Every so often I see publishers and the ACM complain about open access being expensive and needing difficult planning to create sustainable open access models. And every time I see it, it seems like so much hogwash.
As far as I can tell, the only costs they pay are, well, for publication, which today basically means file hosting. Depending on how many graphs or other images a paper has, it might be a few megabytes large. However, many papers are just measured in kilobytes. The entire corpus of research papers published in ACM venues must weigh in at just a few gigabytes per year. The ACM was founded in 1947, so the back catalog only goes back 73 years. Supposing their catalog takes an average of 10GB per year, the entire catalog could fit on a single 1TB drive.
Obviously hard drives aren’t the only cost of secure long-term hosting and archival. They need redundant, resilient storage in multiple locations, for instance, and they need servers, electricity, and bandwidth. However, the ACM is asking authors to pay hundreds of dollars per paper for gold open access when it is an option. That is far beyond the cost of archival and hosting.
What other costs are there? The reviewers and conference organizers are all volunteers. Researchers do their own typesetting. Often the paper that is actually used is the preprint that the publisher hasn’t touched in any way that is hosted by the author or by arXiv. Speaking of which, arXiv seems to be able to operate on a tiny fraction of the budget the ACM and other publishers claim to need.
What other costs am I missing, besides monumental profits to publishers? Do these other costs provide real value to society and to researchers? Do these costs matter in computer science, or are they limited to certain fields?
If I’m missing something I’d really like to understand it. But it seems like open access for all CS research is not only the obviously right thing to do, but that it would be inexpensive and easy to accomplish as well.
If anyone with influence at the ACM reads this, will you please publish a breakdown of the costs of publishing, rather than just repeating the mantra of “sustainability”. If anyone with insider knowledge of publishing reads this and has answers to my questions, please email me. I’d really like to know the answers.