5 ms·
I work principally in academia, and I recently bought a Kindle to cut down on printing costs for PDF files (acquired from various sources, principally JStor). I
by rpedroso 15y ago
I work principally in academia, and I recently bought a Kindle to cut down on printing costs for PDF files (acquired from various sources, principally JStor). I find it increasingly unsatisfying.
1. I have to cross-reference page numbers for every citation. When you're talking about hundreds of citations from dozens of sources, you come to regret not reading the originals in the first place. I don't see the day when citing Kindle locations becomes acceptable among the standardized citation guides (MLA, Chicago, etc.) since many users convert PDFs by themselves, and different settings (e.g. omitting introductory pages, dedication pages, title pages, etc.) mean different locations.
2. Annotations. Everyone tells me how easy it is to insert notes and highlight passages, but the currently available methods do not give me the same flexibility as a simple pen:
2A. Discontinuous highlighting/underlining. Sometimes I don't want to highlight every word in a particular passage, but Kindle does not allow me to do this; as soon as the highlight ends, the next highlight can only be treated as a completely new annotation.
2B. I can't draw. Arrows, lines, maps, even illustrations are an important part of my annotation process. On the bottom of one section of Plato's Meno, for example, I have about a half a dozen drawings trying to make sense of a particularly interesting geometry problem (about which several papers have been published) [1]
2C. Multiple language/character support. This is a particularly difficult issue when dealing with translations. Even if I'm not reading a Greek text, I may be on-the-fly translating between English, French, German, and Greek (often 3 or even 4 of the above within the same paper!). As such, quite a few of my notes are in Greek, and so you can imagine my disappointment with the Kindle.
3. Jumping between passages is difficult. Sure, I can search the text - but I guarantee that's actually slower than flipping through pages. And if I direct students to a particular passage? That one kid using a Kindle in the corner of the room is still looking for it by the time we have moved on.
These are not all just Kindle problems, they're obstacles that eBooks and their complimentary software have to overcome to actually obsolete the physical book. I don't mean to suggest that these issues can't be overcome [2], but I believe the paperless revolution is much further away than some like to believe.
For the small amount of for-pleasure-not-work fiction that I consume, I'm probably not going back to paper.
[1] http://homepage.mac.com/jholbo/writings/dialogues/meno/meno9.html http://homepage.mac.com/jholbo/writings/dialogues/meno/meno9...
[2] No doubt the failure to adopt these features stems from low demand -- most of the demand in the industry is for popular literature, not the sort of people who are writing all sorts of things in the margins of their books. There are also some curious UI/UX problems that need solutions, e.g. scanning pages.
- coliveira 15y agoI think the general answer in that the current generation of Kindle and similar readers were created to read paperback novels, not for technical documents. The same way it takes a lot of hard work to create a word processor for technical documents (compare TeX to MS Word), it will take a lot of work to create a good reader for such documents.
- absconditus 15y agoPlato's Meno is not a technical document. His complaints do not appear to be related to technical documents at all.
- kragen 15y agoIt is; it's just a very old technical document.
- icebraining 15y ago1) is pretty easy to implement. Ebooks are searchable; page numbers are hacks around the limitations of paper books. Example: <quote source="Dune (Herbert, Frank)">A process cannot be understood by stopping it. Understanding must move with the flow of the process, must join it and flow with it.</quote> This perfectly identifies a location in the work, regardless of page numbers and other nonsense. A smart viewer will have no problem linking it to the correct place.
- daniel02216 15y agoI've found my iPad to be pretty good at dealing with academic PDFs, but there are still some difficulties that make it annoying. I started taking it to my discussion classes instead of printing the day's paper out, and it worked great. The PDF is reproduced exactly as it prints, and you can quickly scroll from one page to another, search, and zoom in on interesting bits. The ACM two-column format works well with the iPad's PDF reader and iBooks. Later in the year the professor showed up with an iPad as well. :) (There are apps to organize and cite papers with, like Mendeley, that help somewhat, but I haven't found a great one yet.)