11 September 2007

Journals are on the Mind

Anyone who has ever read this blog before knows that I'm a huge supporter of moving our Computational Linguistics journal to open access. Those who talk to me, know I'm in favor of more drastic measures, but would be content with just this one small change. I'm not here to beat a horse (living or dead), but to say that this appears to be something on many people's minds. I just got an email for voting for new members of the ACL board. I think only members get to vote, but I don't see anything that says that everyone can't view the statements. The list of candidates with their statements is here.

What I find promising is how often open access CL is mentioned! Indeed, both candidates for VP-elect say something. Ido writes:
Among other possibilities, the proposal to make CL open access, with a shorter reviewing cycle, seems worth pursuing. Additionally, we can increase the annual capacity of journal papers and include shorter ones.
And Jan writes:
Third, ACL's role is to help the widest possible dissemination of the field's results, including novel ways such as electronic and open-access publications, training workshops and summer schools (to attract excellent "new blood").
The issue comes up again for one of the Exec members. Hwee Tou says "I believe some issues of current concern to the ACL membership include the role of open access journals..."

Obviously I'm not here to tell anyone who to vote for. Indeed, since both candidates for VP mention the issue, it's not really a dividing point! But I'm very very very pleased to see that this has become something of an important front.

16 comments:

  1. Yeah, the situation about the journals in our field is really bad now. I guess we not only need open access CL, but also some new journals (at Elsevier, IEEE, ACM, Springer, ...). In many countries, they evaluate departments only based on impacted publications, so that your department may have, say, 3 heavily cited papers at hard-to-get-in ACL, but receive zero points. And when they compare it with other CS departments focusing on, say, computer graphics, you always lose. Same situation may be with your academic career - only journal papers count. It'd nice to have some impact factor for good conferences as well. Any idea how to promote this idea?

    ReplyDelete
  2. There are many reasons to shift the emphasis in our field from conferences to fast-turnaround journals, from the better reviewing that is possible in journals to easier inclusion in impact-measurement systems. There were many discussions around these issues at ACL in Prague, including with candidates for ACL office, and it is gratifying that several of them mentioned publications and open-access in their statements. Let's support them and work together to improve our scholarly communication processes.

    ReplyDelete
  3. I must be missing something here, and maybe someone can explain it to me ... sure, I think open-access would be a Good Thing, but I don't see it making much difference in the conference-oriented culture that CL currently has. There are already quite a few journals in CL and related fields, and they're not overloaded with submissions (quite the contrary). When I mention this, open-access proponents say something like "just going open-access would change the culture and encourage more people to publish in journals". I suppose stranger things have happened, but I don't see why open-access (rather than, say, including poster-sized pinup photos of authors in every issue) would be likely to encourage journal submissions.

    (If we really wanted to make our research culture more journal oriented, there are some obvious things we could do -- e.g., stop publishing conference proceedings -- but I personally wouldn't advocate this).

    ReplyDelete
  4. mark: i don't disagree -- i don't see that the move to open access would, alone, change much. i think the co-committant changes (faster turn-around, no prescribed page limitations) will. i've talked to several people who have said that they have published what is essentially an NLP paper in JMLR because they didn't want to wait on CL. this needs to change.

    ReplyDelete
  5. I've been trying to avoid contributing to this discussion until I had discussed a number of matters with the ed board and ACL Exec, but I admit I'm being worn down by what I consider is a constant missing of the point.

    Let's get something straight: slow publication and page limits are *not* intrinsic flaws of toll for access publishing. Electronic workflow (which we are in the process of implementing at CL) and electronic publication (under discussion) do not require open access.

    For me, the major issue with open access is the maintenance of production quality. Yes, there is a production cost even in open access. Some journals (check out PLOS) address this by charging authors for publication; I've seen rates between $500 and $2000 per article. Other journals (JAIR I believe is one example) abandon copy editing and rely on volunteer labour for every step in the process. As editor of CL, I'm concerned about the likely loss of quality that removing the copy edit process would introduce, especially given that we are, after all, a journal whose focus is language. Remember, I get to read every article we print, so I have a sense of quality of the texts we receive. I think we need copy editing, but I'm not sure our community is willing to pay for it. Yes, we could cover some of the costs here from the out current journal production costs. but if the journal is no longer a benefit of membership, how long before the membership base declines to the 50% or so that attend conferences?

