Manybooks.netDownload statistics of the different ebook formats on manybooks.net. In 2006, out of the 260,000 e-book downloads of public domain works, here’s the breakdown.

PDF 139,995
eReader 28,304
Large Print PDF 17,940
RTF 15,142
Mobipocket 14,860
Sony (Librie, Reader) 14,561
Palm Doc 11,253
iPod 7,528
iLiad PDF 2,889
iSilo 2,886
Plucker 2,835
Custom HTML 1,728
zTXT 1,643
Rocketbook 1,526

You can’t read too much into PG statistics; for example, I would assume more mobipocket ebooks are sold than the above statistics might suggest (mobipocket is closely tied with amazon.com). Also, many kinds of readers support multiple kinds of formats (such as fbreader). Someone using fbreader might download pdb one day, rtf the next, and mobipocket on another. 2007 will surely mark the introduction of new software, new ebooks, new distribution methods and new devices. Still, it’s undeniable that PDF seems to be the downloadable format of choice. The fact that Google is serving pdf’s of scanned works and Sony Reader has prepared a great user guide for optimizing PDFs for the Sony Reader (warning: big PDF!) makes it easier for content creators to produce stuff for Sony readers. Now, if only the price of Sony Reader could fall a bit…

Did you know that Project Gutenberg’s DVD project lets you download gigs of ebooks through p2p? (Here’s a link to a bit torrent site which even lets you create custom PG text collections for torrents). As much as I appreciate these distribution methods, it always helps to have a live human do the selections and layout. Some of those PG files look really crappy (however faithful they are to the original).

In March, I’ll be trekking to attend South by Southwest Interactive in Austin. I’ll keep a few free gigs of free ebooks in my pocket to share with other attendees (I’ll probably be giving a presentation about ebooks at a side conference across the street). Yes, sneakerware still can be an efficient delivery mechanism–as long as the stranger doesn’t accidentally walk off with your 4 gig keychain.

8 COMMENTS

  1. I really can’t understand the popularity of PDF but I’ve seen it on my site (BooksForABuck.com) as well. PDF reads badly on the Palm, it’s horrible on the PC (for those of us with aging eyes who need to blow up the font size), and you can’t read it at all on the eBookWise. HTML seems like the obvious best-choice, with Palm or Microsoft Reader a next-best if you’ve got a Palm or Windows PDA/smartphone.

    Surely people aren’t still printing out eBooks and reading them on paper? What a horribly unecological approach that would be. Not to mention expensive.

    Rob Preece
    Publisher, http://www.BooksForABuck.com

  2. There doesn’t seem to be any mention of TXT or ordinary HTM files here? I presume Custom HTML is something different. With most of Gutenberg’s books in these formats they would probably still be the most popular download formats.
    Perhaps one of the reasons PDF is still popular is that we all have PDF readers on our computers. I sometime print out bits of textbooks or particular passages I am studying at university, but printing a whole book on my rather elderly printer would take longer than reading it !

  3. PG’s volunteers are and always have been free to post in whichever formats they like. What has really happened is that more and more volunteers started posting in HTML alongside the (obligatory) TXT format. This was probably caused by two things: 1) more and more volunteers using Distributed Proofreaders, and 2) more and more Distributed Proofreaders using the GUIguts program, which allows them to make both an HTML and a TXT version from the same source document.

    Another factor may have been the fact that we are processing more and more e-texts from illustrated novels; for those it would be a shame not to produce a rich text version.

  4. PG’s volunteers are and always have been free to post in whichever formats they like.

    Let me qualify that statement, because it is not entirely correct: Project Gutenberg likes to be able to maintain these contributions. So if for instance you contribute an HTML file, the formatting of that file has to follow certain rules, so that when readers spot and send in errors PG can fix these themselves without having to hunt down the particular contributor for that text. (Remember that PG is in it for the long run; it should be able to fix texts without having to rely on the availability of any one person.)

The TeleRead community values your civil and thoughtful comments. We use a cache, so expect a delay. Problems? E-mail newteleread@gmail.com.