv its own license for the purposes of arXiv.)

        As long as there is no conflict with Plan S or arXiv, perhaps the others could be offered as alternatives, even if CC-BY is the default.

  3. “It’s two in two days because we have published another new paper at The Open Journal of Astrophysics.”

    You’re still counting. 😀

  4. Phillip Helbig Says:

    At arXiv, many OJA papers, including mine, have volume, issue, and article-ID numbers. Those are also reflected at ADS. (Not all ADS entries have that type of bibcode, though; perhaps only those where it is indicated at arXiv—I don’t know.) Is there any way to see this at the OJA web page? If not, will there be?

    Considering that some bibliography styles are the traditional volume, page, year (as opposed to including URLs for DOI, ADS, to the journal itself, arXiv, etc.), it would be nice if this were available for all OJA articles at the OJA page: both for quick checks without having to go to arXiv, and since not all OJA papers at arXiv have a Journal-Ref. field. This will also allow more citations to OJA papers to be tracked automatically.

    • telescoper Says:

      ADS assigns its own identifiers. Since we are online only we do not have pages or issues. Crossref is the system that ADS uses to track citations but this system really only needs the DOI.

      • Just this morning, I got an email from someone asking how to cite an OJA paper and mourning the fact that his BibTeX fields were empty. My paper is vol. 3, issue 1, article-ID 1. I got this information from you. Where did you get it from?

        ADS has its own identifiers, but they follow a very strict scheme and are based on information from elsewhere: http://adsabs.github.io/help/actions/bibcode

        While it might be a good idea to provide (links to) the DOI, ADS, arXiv, the electronic version of the paper at the journal itself, etc., I think it would be a good idea if—since they obviously exist—a reader could get the volume, issue, and article-ID from the OJA site. Again, this will get more citations to the OJA, or at least more can be registered, which should be good for everyone (except the competition).

      • telescoper Says:

        OJA includes the volume and issue in the metadata sent to Crossref. This is really just to help ADS. The only thing you really need to cite the paper is the DOI. It doesn’t matter if some Bibtex fields are empty. We don’t have page numbers for example, as do many other online journals.

      • OK. But since you have the information anyway, why not display it along with the paper, along with the license information (which probably fewer people are interested in)?

        The revolution will be a long one. At least for a while, OJA papers will be cited by papers in other journals, which might want a volume or even issue number (the article-ID can go into the PAGES field). The alternative is that they might not be cited, or that the citations might not be recognized as such. There are other journals, even some traditional ones, which are online only and have volumes and article-IDs.

        Yes, it is a good idea to mention the DOI in a citation. But it is not immediately obvious that the DOI is an OJA DOI.

        (The electronic equivalent of a citation which mentions the journal name would be the link to the article at OJA, but these are very long URLs—that’s good because one can guess the contents from the last part of the URL, but not good for a formatted bibliography.)

        Also, having this information on the OJA page allows the user to cut and paste it into the Journal-Ref field at arXiv. Many people use the Journal-Ref field to check whether a paper has been accepted.

        Until then, there is the following workaround: from the OJA, go to arXiv, then from there to ADS, where one can see the volume, issue, and article-ID (at least after the paper has been out for a while; initially, ADS has an arXiv-based bibcode.)

      • telescoper Says:

        You don’t need to use the html form of the DOI. See here

        https://library.uic.edu/help/article/1966/what-is-a-doi-and-how-do-i-use-them-in-citations

        The point is though that crossref attaches citations to the DOI. For that purpose the other numbers are irrelevant, though they might be useful for other reasons.

      • Yes, just the bare-bones DOI is sufficient. My point was that it is not obvious (to a casual reader) that it is an OJA DOI. A working link which obviously goes to the OJA is possible, but long: https://astro.theoj.org/article/11632-calculation-of-distances-in-cosmological-models-with-small-scale-inhomogeneities-and-their-use-in-observational-cosmology-a-review

        As you say, they might be useful for other reasons, so why not display them along with the article?

      • telescoper Says:

        The first reason is that it we didn’t see the point as they are unnecessary.

        The second reason is that we are using a third-party platform with limited possibilities for customization and there is no field on the overlay page to display this information (probably because of the former reason).

        One thing I could easily do is add a sentence to our acceptance emails to give the Volume etc so that authors can put it in the Journal field on arXiv.

      • “The first reason is that it we didn’t see the point as they are unnecessary.”

