I would be happier if IEEE/ACM would typeset the damn articles. The International Conference of Software Engineering this year has hired a publishing company th

I would be happier if IEEE/ACM would typeset the damn articles. The International Conference of Software Engineering this year has hired a publishing company that have been sending manuscripts back for even the slightest failure to meet (their interpretation of) the ACM style guide. I've gone through the process 4 or 5 times across multiple papers. You can't charge $3000 for the benefit of putting a PDF on a website when you do absolutely nothing for the creation of it. If they did the typesetting themselves, I might (might) accept that there are some costs per article, but at this point it's obviously complete nonsense on similar old-world scales as the NYT's $41m pay wall. Putting PDFs up is a fixed cost, which, even amortized over the development of the site, is nowhere near $3000 per article over and above the revenue they are receiving from libraries and memberships. We keep getting told that there are overheads in running a journal; that open-access can't work without significant investiture. I disagree. When all you are doing is facilitating communication (academics do essentially everything for free: write, typeset, review, accept/reject, organize conferences, and new things like storing source code and data for replicability like PROMISE or SIGMOD), you are just a middleman. All that's being provided at this point is "authority" because ACM/IEEE-stamped articles count towards tenure, and the rest don't (at least in CompSci). This sort of thing is ripe for disruption: a competently written web app could annihilate the processes that are in place. The only problem is selling the authority. A company like Google or Microsoft Research taking to the right people and bringing them on-board for a rival publication/conference would do that (MSR are actually quite well-versed in doing conferences, the Foundations of Digital Games conference started with them, and that's now ACM for authority). I think most academics are totally fed up with the status quo, so I can't imagine that's all that hard, you just need the money to have a dedicated person to do all the "selling."