    R

    ReplyDelete
  6. robert -- i think maybe the issue is a conflating of points, not a missing :). there are basically three issues:

    (1) should CL be open access?
    (2) should CL be available electronically?
    (3) should the volume of papers in CL be increased?

    the only argument i've heard against (1) is a cost issue. stuart (who may join in?) seems to beleive that this is not a big deal, though others feel otherwise.
    the reason we didn't address this issue is because we just don't have all the data to know what's really going on. none of the public sources say exactly how much money CL makes from different sources (memberships, library subscriptions, etc). when we were at ACL, fernando, stuart, ryan and i tried to get some of this info, but it's not easily available. because we didn't have the data, we decided not to mention this at all. but it is without a doubt the most important issue with (1), since i've never heard a convincing argument against going open access other than cost.

    i think that (1) alone is a benefit to the organization. whether it's a benefit justified by the cost is a question that we can't answer until we know the cost. (note that depending on cost, it's entirely possible to continue to employ copy-editors.) so i also don't think that (1) implies lower quality. i love getting my copy of CL in the mail, but i probably read about 0.8 articles per issue and then it gets shelved. in contrast, with JMLR, i get emails whenever new papers are published and probably read about 10% of them. what annoys me is when i'm at home, trying to work, and i have to VPN in to work so that i can access CL online. this is just lame.

    the conflation between (1) and (2) happens, i think, because of accidental correlation. there just aren't many electronic journals that aren't open access, and there aren't many open access journals that aren't available electronically. indeed, the incremental cost of making things available electronically (given our existing infrastructure with aclweb and the anthology) is likely reasonably low. IMO it just wouldn't make sense to have an open access journal that wasn't available electronically.

    again, i think that (1) and (2) alone would be very good for our field.

    which brings us to (3). i think the conflation that happens here is with (2) and (3). (2) simply enables (3). it everything is print-only, then it's basically cost-prohibitive to increase the number of papers published in CL. one of my personal beliefs is that (3) would be a good thing: people may certainly disagree. but i think that if you believe that (3) is good, then it pretty much requires (2), lest costs soar.

    in a nut-shell, my position is that each of (1), (2), (3) is good on its own. however, i think (2) is necessary for (3) to happen and (1) and (2) just naturally go together.

    the big open question for me is: what will it cost? this is information i just don't have, but must exist somewhere.

    ReplyDelete
  7. This comment has been removed by the author.

    ReplyDelete
  8. Mark: it's not a question of culture, it's a question of the capacity of the journal system. A print journal has limited capacity because of the cost of increasing issue size, which we discussed before. A closed-access journal, even electronic, imposes a lag in the distribution of published material in comparison to oour current conferences with their liberal copyright policies (If there was no lag in making the papers generally available, there would be no reason to subscribe, and the closed journal would fail). Electronic open-access maximizes throughput other things being equal, making a journal more attractive as a fast-turnaround venue. For example, my group is currently working on three papers for JMLR and we've just submitted another paper to PLoS Computational Biology. At least one of the JMLR-directed papers could also be sent to CL, but we won't do that for these reasons. We want our results out there for everyone to find as soon as possible.

    ReplyDelete
  9. Is there any time frame in place for a decision on whether to change the publishing model of CL or not? Is it even clear who has the power to make such a decision?

    The alternative to changing the model of CL would be start another journal. It worked for JMLR. All you need are a few good papers and some folks to agree to be on the editorial board.

    Is copy-editing worth the time and expense? Assuming journal articles would have the same quality as ACL papers, I don't have a problem with foregoing copy editing.

    For some amusing open-source publishing FUD, check out this comment on PRISM, which is the Partnership for Research Integrity in Science and Medicine.

    ReplyDelete
  10. Yes, I agree with Hal we should make CL journal articles freely available electronically as soon as they've been published, and I'd like to see continuous, electronic publication. But I think we should do this things because they would be valuable to the ACL community in their own right.

