I have a PhD in chemistry. A professor presenting unpublished work at ACS, etc., is the exception. It does happen now and then and we got scooped once because a postdoc did it. The main exception are GRCs, which are not public and where the audience is not permitted to share anything about what was presented. There, they require new material. No press allowed.
The rule might put a damper on the scores of talks by grad students that nobody attends anyway. They’re mostly just giving them because their departments require it.
No need to fundamentally change how science is done, as my people are already mostly doing it this way.
I’m in the “evidence is ambigious” camp and am really more interested in the story from the meteoritics angle anyway. But there doesn’t need to be an implied agenda for me to be posting links related to a science news story.
It’s certainly not the case in Ecology or Ornithology (including work on fossil birds). I have also seen lots of public seminars here at the Smithsonian that contained unpublished information on genetics and other field. It surprises me that the practice in chemistry might be so different, but I imagine that it is because, like medicine, there are commercial applications.
In other words, lots and lots of talks present unpublished work.
I think this shows the difficulty of enforcing an embargo on description of results in public before publication in a peer-reviewed journal. As far as I can determine, the New Yorker article and the press release by UC Berkeley (which is the same as the Science Daily article) came out on the same day, three days before the PNAS article. The Berkeley article also contains some information not contained in the PNAS article, including mention of an additional dinosaur fossil. The New Yorker article contains much more, but also presents much of the most important information about the site. Would the New Yorker be compelled to exclude all the unpublished material? In fact, I assume that most of the unpublished content, including the information about the dinosaur fossils, is included in articles that are currently in review or preparation.
No, it’s not fundamentally changing how science is done because they aren’t “doing science” when they present in front of six people to check an inconsequential box for their department. That’s even assuming they’re presenting unpublished (or often unpublishable) results. I sure as hell wasn’t allowed to.
All I can say is that what you describe seems completely bizarre to me. Grad student presentations at conferences and seminars I go to are attended by dozens of people, who are often very interested in the work. (Conference sessions are usually organized by topic as well, so senior scientists could be presenting in the same session as even an undergraduate. Of course talks by well known researchers will attract more interest, but many people will stay in the room throughout the session.) Many times I have had to cite conference abstracts for some interesting results hat have not been formally published yet. Also, I have to say that if you are not learning about new work but only research that has already been published these conferences must be exceedingly boring indeed. I have to wonder what’s the point.
The point is to network and to learn about research that I don’t actively follow. If I already read all of Alan Goldman’s papers, I’m not going to go to his talk unless I want a refresher or if I want to see the other folks he’s been binned with in that session. I can’t imagine citing something that other people can’t look up. That’s my main objection to pop science pieces that don’t cite published research. A conclusion is pretty much useless without the data. It seriously hampers the evaluation/improvement/confirmation side of science. Talks and magazine articles are great for generating hype but do little to further scientific inquiry.
Then the objective of chemistry conferences seem to be very different from the ones I attend. One of the big purposes would be to find out about new research that may not have been published yet.
While it’s preferable to cite something in print, it’s fairly routine in my field to cite (pers. obsn.) or (pers. comm.) or (in lit.).
That’s pretty much all of it, then. It’s rare for a popular magazine article to cite a particular article in a scientific journal.
Of course, that’s not the purpose of popular articles. The purpose of popular articles, to be blunt, is entertainment. Of course, some sources are more reliable about the factual nature of what they are reporting, but in no case can you expect them to adhere to the standards of scientific articles.
Are you under the impression that general-audience magazines like the New Yorker have aspirations to be scholarly journals, but just aren’t good enough? Or are you under the impression that their sole purpose is to generate “hype”?
You appear to be under both impressions, but neither is generally correct. The purpose of what you dismissively call “pop science” is to provide informative background on important matters of science to an interested general audience. Knowledge and curiosity don’t need justification, but an important societal justification for articles like the one we’re discussing is to help promulgate scientific literacy and reverse the sad phenomenon described a few years ago in a New Scientist cover story under the self-descriptive heading “Unscientific America: A Dangerous Retreat from Reason”, lamenting the growing gulf between scientific advancement and the scientific ignorance of much of the American public – for example, on climate science and some of its foundational principles.