        I haven’t yet cited an OJA paper in a non-OJA paper. If such journals and their readers and ADS are happy with just a DOI, then that’s fine. (I don’t know whether this is the case.) Certainly I can report from the trenches that there is some confusion. 😐

        In any case, there should always be something which the arXiv Journal-Ref field treats as a valid journal reference. Yes, they have a separate field for the DOI, but there are still some publications without DOIs, so the Journal-Ref field still makes sense and is used by some readers as a quick check whether a paper has been accepted. (The presence of a DOI would indicate this as well, but the absence of a DOI would not be evidence of lack of acceptance, at least not in general.)

        “The second reason is that we are using a third-party platform with limited possibilities for customization and there is no field on the overlay page to display this information (probably because of the former reason).”

        OK, makes sense. In general, both the OJA web page and the Scholastica web pages work fine and are easy to use. As anyone with an old browser and/or slow computer and /or slow connection will notice, there is apparently a huge amount of JavaScript stuff going on. Not an issue for most users, but some careful people are suspicious of JavaScript. 😐

        “One thing I could easily do is add a sentence to our acceptance emails to give the Volume etc so that authors can put it in the Journal field on arXiv.”

        Definitely a good idea!

  5. Jonathan Thornburg Says:

    Congratulations on getting OJA running!

    But it would be nice to have a low-graphics non-Javascript variant of the website available.

    Looking at the website, I’m also struck by how difficult it is to read the each paper’s section & date on the front page — they’re white letters on an uneven light-grey background (e.g., the words “COSMOLOGY AND GALACTIC ASTROPHYSICS” and “January 07, 2020” in the OJA front-page entry for Phillip Helbig’s paper).

  6. For my recent paper, I see 2019OJAp….2E..12K in ADS. So presumably ADS is using some sort of its own numerical scheme for indexing OJA papers. inspirehep only reports the year.

    • Maybe Peter can explain it. Apparently the OJA gives volume, issue, and page numbers to ADS (or to something which ADS uses) for the construction of the bibcodes according to the usual scheme. I think that there are only three OJA papers which have these in the Journal-Ref field at arXiv, though from the ADS bibcodes one can extract the volume and article-ID (a.k.a. page number).

    • It’s a bit puzzling though. Apparently volume 1 is everything before 2019 (when it really got started), volume 2 is 2019, volume 3 is 2020, etc. So far, so good. Syksy’s paper has ID 9 instead of the expected 2, then come 2, 3, 4, 5, 7, 8, then two with “arXiv” in the bibcode, then 11 and 12 (so can we assume that the two with “arXiv” are 6 and 10?). The two newest still have “arXiv” in the bibcode, but presumably that is to be expected.

      Why the couple which are out of sequence? Why two relatively old papers with “arXiv” still in the bibcode? Neither of the latter two have a Journal-Ref field at arXiv, so maybe that explains that.

      ADS says nothing about the issue number; is there one issue per volume?

      • Sorry, not Syksy’s paper, but the following one has article-ID 9.

      • telescoper Says:

        Yes, the problem in that case is that the author did not update the field on arXiv promptly so it was assigned an article-ID out of sequence. ADS does assign its own IDs, presumably just in chronological order of receipt of the metadata.

        I am correspondence now with NASA/ADS to see if there’s a better way of getting papers indexed than relying on authors to put the metadata on arXiv.

        One possibility would be to have a direct feed to ADS from the OJA. I suspect more of our authors use ADS than, say, Web of Knowledge.

      • telescoper Says:

        This is because the order in authors add their info to arXiv and are picked up by ADS is not necessarily the same as the order in which they are published oby OJA….

      • That would explain why papers with no Journal-Ref do not have a “proper” ADS bibcode. Was it true in the past that the article-ID was determined by ADS but is now determined by the OJA, via data to CrossRef or whatever?

      • telescoper Says:

        First, Volume 1 is the `legacy’ papers published before we moved onto Scholastica. Volume 2 is 2019 (12 papers). Volume 3 is 2020, and so we mean to continue.

        I am in correspondence with arXiv to sort out what has gone wrong with the sequencing. I don’t know exactly what ADS does but I think it’s because ADS has picked up one or two papers out of order, possibly because authors have put the DOI on arXiv in timely fashion.

        I am correspondence now with NASA/ADS to see if there’s a better way of getting papers indexed than relying on authors to put the metadata on arXiv.

