Skimming is the least of it. Current ebooks, in their linear form, are essentially the equivalent of 19th century horse carriages with motors. "Books" need to be reimagined. While reading is and will remain a fundamentally linear activity, there is no reason why books need to be.
For example: Why is text presented flatly? Isn't that a direct consequence of the limitation of physical paper? Where are the ebooks that allow you to dynamically fold text according to hierarchical order? In a mathematical text, say, sometimes I'd like to highlight the core results; other times, I want to home in on a particular result and highlight the "path" (i.e. the "helper thms" that are needed to establish the core result). Yet other times, I'd like to see the connection between results across sections _on the page_, i.e. without paging back and forth and holding everything in short-term memory. Sometimes I want the author's comments and discussion and "connective tissue"; other times, I don't want to see anything but the structure.
Apple did a good thing, a first step with the option to generate flashcards from your highlighting and notes. This is one extremely obvious thing that ebooks were missing, although I think the implementation is not yet very good, rather clumsy in fact. Sometimes I get frustrated that people keep making ebooks as if they were still limit to the same physical constraints as paper books. The whole form and structure and interaction could be made much more malleable, such that the presentation of ideas "adapts" to your brain rather than your brain adapting to a successive, linear, "rectangular", fixed presentation and compensating for it. It's not like "ebooks" simply means adding a bit of "multimedia" like it's the 90s...
I read a lot of math too and the points you raise are good ones. I bought a Kindle DX in hopes of jettisoning a large part of my hardcopy math library; thanks to TeX, a huge swath of math literature is available by PDF (most of it free, if you have university access). I guess I wasn't prepared for how cumbersome it would be. Flipping back 10 pages to check a lemma takes, well, 10 seconds. It really detracts from the study experience.
That said, if I may play devil's advocate for a second, another way to look at the linear structure imposed by print is as a useful constraint for conveying thoughts in an effective manner. Every once in a while, I come across a "classic", math or otherwise, where the author has put so much thought into structuring her argument that I don't need to flip around. Invariably I absorb more from these rare gems than I would from a sort of hypertext arrangement, simply because somebody thought hard about the best way to learn the material.
It's funny you mention Apple because this strikes me as a very Apple thing to do--impose something on its users which initially seems quite cumbersome (one-button mouse, no floppy, one-button cell phone, no CD/DVD, hell the whole HIG), but actually turns out to be a clever mechanism for enhancing usability.
I totally know what you mean, I appreciate the craftsmanship too that went into such gems. It's like knowing you are in "good hands" ;-)
Still, I would appreciate if the form would give me the option to "mess" with it sometimes, e.g. suppressing some stuff that I may or may not come back to later after a first glance; or generating quick links between passages (for example, a proof uses a previous result, can I quickly link back to it); or when I point at a technical term, can I see the formal definition from the text. etc etc. That would be amazing.
It occurs to me, you could submit a feature request to Amazon... "while reading, type CTRL+5 to go forward 5 pages, ALT+5 to go back" or something like that.
Of course this would be cumbersome on the keyboardless models, but they are designed purely for novellas et al anyways.
I'm not a fan of ebooks generally, but would love to have a tablet that could replace a bunch of technical books and papers. Any of the e-ink readers refresh too slowly to really allow skimming, but maybe the iPad or Android tablets would be better. Something like a split pane view for PDFs where you could 'dock' a bunch along an edge for quick reference, or even have multiple papers open at once. I don't know that current devices are either quick enough to render pages quickly, or have enough memory to cache the current page +- 10 others.
Yes, we really need some kind of "hypertext" that allows an expression of "hyperlinks" between concepts so that a reader can follow conceptual paths through a corpus of text at will :)
I agree with you wholeheartedly, actually. It has been disappointing that the advances in electronic textbooks have, as you say, given us the equivalent of scanned-in copies of the old textbooks, but with some embedded encarta video clips.
The main problem, as efforts with XML and microdata and the semantic web (etc, etc) have taught us, is that this (parsing structure, displaying it, and allowing manipulation of that display) is really hard to solve in the general case, and even solutions that get you only part of the way may be so onerous that no one wants to actually author content at that point.
There are a few startups that are instead trying to solve this by producing great textbook content on a case-by-case basis, really thinking about what it means to try to transmit knowledge in a totally different medium. I think there is a huge opportunity there, so I'm hopeful that we'll see some really great innovation in that UI/UX space soon.
It's actually pretty disheartening how few e-reader-native e-books there are these days. Most of the e-books I read are simply reformatted paper books. The better ones -- and, anecdotally, I'd say these represent 5% of the e-books I own -- have hyperlinked chapters. It's pretty clear that the big publishers are still treating e-book editions as afterthoughts.
