Recipes for publications

ADS collections.   Since many years I collect complete bibliographies of all ADS abstracts for all my solar-physics colleagues whose work interests me, about 900 by now.  This solar author abstract collection consists of plain-text author-named files (from abbett to zweibel) that are easy to search with grep or an editor while limited to solar physics only, more effective than searching the whole ADS or Google scholar.  I expand and refresh this collection irregularly (“as my whimsy takes me”), together with its 1:1 companion bibfile collection.  Newest zips in and ready for download.  Since the 2019 ADS “classic” to “new” revamp I distill these abstracts with a fast script contributed here

From the plain-text abstract collection I also generate “solabs_author.html” abstract files with links to ADS.  The prefix “solabs_” gives link uniqueness so that googling “solabs_<name>” (from solabs_abbett to solabs_zweibel) is often a fast way of getting a compact complete solar-physicist bibliography on your screen.  My test of each bibfile for error pruning also delivers yet more compact personal publication lists with ADS bibcodes and links.

Some authors have hard-to-separate namesakes (most =author:"Rutten, R." ADS entries are not mine; googling “solabs_rutten” gives two bibliographies).  I weed out many by excluding the physics database and defining fine-tuned ADS search name strings.  These also serve for directly getting ADS info for a solar physics colleague from your terminal command line by using scripts ads-name-open and ads-name-metrics with my shortcut authorname as parameter.

Citing with ADS.   Fully automated with the above collections.  I usually find the ADS bibcode of a publication I want to cite by searching off-line in my abstract collection.  For their citing I use my citeads LaTeX commands that add ADS links to the resulting citation in the pdf.  I simply copy-paste the ADS bibcode into a \citeads command in my manuscript, that's all.  One copy-paste, done!  I often add page links so that the reader directly sees the cited figure or equation along with the reading, Wikipedia style. 

The citations and .bbl references are then produced automatically by bibtexxing with my entire concatenated bibfile collection.  BibTeX uses the first entry for a bibcode and skips duplicates from coauthors.  This way I never have to distill a specific bib file for a manuscript.

Reading with ADS.   I choose which publication to read by searching in my abstract collection.  Wherever I am I then get any publication that is available as pdf on ADS or via ADS directly on my screen by simply copy-pasting the ADS bibcode into terminal command
         acads "1991A&A...252..203R"
where script acads finds and opens the pdf file in my laptop if I had it already.  Otherwise it gets the pdf from ADS via a remote server with IP-checked library licenses, using wrapper script getads which remotely runs script adscode2tmp to put the pdf in the remote /tmp dir, then pulling that over and putting it as 1991AAp...252..203R.pdf in my local /tmp (in this A&A example the double quotes accommodate the linux-troubling ampersand in A&A which the macro changes to AAp.)  It then opens this publication in my prefered pdf viewer.  That usually appears within a few seconds.  One copy-paste, there it is! 

This works from anywhere (say a pensioner or covid home office) for any publication with silent license check, i.e., having a publisher pdf-symbol button on the ADS abstract page.  These include A&A and ApJ (my staple diet), also for their newest not-yet-public articles so that acads opens also these directly.  The alternative is to use the Lean Library browser extension, but acads is faster and requires no browser clicking whatsoever.

When I add an author name as in
         acads "1991A&A...252..203R" rutten-rene
acads stores this article by my famous namesake in directory .../lit/perauthor/rutten-rene.  The next time acads will find it there and open it yet faster.

Reading silly.   Publishers that require passing through their website I call “silly”.  These never yield a silent-check publisher pdf-symbol button on ADS, not even for open access.  Even then, and also when you are subscription-licensed, they impose a silly browser-clicking detour of pdf-button hunting through their silly website that they are sillily proud of.  My script acads does not function for them and they can not be page linked via ADS.  I then use acads with the ADS arXiv altcode found in my abstract collection to read the arXiv version, assuming its content is not too different from the published version.  For page linking via ADS I also use the arXiv altcode.

Non-arXived silly-publisher articles I have generally ignored because of the irritating effort formerly needed to get them: start vpn into a remote licensed library server, hunt laboriously via that server for the journal link there to the publisher and then through the usually bad publisher website for the desired publication and then finally for its local-license pdf button.  Since 2018 they are easier to get with the Lean Library browser extension bypassing most silly-publisher detours if your remote library offers that.

