This depends a lot on what page size the PDF was formatted for. My guess is that eBook type digital editions will be formatted for small screen, or at least have text reflow. These should read very well. I think eBooks on the reader look great, and even photos and complex illustrations look pretty good. I’ve never been a huge fan of color printing, so greyscale is mostly OK for me. The only problems are with color fonts and highlighting. A light red font or green highlighter behind black text are easy to see on a color screen, but the low-contrast grey they get translated into can be very hard to read.

Sadly, I don’t care that much about reading eBooks on the 505. I spend a lot more time and eye strain reading technical manuals and scientific papers, and I’d rather read them on the 505 then print them. These types of paper do not, in general, work well with the 505. There are two major problems.

The first is screen size. Most scientific journals format for 8.5×11″, and use a small font too. Even in landscape mode, the 505 screen isn’t good enough to represent these papers without causing major eye strain. The resolution also isn’t good enough, and some line graphs and such can be lost or distorted.

Sony could fix this problem for many types of publication by providing pan-and-zoom or re-paginating features. Most journal articles are written in two columns, so it is possible to represent one 8.5×11 page with four views (top of the left column, bottom of the left column, top of the right column ….) and get a nice image. I can actually do this currently by running a python script on my desktop that duplicates the pdf pages and sets the crop boxes appropriately for the 505 screen. You can also do it with the built in image viewing mode on the 505 if you export a PDF page as an image. Sony could easily add this sort of viewing feature in a firmware upgrade.

The second problem is page rendering speed. The 505 is not nearly as fast at page turning as my Palm TX, but for eBook content it is fast enough to provide a seamless reading experience. For complex scientific pdfs, though, a page turn can take a long time. 5-10 seconds is pretty normal, and I have tried a few documents that took several minutes (during which the reader interface is locked up), and two or three PDFs that crashed the reader outright. I can see how a firmware update might speed up PDF rendering.

I suspect Sony could please a number of folks, and sell a number of 505s by offering digital editions support, but if I had a say in their selection of new features, I’d want faster rendering and pan-and-zoom for viewing of un-encrypted PDF much more than I’d want digital editions.

If you use a stylesheet in an epub (you don’t have to), then the behaviour is the same as a web browser. If the reading device has the specified font, it will be used. If not, a substitute font will be used.

The real problem that I see with this is that the epub spec leaves the font display issue on the reader totally open. If I don’t like the font specified in the epub, can I change it to one I do like? What about using a smaller or larger font size? Digital Editions does give you a range of sizes you can use (four, if I remember correctly). Reader devices and software are free to handle these issues any way they want, or not handle them at all.

Maybe the IDPF didn’t think that they should include this in the spec, or maybe they left it for a later version. In the mean time, we can either live with the choices that the stylesheet specifies and that the reader allows us to override, or we can unzip the epub and manually change the stylesheet.

These font changes can be undesirable in some cases, e.g., when the font choices carry a meaning in a technical document. For example, in a mathematical document, a script L means something completely different than a regular L in the same document. The use of fonts in a SVG embedded in an epub file can pose similar problems.