No doubt this will change. E-book consumption is rising rapidly, and it's abundantly clear that the e-reader/tablet/etc. is the future of the industry. But right now, the e-book is still in its infancy, and publishers haven't yet shifted their priorities in its favor.
Your ideas remind me of "Up and Down The Ladder of Abstraction", a great piece on the future of interactive documents and how far existing standards can take us.
Wow, fascinating ideas. It'd be amazing to have a textbook that had "adaptive abstraction levels" that let you read only the parts you needed detail on in detail, while tying the rest together in a high-level model that made it more useful; and had several different models into which the data fit, letting you read through the same material in the logical order of each model.
Before I migrated my entire library to my computer, I sometimes made carbon copies and cut the theorems out just to reach this sort of effect. Then I could reshuffle and have what I want in front of me instead of flipping around in a book (or across several books) which I hated for the sheer physicality of it.
Nowadays, when I read mostly papers in PDF or books I have scanned in or ebooks, I do screenshots and copy/paste to OmniOutliner, an outliner program.
It's all still a hack though. I'm still waiting for a clean way from within standard ebook reading apps. For example, an API by iBooks that would allow any 3rd party flashcard app to interact with my notes/highlights.
I used to think this use case is so obvious that it'd be a matter of months that someone would come up with a way to make it more easy to both read and write/organize content this way. As it is, now I'm increasingly asking myself whether I'm an exception or whether this is not the way non-literary books _should_ be read after all.
Having some structure is useful when you don't know what you're looking for, or even what kind of thing you're looking for. And I find that's true of a lot of the technical documents I read. I also quite enjoy reading a physical book with a dozen bookmarks and fingers, being able to see several full pages of text at once.
I could see your way of doing things being useful, though. I suspect it hasn't been partly because it's harder than it sounds (look at the design process for HTML, and all of the mistakes that were made and alternatives that were proposed) and partly because so much of the relevant IP (and thus market) is already locked up pretty tightly.
I agree about the structure. The thing about most technical, or how to call them, non-literary? books is that clearly you are not expected to read everything. In a sense, you start with the given structure but make the book your own as you go along. You skim, see that you already know about this, that you are not interested in that, and are _very_ interested in that. What I'd like to see is a format that makes it easy and gives us new possibilities to make a book our own. What I didn't see is your second point. Perhaps I imagined it too easy, maybe it is actually a tough nut to crack. After all, the format must accomodate all sorts of books and styles. And the IP issue, oh my...
Allowing 3rd party apps to access iBook notes and highlights would be useful, but will probably never happen. It's one of the weak points of the app model that there's not any good way to work with other programs' data. On the desktop, you could probably write a script to read annotations from a PDF and dump them into something like SuperMemo, but on tablets you need everyone to agree on a format to export so you can email them to yourself. Maybe Android is better about this.
I don't read lots of fiction lately but the one thing that bugs me is how clumsy sharing (of short passages, memorable quotes etc) is.
Imagine you could just share a memorable sentence or passage to twitter or FB or whatever directly from within your ebook app. And the quote should have a link to a source, for example Google Books or the Kindle web app, such that your friends or followers could see an excerpt of the source if they clicked on it. If you were so inclined, you could even turn the generated link into an affiliate one.
I'm don't know about the Kindle app, but Kindle devices have Facebook and Twitter sharing built in. Once you link to your account, you can just select some text and tweet it.
I've wanted "expandable" math ebooks, which in some ways are similar to what you are describing.
The way these would work is that in their initial, unexpanded state, everything is presented the way it would be presented to an expert in the field. For instance, if you are interested in seeing how the prime number theory is proved, the unexpanded view would present the proof the way it would be presented if it were a new result being published in a top analytic number theory journal.
This would, of course, be over the head of most people--even most mathematicians who are not number theorists. This is where the "expandable" part comes in.
For any part of the proof, you would be able to expand it. There would be several kinds of expansion available.
1. Horizontal expansion. This keeps the argument at the same level, but fills in more steps.
2. Vertical expansion. This is for when you run into something you don't know. It expands to give you the necessary background to continue.
3. Calculation expansion (I need a better name for this). This would be used for things like evaluation of integrals or solving differential equations or similar things that show up in the proof. It's for when you know how to evaluate integrals or solve differential equations, but just aren't seeing the particular substitution or integration by parts or whatever that makes this particular calculation work.
The difference between calculation expansion and horizontal expansion is that calculation expansion is for where you need something expanded that is relatively elementary compared to the current level, whereas horizontal expansion is where you need more steps at the current level to follow the argument.
