Hacker News Reader

Comment by saosebastiao | Hacker News Reader

I just had a conversation about this last friday with a coworker. We both basically came to the conclusion that the only e-books that work are prose-oriented, not learning-oriented.

With your average novel, you only read from beginning to end, and you only read words. With a textbook, you not only read words, but you read diagrams and figures. You also skip around...you follow footnotes, references, and you also might follow a curricula that was not defined by the author. You will often refer to two or three different hotspots while studying a single chapter...for example, while working on practice problems, you will flip back and forth between the practice problems, the chapter intro, the context for that problem, and possibly a chapter recap.

This user-adoption gap is not one of the underlying medium, but rather of the user interface. Flipping pages one at a time is not sufficient.

View on HN

There are tons of text features that suck or are absent in ebooks. Even footnotes aren't great, let alone things like facing-page translation—or anything else that relies on two pages being visible side by side at once, which is quite a bit if you're reading much outside of mass-market fiction. Plus, yeah, diagrams, figures, et c.

Poetry often blows in ebook form. A good paper book of poetry has a lot of care put in to precise placement of the text, and it can add a lot to the experience. Ebooks can't do that. Really, any book that benefits from fixed page boundaries has similar problems.

Even one of ebooks' big advantages, full-text search, is worse than using an index unless either 1) you need to search for something that's not in the index, or 2) you're on a device with a full, real keyboard, so not a tablet or eink device.

Ebooks are also very expensive and there's no used market—I'd effectively have to pay ~3x what I did to assemble my physical book collection to replace it with ebooks, provided there were an ebook version for all my dead-trees (not the case). It'd be even worse if most of my books were common paperbacks (the thing ebooks are best at replacing) since you can pick those up for next to nothing if you keep an eye out. 10x the cost to take such a collection digital, I'd imagine.

Sure ebooks can have precise text placement, just use a PDF. As for indexed vs full text search, an eBook can have both, and many do.

No argument on the price, though it didn't hold for works out of copyright.

> Sure ebooks can have precise text placement, just use a PDF.

Not so hot on most e-ink devices, but yeah, you can use a print format to achieve print-like documents that throw the middle finger at your screen's dimensions/resolution with sometimes-nice results.

> As for indexed vs full text search, an eBook can have both, and many do.

Ah, but their indices are usually more of a pain to use than an index in a real book. Like footnotes, but even worse. As with full text search, though, it might beat the dead-tree if you're on a laptop/desktop (bigger screen that can show the index to the side of the text) but at that point there's not much distinction between an "ebook" and anything else one might read on a computer.

> No argument on the price, though it didn't hold for works out of copyright.

Unless they're not originally written in a language you can read, in which case odds are good that the best translation is still under copyright. In fact, it's not uncommon for there to be several copyright-protected translations all of which are better than any of the public domain ones. So technically, yes, you can go to PG and grab a ton of classics for free. Practically speaking, for non-[language(s) you can read] works, if it's worth putting in the time to read it you'll be well served by buying (or borrowing) a newer translation. All those 19th century ones have a way of being both less accurate and less pleasant for a modern reader.

I'd love to see House of Leaves as an ebook. Even a PDF has limits, as do e reader screen sizes. If anyone makes that book work as well as paper, I'll consider ereaders a solved problem.

OneNote on my Surface Pro is about the best thing I've found, and yet it still has major flaws (this is without even getting into how bad its pdf import is). Capacitive touch + active-digitizer stylus can solve a lot of the problems with ebooks (and indeed does, particularly as far as annotation is concerned), but the software is still severely lacking.

The main UX challenge that I haven't found anything has really solved yet is the ability to show 2 or more pages at once (and do so in a way that's simple & intuitive). Also related is some way to mimic the ability to, with a textbook, stick a pen or finger in between two pages as a "temporary bookmark", to make it easy to flip back & forth.

Now that the hardware has caught up to where it needs to be, I think e-textbooks have the potential to get a lot better as software improves. I don't think the physical book will ever die, but I do suspect that ebooks will be much more prevalent in 10 years than they are now, when the software has finally solved UX challenges like these.

