November 12, 2013 9:12 AM   Subscribe

Fuck yeah. Not my field but this is (at least tentatively) awesome news.

(Though I'm tempted to pronounce it "bi-orchive," as if it meant "pertaining to things with two testicles." But that's probably just me.)
posted by Now there are two. There are two _______. at 9:16 AM on November 12, 2013 [3 favorites]

The fact that it's being launched by Cold Spring Harbor Laboratories, a really legendary institution in molecular genetics, should help it find wider adoption. Most of the previous attempts to do this have been top-down, where a new journal was brought into existence by an existing publishing group (e.g. Nature Precedings); this seems to be more of a groundswell.

(Nature allows preprint posting; Science does in some cases; most Cell journals do not.)

If this forces Cell to change some of their policies I'd be quite happy about that. Cell has sort of a unique style in addition to obviously being extremely high impact, but they have been the most emphatically closed-access out of the Nat/Sci/Cell triumvirate. The director of the institute where I used to work wouldn't submit anything to a Cell Press journal on principle.

Then again, Cell has also been slower than Nat/Sci to publish results in computational and quantitative biology - or really anything that isn't Western blot, Western blot, histology, bar graph, mouse with a tumor, survival curve (last one optional). So if BioRxiv has a more quant-y constituency then they may not really notice any increased pressure.
posted by en forme de poire at 9:56 AM on November 12, 2013 [1 favorite]

If any organization could pull this off without fucking it up it'd be Cold Spring Harbor, especially if they've still got Jim Watson's plucky indifference to other people's problems along with all the cash his directorship brought, but there are big systemic issues that will make it at least really difficult and certainly really painful if it works.

Physics not only generally has a strong culture of respecting unpublished results that allows it to be lax but also have a comparatively uniform culture of unpublished results, while biologists range from very strong free cultures like in old school phage biology where an awful lot got passed along as pillow talk to very restrictively weak cultures like in some of the more terrifyingly ruthless parts of the stem cell world. Nothing of value is ever really yours until its published under your name and gets cited with you as an author, thus if authors can't trust that their results won't be abused then they won't contribute.

If this works it will also likely do a hell of a lot to shake up the ownership of journals and not necessarily in a good way. Elsevier and their ilk have been able to buy up and exploit so many journals, particularly in the biological sciences, because the business model is incredibly weak and requires active and continual sacrifices from scientific communities and generous support from departments to run independently. My understanding is that part of how arXiv has worked so well is a comparatively strong support for journals from physics departments that Cold Spring Harbor certainly can't conjure up for Biology out of thin air.
posted by Blasdelb at 10:14 AM on November 12, 2013 [2 favorites]

This will be interesting, if only to see papers we wouldn't normally see through the bigger journals. We had a tough time getting one of our computational papers through, because one of the reviewers was the author of a competing program that wasn't designed as well. We got it through, but being able to skip past that nightmare scenario will help novel work (particularly computational) see the day of light. Plus, it'll make sharing papers easier. One key will be to get major institutions and investigators to legitimize it by sharing their work on it.
posted by Blazecock Pileon at 11:14 AM on November 12, 2013 [2 favorites]

Good on CSHL, and I hope this permeates wet lab biology!

For more computational work, arXiV has already been instrumental. Echoing Blazecock Pileon's comment, the latest tool of one of the leading computational biologists has been permanently banished to arXiV because of an odd peer review spat. This particular researcher's work is well known enough that his latest work gets publicity wherever it shows up. Now biologists will have that same luxury.
posted by Llama-Lime at 12:13 PM on November 12, 2013 [2 favorites]

Hey, so I wrote a lot of the software for this. Very cool to see it getting attention in my wider internet circles.
posted by Space Coyote at 12:55 PM on November 12, 2013 [6 favorites]

It's really interesting how peer review papers over situations like this swa-mem thing Llama-Lime linked to. In some sense Heng Li definitely "overreacted" by not submitting the paper anywhere else, but it does seem like he also got screwed by a reviewer with a conflict of interest. I do really like the trend towards open peer review as shown in journals like eLife. They really give you a better sense for a paper as a necessarily imperfect and incomplete, but absolutely still valuable, scientific document. This seems to me to be opposed to the C/N/S vision of a good paper as telling a "complete story," which more and more seems like the academic equivalent of a model whose image has been Photoshopped and airbrushed into inhuman perfection -- you know it's not real, but you still want to emulate it.

The problem with moving wholly towards post-pub peer review, in my view, is that it could make it even harder for new scientists to "break in" and establish a reputation for themselves: somebody still has to decide your paper is worth even reading and evaluating in the first place, and people who have the benefit of institutional stamps of approval (Heng Li works at the Broad at Harvard/MIT, and has an impressive publication record already) will tend to seem more worthy of consideration than your average struggling faculty member from Directional State University of City. Of course, BioRxiv doesn't commit itself to post-pub peer review exclusively.
posted by en forme de poire at 1:01 PM on November 12, 2013 [1 favorite]

Echoing Blazecock Pileon's comment, the latest tool of one of the leading computational biologists has been permanently banished to arXiV because of an odd peer review spat.

There was a little bit of discussion of the Heng Li spat over at biostars. Here's a comment from the site moderator:
"My advice to anyone in the same situation, if you feel that you would be enraged by an obviously unfair decision then you need to pick a different field of work. Scientific work evaluation is subjective and opinionated and usually depends on a small number of people so you can' even expect the outcomes to even out on long term."
posted by Blazecock Pileon at 1:19 PM on November 12, 2013 [1 favorite]

I don't know if I should say this or not, but we may have had similar COI issues with a different reviewer at the same journal. Given issues of subjectivity and ownership, journals apparently picking reviewers who've authored or who are otherwise closely connected with current software options is perhaps not a good way to push the state of the art. I hope this project succeeds on that basis, alone.
posted by Blazecock Pileon at 1:39 PM on November 12, 2013

I've found that reviewer COI is depressingly common across the biological sciences, as well as simple vanity and self-promotion (how dare you not X the Y when I've published 45 papers saying that you must always X the Y, coincidentally requiring that you use my favorite tool/piece of equipment? your entire research program is invalid!).
posted by en forme de poire at 2:15 PM on November 12, 2013 [2 favorites]

« Older A weird, sexy, completely untrue version of a...   |   Places Are Made Of A Thousand Stories Newer »

This thread has been archived and is closed to new comments