All of these expansions would in an ideal expandable ebook be applicable recessively, with vertical expansion going all the way down to high school math. The net result would be if you took, say, an expandable ebook on the prime number theorem, and did maximum vertical expansion, you'd end up with in effect a course of study that takes you from high school to understanding the proof of the prime number theorem--it would teach you all the number theory and calculus and complex analysis and so on that you need to understand that proof.
I happily trade the ability to skim for the ability to search by keyword.
The bigger problem with eBooks is that there are a lot of lazy publishers who don't bother proof-reading their ebooks, with the obvious impact on quality.
I've personally resorted to buying the physical book (the cost is the same) and then acquiring high quality "fan made" versions for use on the PC/tablet.
This is why I prefer reading on the iPad or a laptop as opposed to a Kindle. While a Kindle is certainly easier on the eyes, being able to quickly flip through pages and search through a book is just too great to give up.
The big problem with LCD screens is that they give some people headaches. I can't do any serious reading on one. The attendant battery life problems are also a pretty serious issue.
Some examples of how an iPad/PDF saves me time vs a Kindle:
-Let's say I'm looking up how control flow works in a programming language I have little experience with. I can skip to the index, find control flow, and click on that in a few seconds.
-The current chapter covers a new technique that I think could be applied to a previous example, and I want to check that example out. With the iPad/PDF I can easily move back and forth, with a Kindle I can't.
-The current page references a technique from a few pages ago where I don't remember the exact details. I can easily scroll back and forth with an iPad, not so much with a Kindle.
-I want to see if a book covers a relatively obscure topic. The iPad or my computer let me do a search through a PDF for keywords and preview each section containing the keyword, a Kindle doesn't.
I too like reading on iPad more then kindles, but iPad is unfortunately not very fast either. Opening an interactive ebook (the ones they introduced last month) on my iPad 2 takes about 2 seconds...
But if I repeatedly press "next page", my Kindle doesn't do a full refresh on each turn. It's actually redraws the pages quickly enough to be skimmable.
I absolutely agree. This is especially noticeable with more technical documents like research papers or textbooks. On the kindle, to flip to a particular page, you have to go into the menu, pop up a symbols list, navigate with arrows to each digit, then click go. And each of these keypresses takes a long time -- probably half a second -- for the screen to refresh. Then, you repeat log(n) times until you've found (via binary search) the section you were looking for. Poorly thought out.
You don't need to use the symbol list to type numbers — alt–Q will give you a 1, etc. That still doesn't make the process you describe good, but if you must do that, this should at least speed it up considerably.
The worst ebook feature is that the Kindle only stores max(last_read_page, current_page). This means that if you follow a footnote to the end of the book, your reading position is lost forever. This makes flipping through the book a mentally-expensive proposition. (The screen refresh also makes this impossible.)
I'm a big ebook fan, but only for novels or books that read like them. (Which is 90% of the reading I do, so still a big win.)
Yes, and it has a full stack too, so you can, for example, look up the dictionary definition of a word in a footnote, and then follow another link in that definition, etc., and then, when you're done, just repeatedly go 'Back'.
Did that on pg's Hackers & Painters. Back key works just as expected for me. It's like a stack - you can pop significant transitions from it, like following a reference or leaving the book.
I was thinking it would be cool (well, it might only work for me...) to have a "skim" mode on e-readers which let skip, say, 8 pages at a time - but the screen would be horizontally split into 8 sections to show the text from each of the 8 pages.
I only need to see a snippet of text to remind me "oh yeah, it was around here somewhere..." I should whip up a proof-of-concept to test it, I guess.
Do iOS apps have access to one's personal library in iBooks? (The same way music apps like LyricWiki can access your music on an iPhone, I mean.)
If so, we could keep the canonical version of our ebook libraries in iBooks and have additional apps that can access the content in myriad and interesting ways. It would allow the main iBooks app to remain simple and focused, while allowing useful innovations to sprout up around it.
What books need is more aggressive editing. Skimming wouldn't be necessary if the goal were maximizing signal versus noise. I'd pay a premium for the same book more aggressively edited. There's a cost associated with the benefit of reading. Time is now at a premium and part of the cost of a book along with the monetary cost.
Most books are around 250 pages, but I find that most of those 250 page books really only have about 120 pages worth of content.
A while back I asked about it on Quora and discovered that most book publishers aim for 50,000+ words. With the World reaching peak attention, especially for those that read a lot, it's ridiculous to keep insisting on books at 250 pages, especially in the case of technical books and non-fiction
For me, this problem is manifested in referencing bible verses. I love my Kindle, but use it once and you quickly learn that it wasn't designed for random access.