>The main UX challenge that I haven't found anything has really solved yet is the ability to show 2 or more pages at once (and do so in a way that's simple & intuitive). Also related is some way to mimic the ability to, with a textbook, stick a pen or finger in between two pages as a "temporary bookmark", to make it easy to flip back & forth.

Maybe a spot on the screen or a button on the side you can press to bounce to some "reference" bookmark, letting you navigate there normally until you release it, at which point you instantly return to the page you were reading.

Depends on a) the quality of the e-book adaptation's typesetting and markup, and b) the size and quality of the device. The textbook publisher needs to invest a little (though possibly not as much as conventional typesetting) and the student needs to shell out on a full 10-inch screen (which need to be used on several books for the investment to amortize out correctly).

Layout/markup: for well-done academic volumes (using Google Play and Sugimoto's "An Introduction to Japanese Society" as my reference piece) things like footnotes are actually a better experience than print - you tap a blue-highlighted superscript, and you get a model that pops up inline for you to read and then swipe away. No looking up the number at the bottom or (even worse) in the endnotes. Tables are all linkable and back buttons work, etc. On the other hand, a straight import of the original text and graphics without any semantic markup can be truly painful to use.

Device quality: this matters more for things like academic volumes with charts/figures (as you mentioned) and graphic novels. If you have content that needs to be viewed on a single screen, and it's organized to fit on 10-inch-diagonal paper page (or larger), it's going to suffer a lot on a 7 inch tablet. especially on small form factors. If you try to read a book with tables and charts built for a full-size hardcover on a 7-inch tablet, it will go badly (I have tried this). On a ten-inch tablet with a screen closer to the hardcover size it's actually quite pleasant.

These two issues interact a bit with the layout/markup issue; for example, I've seen linguistics books that try to give the same text in three versions (foreign language, individual-word translation, and idiomatic translation) as three-line clusters, like so:

    mah     nishma`?
    what    hear.PASS
    How are you doing?

The typesetter didn't know how to make these reflow correctly using e-book markup (which may not be possible given the tools available?), which means that the line breaks are hardcoded, and the ends of long lines will be unviewable past the edge of the screen on 7-inch tablets. On a 10-inch tablet this works quite well.

Doing the things college students do with textbooks (flipping back and forth between a pset and the material, for example) requires a bit of familiarity with the software but it's very doable and I think very pleasant once you get used to it. It is a barrier to entry, though, and I think e-book UIs need to make it easier to learn. This might just be my personal experience, but I find full-text search, even with a tablet screen, much faster than index lookup in print.

That seems abundantly obvious to me. I've been a fairly avid fan of Kindles for about 4 years, but I've never even considered reading a textbook, programming book, or reference book in that format. All I have ever read on my Kindles are novels, comedy, popular science, and biographies, all of which are very prose-oriented and linear. One pop econ book I read recently had a lot of endnotes, which was admittedly a bit annoying, but I don't think it would be significantly less annoying than a physical book with a lot of endnotes. Luckily the endnotes were not very crucial to the text.

I have purchased several physical books from Amazon and chuckled seeing them offered for Kindle. The most recent examples were a large and dense musical theory book that's very reference-oriented and has a lot of diagrams, and a book of Bay Area hiking spots that is obviously very map-oriented and meant to be thumbed through. I can't imagine using either on an e-reader.

I wish I knew more details about what sorts of material they were having these young people read, but I suspect it was largely not prose-oriented.

> We both basically came to the conclusion that the only e-books that work are prose-oriented, not learning-oriented.

Completely agreed. E-books are great for linear text, but the moment you have any kind of cross-referencing it falls apart (even footnotes are troublesome; why can't footnotes be display inline?) Most importantly, they lack discovery; it's a lot harder to flip to a random page.

That said, I have started reading a programming Apple iBooks textbook [1] and I find the format great because it's interactive and because [edit: in principle...] you can copy and paste code snippets directly. In this sense, I think it's actually a lot better than its paper counterpart–but perhaps coding books are a niche (and, I'm sure, some would argue that having to re-type code yourself is a worthwhile exercise in its own right).