Publishing silly.   If you publish with a silly publisher you should always post a prepublication on arXiv (don't use “preprint” or “paper” anymore in these electronic times), preferably the accepted and language-corrected version after the referee iteration to ensure that the content is the same, or update an earlier one to the eventually published content, or even keep the arXiv version “living” beyond the published version through subsequent improvements.  Use the ADS-copied arXiv comment entry to specify identity or differences with the published version (as I did here and here, respectively).

Publishing in A&A.   Sharing my latest A&A lessons:
        Consumer report - Everything (submission, peer review, language editing, production by EDP) was handled competently and went smooth and pleasant.  The web interfaces are transparent and easy to use.  The A&A language guide is particularly instructive.
        Citation linking - In its readme.txt for A&A authors EDP recommends my latex citation linking to ADS since 2011.  The macros given there were for “classic” ADS but keep working.  Updated versions for “new” ADS are given in my student-report example.tex.
        Page linking - My newer ADS page linking latex command to open specific cited pages with pertinent figures, equations, etc. alongside the article viewing works well in A&A.  Example: “Fig. 3” in the third line here.
        Movie and blinker linking - In this article we added a new latex trick supported by A&A: direct links to open movie versions and blinkers of figure panels given in the figure captions.  Solar physics publications often display selected movie frames as successive image panels, or images or image cutouts side by side with location markers for special features that should be inspected comparatively.  Actual movie playing and image blinking using your cursor as self-controlled location marker are often much better options.  As usual our files for doing this are supplied as zips in the A&A online-material data base, but here we also added single-click openers that furnish movie or blink inspection modes directly while reading the article (by using latex \href links to specific application-opening files at an open site).  These also work in the arXiv version.  Example: blinkers at the end of the caption here (try to spot fibrilar black-on-black retarded correspondence in the 4-6 last one). 
        Acronym popups - One desirable latex trick in which I did <not> succeed is to add popups for acronyms that spell out their definition when you hover your cursor over them in on-screen pdf viewing.  It would be good if (in my little solar backyard), LTE, NLTE, CRD, PRD, CS, SE, CE, MHD, GOL, NLFF, FOV, LOS, ROI, SDO, AIA, HMI, EVE, SST, GST, EST, RBT, AO, MCAO, MOMFBD, SOT, PSP, SolO, AR, QS, PIL, CME, MBP, MC, EB, QSEB, RBE, RRE, PHE, RPW, MMF, MPS, KUL, AIP, IAP, IAS, IAC, ITA, IIA, LMSAL, JSOC, JPL, NRL, NSO, HAO, LASP, CFA, ADS, AA, ApJ, MNRAS, SoPh, JQSRT, LRSP, EPS, SPS, EAS, IAU, AAS, SPD, and the already nostalgic SIU, SOZOU, KIS, LPARL, ESMOC, MDI, EIT, TRACE, VTT, SVST, DOT, NST, DST, IBIS, CDC, IBM, TI, HP, DEC, VAX, VMS, IDL - (LOL), etc. would pop their meaning, better yet with a Wikipedia clicker.  I earlier used command pdftooltip in pdfcomment.sty for such popups but they were insufficiently robust and worked too variedly between different pdf viewers -  even non-functioning in macOS Preview whereas (far too) many solar physicists remain Mac-challenged.  No go so far.  Help?
        Revision marking - The A&A office showed a great trick for revision marking.  I used to do this by defining latex environments as \revisiona{...} and \revisionb{...} that boldfaced and colored new text, setting different colors for successive revisions.  They required embedding any change, large or small or tiny, in such calls, a laborious but seemingly unavoidable chore.  Deletions I marked with a command inserting a special symbol.  I would then turn these commands on for single-column double-spaced referee-style output and turn them off for journal-style output.  Subsequently I simply maintained all these calls set inactive to avoid errors from undoing them and to keep a revision change record for myself.  They then sat also inactively in the accepted-version latex files submitted to arXiv, but these can be freely downloaded whereas you and your referees may prefer to keep peer-review changes private (likewise, I do first remove all comment-line coauthor debates - sometimes very private - with script rmtexcomments).
         The great trick is to not code revisions at all but to use latexdiff for difference marking between the old and new version with a command as:
      latexdiff --flatten --encoding="ascii" --append-textcmd="abstract,revision,LEt" ms_old.tex ms_new.tex > ms_diff.tex
where option --flatten permits calling extra input files as your macros and where any command-defined environment using text inside (as A&A's \abstract{...} and the language editor's \LEt{...} insertions) should be appended as shown.  The resulting ms_diff.tex file produces a referee-style pdf with all revisions clearly marked, including deletions.  This way you do not have to embed-code any change you make: you may alter, delete and add as much as you want without further ado; you only have to keep the preceding ms_old.tex submission as reference.  A lot easier!

