Episode IV: The completerer

Per usual, I throw out some observation or random remembrance and then it nags at me.

I come to the realization that perhaps the kids these days actually genuinely have no idea that there is/was/can be a better way.

Like when I remind you that Science and Nature "papers" were once barely more than abstracts. With a single figure, maybe two. And that the real followup paper was in another journal. Seriously, look back in the early 70s, maybe into the 1980s. The issues are available for your perusal.

This is another example. Two cases in which the same group published (or at least prepared to publish) at least four different papers from a single project. In the first case, it looks like the same three authors were on all four, they put it in the same journal and the first authors swapped on one. In the second case, the author list was more diverse and there were three different journals. (Interestingly, report III seems to be missing. I wonder what happened there? But still, the group published several other papers around the same time and on the same rough idea- perhaps one of those was supposed to be the III article?)

This sort of thing reinforces my criticism of the way Glamour Humping has done bad things to science and careers while not really providing anything more than a sham of the "complete story" in exchange.

If you want to publish several manuscripts on a topic, with different unshared unique first-author and last-author slots, it is possible. You get to throw up far more than a single published manuscripts' limited number of figures. You can elaborate on side themes. Nothing gets hidden from view in the Supplemental Materials. And presumably the speed by which some of the story emerges in published form is enhanced. Which permits other people to see and use the information earlier.

It was possible once. It is possible again.

14 responses so far

  • At least for primate work, this was absolutely the model. A 3 or 4-parter in JNeurophys, with a summary in Science/Nature. As a graduate student, reading those papers was a bit like walking through the library stacks in search of a particular book. If you took the time to browse, serendipitous discoveries were possible.
    I don't know anyone who has slogged through the modern complement of poorly-formatted, barely proofread Supplemental Materials and felt the same way.

  • drugmonkey says:

    Heh. I actually found one of those because it was in the same issue as something else I was looking for. Or actually I think I initially clicked the wrong issue and noticed that one. Random perusal can be useful.

  • "In the past 10 years, we have conducted a systematic analysis of brain stem and cerebellar neurons that may participate in motor learning in the VOR. We present our results in three papers. "

    http://jn.physiology.org/content/72/2/928

    At least for neuroscience, maybe we can blame Zach Hall & co. for starting Neuron in 1988, with its more compressed single report format. And then along came Nature Neuroscience because NPG won't be left out, and then when Science and Nature realized that they could just stick everything in Supplemental, well, the "short" report became the sole paper of record. Nice to see Cell hop on the systems neuroscience train, even if a bit late to the party.

    I'll just leave that here as a hypothesis -- some of the greybeards may actually know for sure.

  • Established PI says:

    Ah, memories of the good old days. My favorite is the classic three-paper series in Journal of Molecular Biology in 1980 on the lambda repressor/cro genetic switch.

    http://1.usa.gov/1m9ruCB

  • Grumble says:

    "Nice to see Cell hop on the systems neuroscience train, even if a bit late to the party."

    Yeah, what's up with that? Who invited them to the party?

  • odyssey says:

    Or there's this issue of J Mol Biol from 1992:

    http://www.sciencedirect.com/science/journal/00222836/224/3

    Check out pages 733-859.

  • Microscientist says:

    The most famous of all of these is of course the papers by Watson and Crick. The last sentence of their 1953 Nature paper is essentially a scientific cliffhanger. If you wanted to know their proposed mechanism of replication, you needed to read the follow up paper in a different journal that had all the hard data.

  • Noncoding Arenay says:

    "You can elaborate on side themes. Nothing gets hidden from view in the Supplemental Materials."

    This bugs me the most. So much good science that will probably only be skimmed or not even read and hence lost on the community. I wish the SM madness would stop, but it only seems to be getting worse in many journals.

  • zb says:

    A corollary was that even labs doing important work didn't publish *everything* in Science/Nature. They published complete reports, follow-up work, and work that didn't pan out exactly as expected in other journals.

    Now work that isn't deemed "exciting" enough for SN etc languishes, waiting for something to make it exciting enough to bother publishing.

  • Noncoding Arenay says:

    "Now work that isn't deemed "exciting" enough for SN etc languishes, waiting for something to make it exciting enough to bother publishing."

    I don't think this is too common.....yet. I know some very good people that publish consistently in top journals, but every once in a while publish the remaining data in much lower ranked journals. That said, these are folks who are sincerely interested in digging deep (or "ascending vertically"...whatever terminology you prefer) into their research question(s) rather than chasing every frickin tangential lead that might look like a promising S/N/C paper.

  • Josh VW says:

    So if you read an article today in S/N that is three pages, with no supporting information, would you trust it? No NMRs/Gels/SEMs to prove that the authors really have the system that does the Fancy Thing, no details of their SpectraMaster2000 that lets them collect the data, etc? I guess I don't see that the Old Way was ever a good idea - why should I trust the short story on the promise that the evidence will come later?

  • drugmonkey says:

    The same way you ever view anything in science. With "trust" or skepticism appropriate to the evidence that has been marshaled. Why is this a problem for you?

  • qaz says:

    Josh VW - the problem is that we actually have two competing goals in science (the field, not the journal).

    One is keeping up with the people and papers in your field (or in your subfield, or in your subsubsubfield). For those papers, you need to have every detail and every control figure so that you can be sure that they didn't accidentally screw something up. But in many fields, there may be as few as a dozen people who can really understand that paper.

    The other is keeping up with science in general (or at least your more general field). There, you often don't have the wherewithal to check the details. On the other hand, you also don't want to miss the general field because there may be something in the broad world that is extremely useful. (Most breakthroughs I have observed have come from translating from one field (or one subfield) to another.)

    In the old days, there were broad field-general journals (Science, Nature) that most if not all scientists read to keep up with each other. These were not popular magazines, but they did detail the latest breakthroughs and important results at a level where a scientist could judge the general method, trusting that SpectraMaster2000's worked as promised and that the authors probably knew how to use a SpectraMaster2000 (that was the role of reviewers). And there were field-specific journals (JNeurophys, JNeurosci) that contained every last detail (some JNeurophys papers were 75 pages long). In your field, you read the field-specific journals to keep up with your field. If you found something in the field-general journal about another field that was interesting, you could go into the field-specific journal for that field to find details of that paper, but working through those details would often take a lot of extra work.

    Unfortunately, the GlamourMags are now trying to accomplish both of those goals, and thus are failing at both.

  • The Other Dave says:

    Nice answer qaz! I think you nailed it.

Leave a Reply