[1] http://scenekitbook.com

I do need to try an iBook, but I think good old fashioned web pages work great for reference material [1], especially if you can combine it with something like Diigo [2] for note-taking. Can't beat the book-markability and searchability of the web. What I wonder is if something like a Surface Pro is good enough for pen-written notes, because if I were in school again I'd probably just use that for everything.

[1] Consider the SVN book: http://svnbook.red-bean.com/en/1.7/index.html

[2] https://www.diigo.com

I agree, if by "eBook" we mean only the specific formats (ePub, PDF, mobi). But I'd argue that you can have all if you make a website instead, and read it on a browser. You can then have multiple tabs, bookmarks, large pages with fast scrolling, etc.

And an epub is just a collection of HTML files anyway, so it shouldn't be difficult to make a website instead. I think what's missing is a "website archive format" supported by all major browsers, so the user can have an offline copy of the "book".

> And an epub is just a collection of HTML files anyway, so it shouldn't be difficult to make a website instead. I think what's missing is a "website archive format" supported by all major browsers, so the user can have an offline copy of the "book".

Seems to me that, since an ePub is just a collection of HTML files (well, there's more than that -- there's also an index file, at least, IIRC), the solution is that browsers should be enhanced to read ePub directly. You've already got the "website archive format", you just lack the browser support.

See https://github.com/futurepress/epub.js

It would only take a simple addon to add browser support. There are several in the Firefox catalog.

Also, browsers tend to have better CSS support than e-readers.

> I just had a conversation about this last friday with a coworker. We both basically came to the conclusion that the only e-books that work are prose-oriented, not learning-oriented.

To me, this is precisely backwards.

I would LOVE to have all my technical library in e-book form where I could easily reference it on a tablet instead of the 4 bookshelves and 42 boxes that it sits in.

I never need my entertainment book on a tablet.

In addition, e-readers are still too slow for me. I can flip a page faster than anything renders. This is STUPID given the vast amounts of processing power we have but continues to hold true.

The only time paper books become easier to use than paper books for me is when the book makes extensive use of really fine print in tables or has extremely long lines of monospace text such as badly written source code.

The resolution of an iPad Mini 3 or iPad Air is sufficient to render all the technical books I use quite adequately.

eInk devices do not have the resolution and screen size required for comfortable technical reading. With "Retina" displays on a decent form factor, magic happens.

I use the iBooks reader almost exclusively since it has very good note-taking and place marking tools.

The only thing iBooks can't handle is position sensitive layout, such as some "art" books which rely on specific typography and layout for conceptual flow. PDF can handle some of this, but if the format is not exactly the same as the iPad screen we're back into "I can tell I'm reading a PDF on an iPad" territory rather than "I'm reading a book".

The Kindle is optimised for flowed text. Of course it is painful for technical books.

Came here to say about the same thing. For regular reading I don't care the medium but will pick ebook most of the time so that I have it everywhere I go and can read a chapter or two in the waiting room on my phone and pick up on my tablet later that night. For text books I have yet to see a good digital interface to them, mainly due to not being able to quickly jump between sections/pages. Maybe an interface where I can "cut" out parts of pages and organize them all on a digital corkboard is closer to what I imagine as "the best" way but still I'd pick book over ebook. The one other thing I will say is that I did get ebooks instead of textbooks a number of times in college because it was cheaper and I was told ahead of time that we either never or rarely use the book (but had to prove we got it or lose points in the class).

> The one other thing I will say is that I did get ebooks instead of textbooks a number of times in college because it was cheaper and I was told ahead of time that we either never or rarely use the book (but had to prove we got it or lose points in the class).

Did the professor write the book?

In at least 1 case yes, possibly more.

With a couple of books I reference often, I've bought the physical version after already having the e-book one. It really makes such a difference.

On the other hand, the convenience of looking for a book online, being able to choose one or a couple, and having it in your e-reader after a few minutes is also something not to discard.

Paper is waiting for ereader screens to increase in size and resolution and to grow a facing page.