    The point of my previous message is that open access is usually touted as a way to change the current ACL conference-oriented culture into a journal-oriented culture, and as I said (maybe too sarcastically) I just don't see how making CL open-access would do this.

    Here's a suggestion: we create a journal where publication handled as much like a purely electronic conference proceedings as possible. For example, we could set page limits on submissions, have reviewers review several submissions in a very tight time window, and only make accept/reject publication decisions (no revise and resubmit). The goal would be to have all of the things that make conference publications attractive without the conference itself ... Anybody think this might work?

    ReplyDelete
  11. Mark: You don't need to make a journal like a conference (page limits, heavy concentrated reviewer load) to achieve fast turnaround. Biology journals prove that every week. Indeed, the whole point of a journal model of reviewing is that it distributes reviewer load more evenly around the calendar, and it allows for revision and resumission with a full history of the previous reviewing, both of which increase the quality and efficiency of reviewing, which are issues with conferences.

    ReplyDelete
  12. (Apologies for monopolizing this subtopic, but probably Fernando and I and a few other die-hards are the only people reading down so far anyway).

    I think it's very important to have publication venues for detailed, carefully presented and thoroughly reviewed material. But if we want to encourage a change in our field's publication culture, maybe we also need journals that have the properties that make conference proceedings attractive to the ACL community?

    So I guess the question is: what makes conference publications more attractive than journal publications to the ACL community? Is it just that conference publications are open-access and have a faster turn-around, as Fernando seems to imply, or are other factors at play as well? Some very popular conferences are not that fast, e.g., NIPs and some ACL conferences, and journal papers are already commonly posted on their authors' web sites as soon as they are accepted for publication, so I suspect there may be other things, in addition to speed of publication and open-access, that make conference publications more attractive.

    For example, I think conference publications are widely regarded as "less hassle"; you don't have to go through endless rounds of revise-and-resubmits, and the page limits mean that reviewers can't expect authors to dot every i and cross every t the way journal reviewers sometimes do.

    ReplyDelete
  13. Mark: To understand the current state we need to look at how we got there. Selective conferences in computer science and related areas emerged as a solution to the problem that there were too few relevant journals, and the ones that there were were extremely slow and prejudiced against new ideas (I still remeber the experience of publishing in some of them back in the 80s). Journals may have changed some since then, but now the conferences are established and splitting at the seams. For example, NIPS had around 1000 submissions this year, up from less than 500 ten years ago. I know the PC co-chairs, they did a great job in the circumstances, but there is no way that reviewing that many papers in that short a time could meet the standards of year-round reviewing and revising that you find in good journals in other fields, like biology. My argument is that electronic workflow and open-access create opportunities to revisit the balance between conferences and journals. My colleagues in other areas, like databases, are thinking along the same lines, for the same reasons. Conferences have reached their limits years ago, and we need alternatives.

    ReplyDelete
  14. In response to Bob's question about who has the power to change things: the answer is the ACL Exec, and there will be further discussion of this when I get around to assembling all the relevant information. As Fernando is fond of saying, what would you like me to stop doing ...

    As I think has been said before, right now we are in the process of putting in place an electronic workflow system for CL, and trying to start up the 'fast track' electronic publication process with MIT Press that has been on the table for a while. This has not been a high priority in the past because we simply have not had a backlog of papers in the last few years. With these two things in place, a number of the concerns with the existing system should be lessened.

    I would expect the open access question will be on the table at the next Exec meeting in January.

    R

    ReplyDelete
  15. 酒店經紀PRETTY GIRL 台北酒店經紀人 ,禮服店 酒店兼差PRETTY GIRL酒店公關 酒店小姐 彩色爆米花酒店兼職,酒店工作 彩色爆米花酒店經紀, 酒店上班,酒店工作 PRETTY GIRL酒店喝酒酒店上班 彩色爆米花台北酒店酒店小姐 PRETTY GIRL酒店上班酒店打工PRETTY GIRL酒店打工酒店經紀 彩色爆米花

    ReplyDelete
  16. Hi,
    This is an excellent and informative work, this will really helpful for me in future. I like the way you start and then conclude your thoughts. Thanks for this information .I really appreciate your work, keep it up.

    Essay Service

    ReplyDelete