Furthermore, articles like that can be inspirational for young people and even launch them into science careers. And these articles do cite research – at least the good ones do: not necessarily already-published, but often an interesting forthcoming paper, as this one did. They don’t have a long list of citations, but, again, they’re not academic journals and formal scholarly communication isn’t their purpose. To claim that such articles are “useless” because they don’t include a mass of peer-reviewed data or two dozen citations, and that “it seriously hampers the evaluation/improvement/confirmation side of science” is a gross, unrealistic, and rather pretentious misrepresentation of who these articles are for and what their fundamental purpose is.
The real problem with many “pop science” articles isn’t any of the things you claim. The real problem is that some of them are written by incompetent pinheads. But the good ones are competently written by knowledgeable science writers and often with direct input from pertinent researchers. They serve a valuable purpose for the lay audience and for the cause of general scientific literacy, and to claim otherwise frankly strikes me as pompous and arrogant.
In my field conference articles are peer-reviewed, are considered archival, and count for tenure, though not as much as journal articles. We also have a bunch of workshops which are only lightly reviewed and often don’t have papers associated with them. They are for industry people presenting work, and for getting feedback.
Just another datapoint.
I see you are in computer science, which is another field with significant commercial interests.
What, do you have to submit a paper for review before giving a presentation at a conference? Is the article in its entirety published in the conference proceedings before the presentation is given? If not, the presentation includes unpublished research.
In my field, all you generally have to do is to provide an abstract in order to present at a conference. Abstracts aren’t generally peer-reviewed, although I expect the conference organizers look at them to exclude any flaky ones. The abstracts are published in the conference materials, but there is no requirement to provide a paper afterward, except for organized symposiums on a particular topic in which participants have agreed to provide papers subsequently for a symposium volume. But these are often published years after the fact. They count for tenure, but of course you can’t really calculate the “impact factor” of a one-off publication.
Wouldn’t this great debate be served better in Great Debates so we lay folk can go back to enjoying what appears to be the most awesome discovery of the century so far?
Whether you like it or not, the controversy over how the results of the site were published has become part of the story. If certain posts don’t interest you, I suggest that you skip them.
In Computer Science, the better conferences are run like this:
You submit an “extended abstract” of your paper. Maybe 10+ pages depending on the conference’s limit.
The program committee reviews the extended abstracts and decides which ones to accept. For a really large conference a subset of the whole committee decides for papers in their area.
Once accepted, you then send in the full paper by a certain deadline. Feedback from the reviews by the program committee may be incorporated. (If you don’t fix something substantial, your rep takes a beating.) Traditionally, these papers were assembled into a printed conference proceedings that was handed out at the start of the conference. Now it’s all done electronically. These proceedings are fully cite-able: title, authors, pages, etc.
Technically the papers are supposed to be polished and turned into journal publications but sometimes you’re running as fast as you can to stay in one place so prepping for the next conference paper takes precedence.
Another aspect of Computer Science are “technical reports”. A department would have a series of TRs that range from drafts to nearly final version of papers. The author would turn in the master to the department, they’d print up a bunch of copies (traditionally) and mail them out on request or to people the author thinks might like a copy. Of course these are all electronic now.
So a paper would go TR -> conference paper -> journal article. Some well known papers never made it past the TR stage and are considered perfectly cite-able (if not so prestigious).
So anything presented at a conference is considered public knowledge at that point (if not before).
Certain courtesy rules apply. I’ve seen drafts of papers (e.g., as a reviewer/referee) and then found better solutions which I wrote up. But I would not release my paper in any form until the original paper had been made public in some way. And appearing in a conference proceeding definitely counted there.
From a Computer Scientist’s point of view, a wait of several years to officially find out what someone has discovered is unfamiliar. But different fields work differently. We tend to be in a rush since someone else might come up with the same idea and you lose out. No one else is digging thru that fossil bed trying to beat the other group.
One of the best known papers I did came about very quickly. I noticed that a result a friend had came up with used a technique very similar to something I had done recently. I figured that there was a general concept at work. We looked at the calender. It was ~Thursday with the next big conference submission deadline Monday. First, work it all out. Took a couple of days. Second, write it up. A long weekend ensued. It got accepted. We presented it. That’s that. We moved on to other things. No journal version. It gets cites. The Red Queen’s race pops up in a lot of places.
Yes, but it has completely taken over the thread and changed what I understand to be the original intent. Mods have chewed me out when I hijacked threads.
That’s all way more formal than any conferences in my field. I would guess that 80-90% of conference presentations present at least some material that hasn’t been published yet, and a fair amount of information never sees the light of day except in the conference abstracts.
I have to wonder what’s the point.