10 September 2012

DIY typesetting and science publishing

Last week, I decided to post one of my original research papers on my blog. As expected, the post about why I blogged the paper has received far more hits than the paper itself. What I could not expect (but hoped for) was the excellent comments and discussion I received in response. Thank you, all.

I spent a few hours Friday afternoon and Saturday morning turning the paper that I self-published as a blog post into a PDF, deliberately trying to emulate what a final typeset article in a professional journal might look like.

Have a look. How close did I get?

I did all that in Microsoft Word 2010. Some writing and typographic purists who will groan at this, then tell me I should have used this or that and I would have gotten better results. Yes, I know. But the point is, this is consumer software that most people have, and the PDF is... not bad. (If there is enough interest in how I did this, let me know, and I'll write a detailed “how to” post.)

This is relevant to the matter of publishing a paper on my blog, because one of the arguments that goes around about academic publishing is whether publishers provide useful services to authors. (I contend that they do.) One of the ways that publishers provide “added value” is with professional typesetting. I still prefer reading typeset PDFs of journal articles to manuscripts.

I spent a few hours with Word and got something that is maybe 90% the quality of what professional typesetters working for academic publishers do. And that’s without much practice, although a lot of experience with Word. I could probably speed up the process considerably.

Meanwhile, I’ve been waiting for over two months for my most recent paper to move from “provisional PDF” to the final typeset form. That is not an isolated incident.

That does not make me a happy partner in scientific communication. It makes me want to say, “Pick up your game, publishers, or these first self-published papers on blogs won’t be the last.”

Additional: I received an email from Bryan Vickery at BioMed Central, indicating that there have been changes in their production process that have caused delays. They know about the problem, and are working on it. Sensibly, they focused on getting the HTML version out first, so there was something that could be read and cited. See more here.

Related posts

Good thing I’m not in a hurry

External links

The Typography of Authority — Do Fonts Affect How People Accept Information?
Analysis: When will your BMC paper be typeset?
The Glacial Pace of Scientific Publishing: Why It Hurts Everyone and What We Can Do To Fix It
The Glacial Pace of Change in Scientific Publishing

5 comments:

protohedgehog said...

Just out of interest, why even bother with type-setting? As long as the text is legible, and the data/information extractable, then what does it matter how pretty it all looks? Or is there some additional purpose that type-setting serves?

Zen Faulkes said...

Proto: Ah, such a scientific reaction: “Only the content is important!”

We could go off on a long, long, discussion of the weighting of substance and style here. Randy Olson’s Don’t Be Such a Scientist discusses this.

I like trying to make things as beautiful as I can. It’s personally satisfying to me to do so. I think other people like looking at beautiful things.

And you do know you’re asking the guy who writes a blog with the explicit mission of trying to make conference posters beautiful, right? ;)

Carl said...

Just because someone reading this will see your comment about "could have used" and ask what, someone else has to ask: "why not use LaTeX?" Free, cross-platform, and made for this.

Despite the existence of editors like lyx that make latex writing look like Word, the answer is probably the learning curve. But you might take a look at pandoc (http://johnmacfarlane.net/pandoc/) which will let you write in the increasingly popular lightweight web markup language called markdown, and generate latex pdfs automatically. Just a for-posterity post...

Unknown said...

I will second what Carl said, and add that many scientists I know are experts in LaTeX and produce gorgeous typesetting that is not improved in any way by the publishers. (In fact, when going over the galley proofs, I often have to reverse the mistakes entered by copy editors.)

To give an example, I created this document specifically to share on the public internet, and typeset it myself on LaTeX using a Tufte-inspired style:
http://dx.doi.org/10.6084/m9.figshare.92789

Isn't that beautiful?
What value would a publisher add to that? :-)

And here's a typical pre-print from my group. Again: LaTeX !!
http://arxiv.org/pdf/1108.5815v2.pdf

I typeset every document that comes out of my research group, and the journals add a little style but, really, it's minimal.

Zen Faulkes said...

Relevant to this thread is this piece from Justin Kiggins:

Why do scientists tend to prefer PDF documents over HTML when reading scientific journals?