This. Try opening right to Mathew on an e-reader. Then try to realize you need to take a look at another scripture so you keep your finger between the pages you are on flip back a bit read what you need then then easily return to where you were. This holds true in programming books as well.
As pointed out note tasking is less natural in digital format.
In many books, such as DnD books, I know I can find certain rules (or other needed information,) by about where they are in the book.
The Kindle version of the ESV Bible makes clever use of the Kindle's index feature. All the verses in the Bible are indexed using abbreviated book names. So if I want to reach John 3:16, I hit Menu -> Index and type "jn 3 16<enter>". I'd rather have proper fuzzy matching ("jhn 3 16" wouldn't work, you need the exact abbreviation), but it's a workable substitute.
Closely related to skimming - can't do an equivalent of a binary search on it, if you have to jump back 100 pages to look at something you saw. I have a recipe book with 1000 recipes on the Kindle and it's a total nightmare to use. A paper copy is so much more usable.
This doesn't seem un-fixable, though. Something like dragging the page indicator backwards/forwards would help, or mirroring the iPod's wheel gesture to move back-forth more quickly. Spin two fingers for chapters, or something.
The issue is not the interface, it's the speed--an e-ink display is naturally slower right now. It is a real problem but one due to technology rather than UI.
Coincidentally, on my 3rd-generation Kindle (the one with the keyboard), you can jump by chapters using the left and right arrow keys. This ameliorates going back in most books.
Also, Kindles have a nice text search which I've found useful on several occasions. This has also helped somewhat, although it's still lacking overall.
Ultimately, I would not want any non-linear books on my Kindle either--I'll stick to pdfs and actual textbooks for now.
I wish the popular e readers had apis for working with book content - skimming could be only one of the awesome enchancements the tech community could make. Among others would be tying words to wikipedia lookups or research citations to the actual content, aggregating all graphs in some sort of navigable 3d interface, page in page reading, the list could literally go on forever.
on the negative side, we'd probably have to sift through tons of crappy word cloud for your book apps.
I'm reading lots of Scandinavian crime fiction (in translation to English) at the moment.
I have no idea how to pronounce any of the names, so some kind of link from the written name to an audio clip would be useful.
Also, it'd be nice to have some "translator's notes". Perhaps there's some nice phrases in the original that just don't translate well; or some idiom that doesn't have an exact English version.
Sure I can just use a computer and browser and a search engine, but it'd be nicer if it was built in.
You could, for starters, install a Scandinavian dictionary and learn to read pronunciation guides. Assuming the author uses common, and not extravagant invented, names.
If memory serves, highlighting the word brings up not only the definition, but the pronunciation guide. And, even if it doesn't, it isn't much of a pain to click through to the dictionary.
I think it's more just the screen update speed. You can flip pages about as quickly as the screen will update, which is fairly slow. Also, the visible page area is much smaller than a normal book, especially a textbook or a journal, and you can only see one page at a time. It makes for slow skimming and sort of tunnel vision that causes a loss of the visual structure.
Corel PDF Fusion [1] implements "flick" view for skimming. Implementation however is nowhere close to right (too slow, mouse gestures mishandled, etc.)
The updated nook color has a new "skimming" bar. It allows to go from the beginning to the end of the book in a swipe, displaying chapter titles and sample phrases. It's really usable, but of course it couldn't work on an eInk device.
Unfortunately it only works on epub, not PDFs alas.
For example: Why is text presented flatly? Isn't that a direct consequence of the limitation of physical paper? Where are the ebooks that allow you to dynamically fold text according to hierarchical order? In a mathematical text, say, sometimes I'd like to highlight the core results; other times, I want to home in on a particular result and highlight the "path" (i.e. the "helper thms" that are needed to establish the core result). Yet other times, I'd like to see the connection between results across sections _on the page_, i.e. without paging back and forth and holding everything in short-term memory. Sometimes I want the author's comments and discussion and "connective tissue"; other times, I don't want to see anything but the structure.
Apple did a good thing, a first step with the option to generate flashcards from your highlighting and notes. This is one extremely obvious thing that ebooks were missing, although I think the implementation is not yet very good, rather clumsy in fact. Sometimes I get frustrated that people keep making ebooks as if they were still limit to the same physical constraints as paper books. The whole form and structure and interaction could be made much more malleable, such that the presentation of ideas "adapts" to your brain rather than your brain adapting to a successive, linear, "rectangular", fixed presentation and compensating for it. It's not like "ebooks" simply means adding a bit of "multimedia" like it's the 90s...