[Tccc] Research without Walls

Henning Schulzrinne hgs
Tue Oct 25 16:12:18 EDT 2011


I suspect we all agree that we want our research to be widely available, with as few restrictions as possible. However, we also need infrastructure to support our publication process. At the moment, this includes archival digital libraries, with search functionality. Since there are no paper archives anymore, we want these to have long-term stability, measured in decades and possibly multiple centuries.

Google Scholar, as useful as it is, is no functional replacement for IEEE Xplore and the ACM DL. (Besides, Google could, at any moment, terminate that service, as it has with other services.) Somebody has to pay for editorial assistants for journals, for review systems, for meta data extraction, editing (for journals) and for the digital repositories.

Thus, the interesting question is how to run such a system. Unfortunately, the authors of the review boycott don't seem to offer much of a positive alternative that has the stability and functionality properties. There are a few options:

- Professional societies, with (largely) institutional access charges. This means that almost all major research institutions and their members have access to the papers, but some smaller institutions may not.  For ACM and IEEE, this is not cheap, but seems reasonable. For-profit journal prices are often not nearly as reasonable, particularly for some of the specialty medical journals.

- OpenAccess with author charges. This just shifts the burden - now the author has to pay to publish, which again is probably more of a concern for smaller institutions, individuals and institutions in LDCs.

Given that the vast majority of authors are probably from the major research institutions, this seems like a zero-sum effort - you're just shifting money from the library budget to the research budgets or a special publication budget. (The current system seems more efficient, since it's one charge for the institution, rather than hundreds of little charges for each conference and journal paper.)

- Some combination of the two. This is actually what we really have today: for example, conference paper authors pay publication charges that are in excess of costs, in addition to the DL subscription that affords zero-cost access to many researchers.

- Government - such as the NIH PubMed system. That's not a bad option, but shifts the burden to the taxpayer, however modest. It would be nice if NSF or some UN organization did this, but that's wishful thinking, not a plan.

- Foundation - some philanthropic foundation runs this. Again, nice (particularly if Microsoft or Google fund it), but somebody has to ensure long-term stability and availability.

If we agree that somebody has to pay, either author or reader (institution) or some combination, the second question is whether the current charges are reasonable. It appears that in some cases, the digital libraries are a source of revenue for the societies and certainly for the for-profit publishers, but it would be nice to have better data. If that is the case, I think it's more productive to have a discussion along the lines of:

(1) Should publications be only self-supporting or is it ok that they are a "tax" that funds our professional infrastructure or possibly private profit?

(2) Would it be more efficient if we had a single digital library for the non-medical sciences than dozens of them, all reasonably similar, but with only partially overlapping holdings?


For (1), personally, I try to submit to not-for-profit organizations, rather than pricey for-profit journals or for-profit conferences and encourage my students to do the same. This is partially self-interested: my perception is that it is a lot easier for students to get access to the ACM DL and IEEE Xplore than to the fragmented libraries maintained by the various journal publishers.

For (2), I do think this is a bit of a waste - if done correctly, I think ACM, Usenix and IEEE (plus a lot of other engineering societies) could get a much better digital library if they pooled their resources, rather than "competing" and fragmenting. To me, that's the more interesting question than who pays how, as it affects the ease of doing research and the quality of research. [See "The Darwin Economy" by H. Frank for some interesting cases where competition can be quite wasteful rather than productive.]

The best thing Google and Microsoft could do is to create the best search platform they can, make it available as a free option to all the societies/publishers, and then let the publisher worry about access control and archival. Google Scholar seems to already do this to some extent, but the coverage is unclear.

Henning

On Oct 25, 2011, at 12:22 PM, Jakob Eriksson wrote:

> Joe,
> 
> I've seen a lot of knee-jerk reactions to this initiative in response to my original post. 
> 
> There may be good arguments in both directions, but accusing pledge signers of being egoistical, when what they are trying to achieve is free access to research articles for all, seems a bit far-fetched. 
> 
> See below for a more detailed response to your points. 
> 
> On Oct 24, 2011, at 1:55 PM, Joe Touch wrote:
> 
>> I agree with others; this notably doesn't boycott submission, only peer review.
>> 
>> That has a few net effects:
>> 	(1) - they're benefiting from the reviews of others without
>> 	pulling their own weight
>> 	(2) - they won't be on program committees anymore
>> 
>> (1) is clearly irresponsible. First, they're clearly benefiting from the reviews of others but not pulling their own weight. Second, they're failing to train their own students in how to review papers, which is a key part of their education as well as preparing them for life after graduation.
> 
> I don't think this is what (1) implies. Anyone can shirk their reviewing responsibilities in this way today, without signing the pledge. Those that choose to sign the pledge clearly have something else in mind.
> 
> If I sign it, my intention would certainly be to continue carrying the same review load as I do today. I would simply reject requests from certain venues, and be more accommodating to others, as I now have more time to spare for reviewing. I would hope and expect that most of the pledge signers have a similar take on this.
> 
> With more quality reviewers available to venues with good copyright policies, these venues will undoubtedly rise in the ranks, gradually achieving the goal of the pledge. 
> 
>> (2) has repercussions as well. They no longer help promote their employer in the community (being listed on TPCs is an advertisement of sorts). They also no longer are able to impact the community through their participation in PC meetings, which include personal networking opportunities and technical direction.
> 
> Again, I don't think things have to happen this way. For example, a signer may reject an invitation to IEEE / ACM conference XXX based on their copyright policy, and instead participate in ACM/USENIX conference YY which has a better policy, or even use the spare time and availability of (highly qualified but pledged) reviewers to start a new online journal with a more appropriate policy. 
> 
>> Finally, I'm assuming that:
>> 	- Google now offers advertising for free
>> 	- Microsoft products are available for public download
>> 	- UC Berkeley no longer charges tuition
> 
> You forget that these venues do not actually produce the papers: we do. They also don't fund them: the NSF, NIH and others do. 
> 
> A colleague of mine chimed in with these statistics today: "Elsevier (publishing, not the Reed Elsevier parent company) received 2 billion EUR in revenue in 2010 and kept approximately 36% of that as profit." 
> 
> What exactly is it that Elsevier does (not their volunteering reviewers and authors), that justifies $2B/year of funding, and 720 million in annual profits? 
> 
> Jakob Eriksson
> Assistant Professor 
> University of Illinois at Chicago
> 
> 
> _______________________________________________
> IEEE Communications Society Tech. Committee on Computer Communications
> (TCCC) - for discussions on computer networking and communication.
> Tccc at lists.cs.columbia.edu
> https://lists.cs.columbia.edu/cucslists/listinfo/tccc
> 





More information about the TCCC mailing list