Dan Cohen has a note about an open access scholarly book project. It’s interesting because the discussion gets beyond the idealistic notion that scholarly content ought to be open access, to examining some of the challenges that work against it; perhaps most notably how administrators and colleagues will value such contributions in promotion and tenure decisions. It’s refreshing the participants found many of these issues to be more-or-less non-issues.
I published my first book with a traditional publisher, and I think it would have probably been insane for me not to as a junior faculty member without an established name. The publisher advertised and promoted the book, and they were the ones that submitted it for consideration for a book award; something I would have never considered.
However, looking beyond that, I have the technical skills to have produced that text in electronic form of a quality virtually equal to that of the publisher. All it takes is TeX and some high-quality professional fonts, along with some design sense. So maybe the next book …
This leads me to one issue that I find problematic with most of the open access publications I’ve been exposed to: their production quality is absolute crap. The book that Dan points to, for example, is typeset directly from Word, using atrocious fonts like Arial and Times and Verdana.
I realize I may be rather unique in my attention to typographic detail, but I’m hardly alone. It negatively affects the quality of a reading experience if one has to read a poorly designed and executed text. Compare, for example, this PDF from the book that Dan discusses to this one. Both are open access publications. All else equal, which would you rather read?
[note: Alf Eaton had a really cool demo of using XHTML and CSS for scientific articles linked from this page, but for some reason it's now 411. Suffice to say I'm not suggesting prioritizing PDF at the expense of HTML.]
Obviously, there are some challenges in getting from here to there: in allowing the production of open access texts to be as easy as clicking a button on Word (or OpenOffice), but the results to be of much higher-quality. Am not sure exactly what the way forward might be, but one idea is to identify, as Alf has, the core structure one might want to see in output XHTML, and to design some openly available XSLT and CSS stylesheets that can achieve this. TeX is, unfortunately, more difficult given all the installation and compiling headaches, but PDF is arguably less important to the future of the open access world than HTML.