Publishing in Solar Physics.   Flower-casting my latest Solar Physics lessons:
        Solar Physics garland 1 - Solar Physics furnished outstanding refereeing.
        Solar Physics garland 2 - Solar Physics furnished excellent scientific editing.
Efficient and pleasant affair up to acceptance (and selection as “Editors' Choice”).  So far so good!  -  But then.........

Publishing with Springer.  Springer's production of the above Solar Physics article made me cast this dozen bitter malapropism flowers after “He’s been a long time in the firm and doesn’t like any nasturtiums cast at it” (Dorothy L. Sayers  “Murder Must Advertise”) where “he” is a manager and “the firm” a whiffle-piffle commercial publisher:
        Springer nasturtium 1  -  Springer journals, unfortunately including Solar Physics, never become public and therefore arXiv posting is a must.  Only ancient 1967 - 1996 Solar Physics articles from the more glorious pre-Springer Reidel-Kluwer era are freely available at ADS and can be page-linked via ADS.  Instead, direct (Springer-proof) page-linking of any Solar Physics article from 1997 up to the present, even when open access, requires an arXiv post.
        Springer nasturtium 2  -   Springer is a “silly” publisher and so needs arXiv posting.  Worse:  Springer permits you to post only your initial first-submission pre-peer-review manuscript on arXiv and forbids even this from the moment of acceptance (unless you pay for Springer open access and keep the copyright).  This policy clashes with the sound and commendable solar physics practice of posting only when accepted, not before. 
             However  -  soon after the acceptance email you will receive a Springer email request for contractual copyright transfer via a webform in which you “sign” a “Copyright Transfer Statement” including Springer's self-archiving stipulations.  Springer then allows you a few days to do this.  I suggest that first ArXiv posting your newest pdf with status “accepted for Solar Physics” within these days is morally and legally in order because the peer review was a nonpaid noncommercial volunteered community service intended to improve your manuscript, the resulting accepted version remains fully written and produced by you and your coauthors, you still hold the copyright, you didn't sign anything yet, and your posting then occurs before what Springer calls its “processing” and regards/claims/guards commercially as proprietary valuable.  At this stage Springer didn't do anything yet.  The Solar Physics scientific editor spent effort in peer review handling and should support and appreciate posting the resulting version.  The referee spent most effort so far and will likely dislike finding only the raw version spread instead, insultingly disavowing his/her scrutiny and improvements.  Springer's contributions so far are only its latex and bibtex macros (clumsy, I helped repair), style guide (bad), and its upload interface (the worst I have tried, unclear and requiring unnecessary and cumbersome remote compilation for the initial submission and peer-review iterations where just sending your initial or latexdiffed pdf would suffice, as with A&A).  Springer supplies these meager services entirely in its own interest to keep you from submitting traditional “manuscripts” as mailed stacks of unformatted hand- or typewritten pages that Springer would really have to typeset instead of your doing this work for free (where a better style guide would make you deliver latex closer to their quaint formats).
            Upshot  -  post at acceptance with corresponding arXiv comment.  You will be in good company: ADS shows that many Solar Physics authors including editors do this, likely in ignorance of Springer's ban of this standard practice. 
             Me too  -  I posted my accepted version before I became aware of this forbidding stipulation, unclearly formulated in a link I got at acceptance  -  but if I had been aware of the ban I would still have posted nevertheless for compliance with my institute's Plan S green open access condition for permission to publish in Solar Physics even though Springer disqualifies that through hybrid double-dipping.  In astronomy “green” open access implies the immediate presence on ADS of a full-text pdf-symbol clicker button with a green open-access dot serving the accepted version directly (side note: ADS might give arXiv pdf buttons serving a not-(yet)-accepted version an orange dot rather than a green dot).  A&A, ApJ and MNRAS are also hybrid but provide publisher pdf buttons on ADS that are green-dotted for paid-for open access or otherwise become green-dotted after 1-3 years, and they welcome arXiv posting without condition and therefore are immediately greenable by adding a green-dotted arXiv pdf button, hence permitted by my institute and page-linkable from latex (also mandatory at my institute).  Springer never yields any pdf button on ADS (“silly” publisher). 
