Gold Open Access publication: Is now the time?

While scientific papers were once available only to those willing to pay expensive fees to journal publishers, papers are now increasingly made available for free, as they enjoy some form of open access (OA)Not all forms of open access are the same, however. While the ACM SIGPLAN Executive Committee (of which I am the Chair) is generally happy with the OA rights SIGPLAN[ref]ACM SIGPLAN is the Association of Computing Machinery‘s Special Interest Group on Programming Languages.[/ref] authors currently enjoy, it may be time to push for even stronger rights. Reasonable people may disagree about the costs and benefits of such rights. As such, we would like your feedback (even if you are not a SIGPLAN member).

Please consider filling out this open access survey. It should take 5-10 minutes.

The remainder of this blog post discusses the issues in more depth. We invite your feedback!

Status Quo: Free OpenTOC and Green OA

We who often publish via outlets of ACM SIGPLAN enjoy OA right now. In particular, all SIGPLAN papers hosted by ACM’s Digital Library are free but only via certain “table of contents” (TOC) links collected at This benefit is called OpenTOC. SIGPLAN authors may also re-post their papers on their own site or other non-profit sites like arXiv and ePrint. This re-posting right is called Green OA. These benefits come at no added cost to authors or conference attendees.

For example, consider the 2015 OOPSLA paper I co-authored, Incremental Computation with Names. This article is hosted by the ACM DL at this link. Without authorization, one must pay to access the PDF. However, as an author I am permitted to post the paper on arXiv (a right of Green OA), so the paper is also hosted at this link and is freely accessible. Moreover, by virtue of OpenTOC support, readers can access the DL version of the article for free via the compilation of links at the OpenTOC. The free OpenTOC is a nice addition to the standard Green OA rights. It ensures that even if authors do not self-archive their paper, a free version will remain available indefinitely as part of the ACM DL.


Despite their clear benefits, Green OA and OpenTOC have some downsides.

First, even though free links to papers may exist, it is not certain that prospective readers will find them. Search engines may return the paywall link first, and/or may not always find the free link. For example, when I use Google to search for the OOPSLA 2015 paper, Automated backward error analysis for numerical code, the first hit is for the paywall link. That said, in my experience, when authors self-archive, it is easy to find their papers. In the above example, the authors’ self-hosted copy of the paper is the second search result (for my OOPSLA paper, it’s actually the first result). OpenTOC links are a new feature, so it is unclear where they will ultimately appear in search results, but they are likely to be lower than the above two.

Second, while authors may retain copyright of their work, they are not permitted re-post on ad-supported or for-profit sites, such as ResearchGate or Non-authors may not re-post papers anywhere, e.g., to their own site when constructing a reading list for a class.

Gold OA: More free, but worth the price?

For those authors unsatisfied with Green OA and Open TOC, they may pay to publish their paper as Gold OA. With this option, the paper is published under a Creative Commons license, allowing anyone virtually unlimited rights to the paper. For example, authors can post their paper to ad-supported sites, and non-authors can re-post the paper as well. Moreover, the author grants ACM non-exclusive publication rights, and the authoritative DL link to the publication is free to anyone.

All ACM author options are described here in detail.

Currently, authors may elect this option by paying $700 (or $900, for non-ACM or SIGPLAN members). Few do, presumably because the cost is not worth the added benefits over Green OA and OpenTOC.

However, conferences now have the option of making all of their accepted papers free to anyone via ACM’s Digital Library, at the Gold level of OA. This option comes at the cost of $400 per paper, but only if all papers go gold. So the question is: Is this option worth the price?

How to pay?

The answer may depend on how you pay.

One option is that the conference pays the cost and increases registration fees to compensate. For a large SIGPLAN conference like POPL or PLDI, which has around 500 attendees and 50 papers, the per-registration cost would increase about $40. For smaller conferences, it would be slightly more. In some sense, the added benefit of Gold OA is felt most by non-authors, since they can now re-post and reuse published papers more freely. As such, it seems defensible that conference attendees should pay.

Alternatively, authors could pay the $400 fee. If a grant or other (e.g., corporate) funding support is acknowledged in the paper, the authors would be expected to pay. Some countries, e.g., the UK, specially allocate open access fees to grants. This option makes sense because funding agencies and authors clearly benefit from greater open access because their work will be more widely read. That said, we should not be asking authors to pay to publish. If they do not have funding support they may have the fee waived and covered by the conference.

Finally, authors and conference attendees could share the costs, recognizing that both authors and readers benefit from Gold OA.

Make your voice heard

Which of the above options makes the most sense to you? Please take 5-10 minutes to fill out our survey. We are interested to know your feelings, generally, about open access. We are particularly interested to know at what price Gold OA is worth it, and how the inevitable fees are paid.