I'm not sure if ereaders will get there in time, because laptops may overtake them. We'll see...

Either way ten years from now readers will be all-electronic slabs of foldable plastic, and won't look so much like the limited single page Minimum Viable Product designs that are popular today.

The limiting factors for textbook content are commercial, and will probably be bypassed. Academic papers are already moving to open distribution, and I'd expect a GitHub-for-Textbooks with wiki-able content to appear within the next few years.

Publishers don't have the time, insight, or interest to enhance textbook content. If enhanced content is open-sourced the work can be spread across an entire population, which makes it much more viable economically.

The only catch is how this works financially for authors. About that, I have no idea.

I have to ask: how much of this do you think is because of the digital format, vs how much of it is because the digital format doesn't support certain physical book methods we use?

The two that come to mind are post-it notes and flipping through the book. If, perhaps, I could press and hold my finger one end of a tablet to flip through it, maybe it would make them more suitable. Also, perhaps if bookmarks I made in the book were always visible at the top of the reader, it would also make it more suitable.

It might even be useful to make the application present a 3d view of a real book based upon the orientation of the device. That way I could angle it to find roughly how deep within the book a bookmark is.

One could and should argue that reading and learning from webpages is an e-book. All the stuff you mentioned about skipping around, footnotes, following references, are all things that were solved with HTML. It's just that the current state of 'ebooks' have regressed and using their own formats.

So if they learn and read from websites then they are learning from ebooks. I think they would prefer to have the internet than to have paper books.

How about on a tablet? I admit, I don't really like reading a screen (I do love me some e-ink for novels, though). I've found for things that require flipping back and forth (think DFW foot-notes) that an iPad Mini has an intuitive enough interface that I don't find it cumbersome. I think the real challenge is making the ebooks not just text, but interactive. That is, ebooks should add value to the text, not just reproduce it in a different format.

I agree -- the need to flip back and forth ruins ebook readers for technical material. There's a similar problem for writing notes on the page. They allow it, but clumsily.

But it seems like a solvable problem: there should be some brilliant user interface that makes it easy, it's just that no one has found it yet. And I'm not sure it's even possible on the classic Kindle (the epaper one), where each view update is clunky.

I guess I'm an outlier because I can't stand to read my academic stuff on paper. Just the ability to search for a specific word makes it so much easier than trying to figure out if something is in an index, if it might be in the index but using a different term, or any of a bunch of other reasons why indexes are not enough.

I don't know about you guys but my Kindle paper white shows footnotes in a pop-up dialog box. No page fumbling. I thought DFW's books were great on Kindle.

But yes, learning-oriented Ebooks I always read on my laptop. Nothing can compete with the power of cmnd/ctrl+F and multiple windows IMO.

Try reading Infinite Jest on a Kindle. Half of the book is in the end notes.

I made the mistake of buying Pale Fire on Kindle. It is basically unreadable due it's structure.

I actually read IJ on a Kindle and found it significantly easier for the most part than reading it in print because of links to endnotes, which removed the requirement to keep two sets of bookmarks (if you haven't read it, some endnotes in IJ are a sentence are two, some are 40 pages). There were some occasional issues with going back to the main text (IIRC the back button's stack wasn't saved across sleep), but overall it was much better for me.

But no, I'd never read a textbook on a Kindle. Can't flip around.

I thought the loss of the back button during sleep was infuriating.

I don't understand why every application/website hasn't started to use end/footnotes of the same responsiveness as sites like http://marco.org

I don't know if that style would even work for Infinite Jest. David Foster Wallace, the author, is known for his extensive use of frequent, long, multi-page, and often multi-level (footnotes having footnotes) notes.

I look at how Grantland has adjusted to using footnotes, since several of their writers (Bill Simmons, Chuck Klosterman) were influenced, in some part, by Wallace and utilize lots of footnotes, and think it's a decent experience on desktop, but not that great on mobile devices.

The Kindle can't do things like that. Also in the case of the book I mentioned some of the endnotes are dozens of pages long. Those won't fit in a little animated bubble. Some of the endnotes also have their own footnotes.

Reply on news.ycombinator.com