Should academic paper publishing embrace EPUB?

Sometime last year I was considering home improvement options to our house, I was thinking about building a large, built-in bookshelf in our upstairs study area. I always loved to see lots of books on the wall, and really enjoyed pulling down a book to have a browse on whatever subject interested me from my own personal library. But there was all this discussion regarding ebooks, and I was thinking if this ever caught on big time, then printed books would eventually go the way of the dodo – the end of their 400-year cycle of greatness was at hand, and the new way to read anything was going to be on a digital screen.

I’ve since come to my senses. I love books – the binding, the texture of fine paper, the fact that it doesn’t require a battery or power cord, and even the smell are all plusses in my book. Books have been around a long time, and they’re here to stay. Ebooks are just another channel of distribution for such content, and I believe that both have their place in the modern era.

However, for academic research papers, I think we can safely kill the paper. Particularly, I think it should all begin moving towards the EPUB format. I read a lot of academic papers in my work, and I find myself wishing that more of this stuff were published as EPUBs. In contrast to my love of books above, I think academic research would largely be much better served in a purely electronic format. It’s already going that way from the reader’s point of view, right?

Typically, when academic papers get published electronically, the format of choice is PDF. Or in earlier days, PostScript. If you’re lucky, someone had the foresight to publish their paper as HTML. The advantage of a flexible format such as HTML is that you can resize the fonts. Text can flow. It’s easier to get a clean copy of a text or data segment out of HTML than it is from PDF for quoting in one’s own paper, because copying from PDF tends to yield horrific line break issues and other artifacts on the clipboard.

PDF is, I’m sorry to say, hard to read on smaller screens. PDF expects paper, and refuses to reflow itself into smaller screen sizes such as an iPhone or Android device form factor. It barely passes on the 1024 x 768 iPad screen. Anything smaller, such as most ebook readers, is going to be unacceptable. Having to zoom in and scroll left to right to read one line of text at a time on a mobile device is not what anyone would call a user-friendly reading experience.

EPUB by contrast works great on mobile devices. Using the Stanza reader on iPhone is quite comfortable. iBooks on the iPad platform is a joy to use.

After reading this tweet by Dave Gutelius today, I was reminded of how much I hate printing out all my academic papers in preparation for travel. Flying is reading time, and printing this stuff out and stuffing it in my backpack is time consuming, a waste of paper, and added weight that I don’t want to carry.

Stuffing those papers onto my iPad and using GoodReader is a step in the right direction. But still, all too often the PDFs are formatted for paper, not for screen, and I am still cursing the format. PDF usually assumes letter-sized or A4-sized paper, and most ebook readers have physically far smaller screen sizes. Far better I think to start providing EPUB options for academic research, so that folks like me who need ginormous fonts and such can read with greater ease.

Or, should it just go to straight HTML? At that point, papers might even be able to add a little functionality to the electronic reading experience – change variables in information graphics, show rendered 3D representations of models, and so on. EPUB doesn’t support anything fun like HTML5 DOM handling or Flash, although CSS3 might work depending on the EPUB reader’s implementation. Either way, PDF ain’t fitting the bill ebook readers, and I think this sort of format will be far more important in the coming months and years as ebook-capable mobile devices become more and more commonplace.

Eight years

As of today, I’ve been blogging at for eight years. Happy birthday, blog!

The original blog was a hand-rolled PHP/MySQL app, before Michael insisted that I upgrade to WordPress. It was actually a fun little SQL mapping exercise to get my old posts integrated into WordPress’ format, and I’m amazed it worked as well as it did, all things considered.

If you’re wondering what San Bei Ji means, there’s a fine article about it on Wikipedia.


Radio is dead. Long live Radio.

New York Times today: Will the Internet Kill Traditional Car Radio?

Ultimately, the incursion of Internet-based music services and radio station streams may be less about annihilating yet another business model than it is about breaking down barriers. For the first time, small local stations will be able to reach an entire driving nation, so some broadcasters may see their audiences swell as more listeners find them on Internet-connected car radios. In the end, it may simply be a case of radio is dead, long live radio.

Radio is the last bastion of the centralized publishing model, and wireless internet enabled automobile devices stand to challenge this final establishment. People increasingly don’t want to be fed centralized content any more, nor do they appreciate being bound to the radio-listening constraints of local proximity to radio stations. They want to have personalized access to content that predicts their tastes, or else they want to specifically select what they want to hear. They want their content to come from anywhere, and to be available anywhere. This is how things work on our computers, our smartphones, and so on – why should our car radio experiences be any different?

It also occurs to me every time I plug in my iPhone to my car’s audio system that developing a separate car interface for audio consumption, or really any other function, is largely a waste of time and money. Why bother with it? Instead, ship an iPod Touch or some Android-based device embedded in the dash. Let users access their existing streaming content, be it iTunes, Pandora,, NPR apps, and so on.

This goes for other car functions as well. The other day I was getting a ride back to the car dealer in their shuttle, and their 2007 R-class Mercedes had a GPS system that was woefully out of date and buggy. I asked the driver why they hadn’t updated the software or the maps, and he said they just weren’t going to do it for whatever reason. This struck me as completely backwards from the current trends of technology – why not embed a GPS system that updates itself continually over 3G? Small changes could incrementally add themselves to the map database, and software updates pushed through an app store or web interface. Why suffer with a broken map application for a car that is only three years old?

Perhaps all the car functions could be opened up. Provide an API for developers and the smartphone app market do the rest.

Doll up your tabular data

In the science realm, we are often confronted with having to look at a dizzying number of flat tables representing experimental data. It wouldn’t hurt to spend a little time thinking about how that information is displayed in it’s native format. And often when we see tables that are ‘nice’, they are rendered as an image, thereby shielding wary and hapless internet searchers from all of the rich and informative data contained therein. I’m often asked to provide feedback on these things, so I was pleased to find a little tutorial with some nice examples of styling tables using CSS3 that I can throw out there. Here you go!

Pimp Your Tables with CSS3 | Codrops