Filed under Policy, Process, Science, Uncategorized

15 Responses to Gold Open Access publication: Is now the time?

  1. It seems only fair that if we do this, the ACM should start providing $400 worth of value per paper.

    • Others have made similar comments in the survey. I will try to put together a blog post in the future that details what I’ve observed that ACM does for us. Off the top of my head:

      • Negotiate hotel contracts, getting us lower rates because of size/clout
      • Pay for insurance of conferences, e.g., to pay hotel contracts etc. in case of 9/11-style events, or to pay for mishaps to atttendees during the conference
      • Check over legal documents and confirm legal documents, e.g., for contractors that provide us services like registration, web services, video editing, etc.
      • Perform reimbursements (e.g., for PAC and PLMW travel grants) and related individual services
      • Manage memberships and membership services
      • Provide free CMS (HotCRP, EasyChair) and other cross-SIG services

      The list is long, and the services it includes would be hard to reproduce by starting over. In my experience, ACM leaders act in good faith. They have come a long way on OA in a short period of time, but they rightly worry about their future, and ours. Costs will almost certainly come down as the landscape becomes more clear, but the moral of my story is that ACM is probably doing more for your money than you realize.

      • Peter Sewell

        Perhaps more transparency of the ACM finances is needed to answer this question. For example, off the top of my head, most of the things you mention are presumably funded by registration fees, and “manage memberships” I would hope would be funded by membership fees. Do conference registrations subsidise publishing, or vice versa, or neither (in which case the issues are separable)? Does the ACM currently get $400 of income from publishing fees on average per paper? How much would a third-party publishing solution that provides long-term archiving and management of metadata cost? (It’s hard to imagine that that would be $400 per paper).

        • The services that ACM does for the SIGs seem to me to be more than what is covered by the 16% overhead fee it charges on conference revenue. But I don’t know that for sure, since I have no window into ACM’s finances. I agree with you that it would be great to have a look at these, but we have tried in the past with no luck. ACM also does SIG-independent things that benefit its members, such as curriculum development, chapter support, outreach, etc. These are its “good works.”

          If I am right that ACM’s expenses in serving SIGs are greater than revenues from overhead, a reasonable way forward would be for ACM to more careful about it connects revenue to expenses. E.g., it could charge higher overhead on the SIGs, but send more DL money to them, too. But the net effect to SIGPLAN would be a wash, more or less: It would raise conference registrations indirectly while lowering OA fees.

        • The last ACM annual report (see page 5 of suggests that publication revenue subsidizes the rest of the ACM, and that conferences about break even.

      • Not to put too fine a point on it, but registration fees already exist to support those things.

        Charging an *additional* 400 dollars per paper is really only justifiable if that represents additional costs related to OA — and that doesn’t pass the smell test. For comparison, the arXiv has an annual budget of about 800 thousand dollars, and currently archives about 1.2 milion papers, taking in between 8 and 10 thousand new papers per month. That implies an average archiving cost of less than a dollar per paper per year. At a 5% discount rate, you’d expect the net present cost per paper to be between 10 and 20 dollars. This figure is consistent with other archives, as well — the LIPIcs series from Dagstuhl (used by ECOOP) has an article charge of 15 euros.

        As a result, I would regard an open access charge of 40 dollars as high-but-okay, and see a 400 dollar charge as an attempt to oppose OA by trying to set an unreasonably high price for it. Open access is important enough that I’m willing to take the offer, but if the article charges don’t come down an awful lot in the next few years, then I think SIGPLAN conferences should seriously consider leaving the ACM umbrella.

        • My view is that we should not think of the OA fees as just being about archiving; it is about supporting ACM in a future in which they do not get revenue from library fees. See my response to Peter’s comment as a way to change this.

          It is easy to say we should just leave if we don’t like the cost. I fear that if we do there are a lot of things we will miss dearly that we don’t now realize. I will try to catalogue these carefully in a future communication, but even with careful effort I fear I would under-report them.

          • Peter Sewell

            I find that lack of transparency, if even SIGPLAN can’t see what’s going on, a tad disturbing for a public-service organisation. If I had a clear view of what the “good works” are and how much they each cost, I might support them, but I might not – many may be US-centric, for example. As Sam says below, restricting access to support them seems questionable.

          • Fair point. I’ll work to get this information and share it as best I can. (And let’s be careful when using statements like “restricting access” which might mislead — the free default is Green OA, so authors are able to self-archive, and most do, and OpenTOC is an avenue even when they don’t.)

          • Peter Sewell

            Thanks – I’ll be interested to see whatever you get.

            The Green OA scheme works pretty well for access to recent work by practicing researchers, who know they can google a home page or whatever – that usually works. But it is a real restriction in other cases.

      • I would appreciate the blog post you mention, but I think we should draw a distinction here between things that are about the publication of the paper (HotCRP etc, some legal issues, working with publication services, hosting the DL) and things that are not (all the other things you menion). For example, the conference organization support is very valuable — but that should be paid for with conference registration fees, not with publication charges.

        More broadly, the current ACM business model is to charge people money to access my scientific work and use the proceeds (which ACM seems to value at hundreds of dollars after the publishing costs) and use it to subsidize various other things (conferences I attend in nice locations, student travel to those conferences, dead tree magazines they send me, lobbying on my behalf, etc). I think this denying people access to academic work for this is wrong, and I hope that we can eventually stop it.

  2. Minor note: all work supported by the US National Science Foundation (awarded since Jan 2016) will have to be made publicly available (“public access”) within 12 months. (See That said, the NSF/DOE repository is still “in beta”.

  3. Dmitry Timofeev

    I am not qualified to discuss OA costs and the services ACM provides to paper authors, but personally I think that ACL Anthology ( model is ideal for third-party users. I understand that this model may be inapplicable to ACM conferences.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.