        Springer nasturtium 3  - Springer starts its “processing” after your copyright transfer or your pledge to pay.  It consists of what Springer calls “copyediting”, meaning language and style adaptations  -  actually mainly done by the Solar Physics scientific editor  -  (which you may not copy to your arXiv post so that the publication diverges in this comma-level textual detail), and then what Springer calls “typesetting”  -  as if they still melt and pour lead to cast type to press on paper to produce proofs, whereas you yourself type 99% of the latex obeying the crummy Springer macros and check their functioning yourself by test-compiling with the crummy Springer upload interface while not getting 99% (or any) of the non-crummy Springer earnings and extraordinary profits from your work.  Nor do Donald Knuth and Leslie Lamport, the true typesetters.  Nor does the referee who after you volunteers most effort in this “bizarre triple-pay system” (“predator” publisher).
        Springer nasturtium 4  -  Springer decided automatedly and erroneously what my affiliation is and whether that institution pays for Springer open access, without asking me and regardless of the affiliations specified in my manuscript  -  then Springer obstinately kept disbelieving my immediate denial during many weeks.
        Springer nasturtium 5  -  Springer could not handle copyright transfer when its webform automation failed  -  then Springer resorted to falsification.  This pleased me because I never got nor saw nor “signed” a “Copyright Transfer Statement” and so was free to post and cast as I like.
        Springer nasturtium 6  -  Springer destroyed my 118 citation links in the proofs, while they worked with the Springer macros and survived the Springer upload processing and copy editing, giving me confidence that they would work  -  but they did not and Springer refused repair, arrogantly, still in the proof iteration when fixing was trivial.  When I protested Springer took a full month to re-affirm this refusal and so motivated these bitter casts.
        Springer nasturtium 7  -  Springer mucked up my text yet further by adding weird errors in subsequent proofs  -  luckily the Solar Physics scientific editor caught most.  Be aware that every Springer proof must be checked minutely throughout, beyond reported modifications and again including everything that was in order in a previous one.
        Springer nasturtium 8  -  Springer may not react non-automated at all when you report failures in Springer macros, interfaces, protocols, production.  Instead, Springer may “ticket” you  -  into a black hole.  I suffered both treatments and had to identify Springer programmers and managers and their email addresses to trigger response.
        Springer nasturtium 9  -  Springer needed 100 days for my acceptance-to-publication processing (instead of the advertized 17) - plus 120 emails (not counting automated).
        Springer nasturtium 10  -  Springer forbids updating your arXiv post beyond your initial submission (nasturtium 2), but nevertheless requires a final arXiv upload that specifies the eventual publication within that text (but ADS shows that most Solar Physics authors including editors do not do this).  Springer stipulates a lengthy ill-phrased formula, but essentially wants the journal reference and DOI inserted in the acknowledgements at the end, an illogical and ineffective place.  I suggest to instead replace the ugly incomplete manuscript banner with these by adapting the (obtuse) Springer style files (as I did here).
        Springer nasturtium 11  -  Springer forbids posting the final version on your institute or sponsor website (permitted explicitly by A&A and ApJ) during a one-year embargo period  -  but instead maintaining the arXiv pdf there not only during this embargo year but forever (as my institute does here) serves well to circumvent Springer mutilations of your article.  Then comment via arXiv on ADS that it is better to read the non-Springer-mutilated arXiv version (as I did here) and also in the arXiv post itself (as I did here).
        Springer nasturtium 12  -  Springer deemed their version of my article the answer to everything and therefore charged 42 Euro per download initially (their Deep Thought yielding triple my customary price at A&A).  This version remains both cheaper and better.

Publishing on arXiv.  The sad Springer saga above made me not offer my next study to Solar Physics but only post it on arXiv as Lingezicht Astrophysics Report.  Its version here is the living latest and retains an epilogue on pros and cons of selfpublishing.  I update the arXiv version irregularly.  Any reader can easily inspect all differences (or lack of difference) between any pair of arXiv versions by dowloading the source material for each and run latexdiff as shown under A&A above.  Similarly, inspection of tex and macros in my source material details my tricks including hyperlink generation.