        The order doesn’t matter much as long is that article-ID is unique, but it would be good to have a more systematic approach.

      • Phillip Helbig Says:

        “I am correspondence now with NASA/ADS to see if there’s a better way of getting papers indexed than relying on authors to put the metadata on arXiv.”

        I’m sure that they have such a mechanism.

        Alternatively, you could tell the authors that it won’t appear on the OJA page and hence officially be accepted until they have updated the metadata at arXiv. 😐

      • telescoper Says:

        In fact I think arXiv also pulls data from DOI feeds so that could be done automatically too.

      • telescoper Says:

        I have to admit that I didn’t really do this when I was a student!

      • telescoper Says:

        OK. So I have now agreed with ADS a way to submit everything directly to them at the time of publication. When I get a moment I will try it with the two papers we have published so far in 2020.

      • Phillip Helbig Says:

        “I have to admit that I didn’t really do this when I was a student!”

        But how much time did you spend at the typewriter? 🙂

        A friend of mine typed up her husband’s thesis back in the 1960s. Their goal: one page of good copy per day!

      • Phillip Helbig Says:

        Since the OJA uses arXiv for distribution, the logical thing to do would be for the OJA, as the publisher, to update articles at arXiv with DOI and journal-reference information. There is already a complete framework for this already up and running at arXiv.

      • telescoper Says:

        Yes, it’s on my to-do list to set up a doi feed for arXiv. I know how to do it, but I need to find the time.

      • Phillip Helbig Says:

        “I am in correspondence with arXiv to sort out what has gone wrong with the sequencing.”

        As you can see at my OJA overlay page, there are two unused article IDs and two papers without article IDs. This is obviously a perfect match; the logical thing to do would be to get those two orphaned papers matched to the unused article IDs.

      • Phillip Helbig Says:

        “there are two unused article IDs and two papers without article IDs”

        Actually, there are four unassigned papers, but two of these are the two newest ones; since they both have journal references at arXiv, presumably ADS will pick them up soon. (ADS already has them, but under the provisional arXiv URL instead of the bibcode.)

        So, if the OJA can feed arXiv with DOIs and journal references, presumably ADS will pick them up automatically within a few days of publication.

      • Phillip Helbig Says:

        “set up a doi feed for arXiv”

        Since you can feed it stuff for the journal-reference field as well, this would also take care of the problem of getting the correct information to ADS.

      • “OK. So I have now agreed with ADS a way to submit everything directly to them at the time of publication. When I get a moment I will try it with the two papers we have published so far in 2020.”

        Good idea. Then perhaps give the two unassigned IDs from Volume 2 (6 and 10) to the two papers without proper journal references: (arXiv:1904.11448 and arXiv:1306.2144).

        Something else: Your own comments often appear out of chronological sequence.

      • telescoper Says:

        Um, id 10 is already assigned: it’s here

      • That assignment must have happened very recently! Good that it went to an unused ID!

      • The ADS link at the arXiv page still redirects to an ADS URL with the (usually temporary) arXiv bibcode, rather than the one you linked to.

      • telescoper Says:

        Yes, not sure why that has not been updated on arXiv.

      • The ADS update was very receent; perhaps arXiv needs another day.

      • Note that at arXiv the ADS link is of the form https://ui.adsabs.harvard.edu/YYMM/arXiv: where YYMM.nnnnn is the corresponding arXiv identifier with the dot. When the paper has the proper info at ADS, it gets a URL at ADS with the usual bicode, but the link at arXiv remains as is. If this information is not available at ADS, then there it has a URL of the form https://ui.adsabs.harvard.edu/YYYY/arXiv so you have to go to ADS; you can’t just mouseover the link at arXiv.

      • Sorry, WordPress tried to interpret the links. See the description of the various formats of the links and what changes where at http://www.multivax.de:8001/oja/ 🙂

      • Phillip Helbig Says:

        “Um, id 10 is already assigned: it’s here”

        Indeed it is. However, as noted above the link at arXiv still redirects to the (usually provisional) arXiv link at ADS. Note that the author list at the corresponding ADS page is slightly different than when using the final bibcode.

  7. So https://ui.adsabs.harvard.edu/abs/2019OJAp….2E…6M/abstract still returns a 404 at ADS. Presumably, arXiv:1904.11448, the last unassigned OJA paper, should get this number.

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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: