EDIT: See this comment on my long-term analysis of this generation of Kindle.

James Fallows says that in order to avoid becoming a Kindle bore, you should “Just shut up when tempted to say or write anything about it. Otherwise you’ll be driving people crazy with your enthusing about how useful and convenient it is, and what its potential might be, and how many elegant decisions are evident in its conception and design.” I’m going to violate that right now by enumerating the number of things the Kindle does right and huge, giant thing it does wrong. If this makes me a bore, proceed to the next post.

Things done right: The screen is very, very nice, as is the tactile feel of the device itself. Although notes aren’t as satisfying to write as they are in paper, they work reasonably well and are easily aggregated. Using the “search” feature allows effectively infinite, immediate concordances in realtime. Shopping in the Kindle store is easy, although I think I’ve only bought two books from it because of the DRM.

The most useful thing about the Kindle for me isn’t actually reading books bought from Amazon—I’m reluctant to spend much money on them, knowing there’s a decent chance that in five years I’ll have a different device or won’t be able to transfer the books I buy now. Rather, Marco Arment’s Instapaper makes the Kindle insanely useful. If I find a longish article online, I hit the “Read Later” bookmarklet in Firefox. About once a week, I log into Instapaper and download all those articles on my Kindle. Bingo: I don’t have to keep printing and losing papers and I still get to read everything I want to read.

Things done wrong: The big-time, number one problem with the Kindle is its terrible software for organizing and managing documents. Actually, scratch that: it doesn’t really have software for managing documents.

The Kindle shows up as a generic USB device on OS X. Want to load it with .pdfs? Be prepared to drag them into a folder labeled “documents.” This process reminds me of .mp3 players… before the iPod. This doesn’t bode well for Amazon, especially now that the iPad is out.

The closest third-party app I’ve found so far is Calibre, which is clunky and doesn’t work that well, especially out-of-the-box. It won’t automatically sync to my Kindle at the moment for reasons not abundantly clear to me; it doesn’t have built-in optical character recognition (OCR) for .pdfs; it doesn’t automatically copy things bought off my Kindle to the computer. The list goes on. The difficulty of writing really good, really intuitive software like iTunes is really, really high.

I’m reminded of this post comparing Tumblr and Posterous, which compares those two “reblogging” tools. The basic point: design counts more than technology. At the moment, the Kindle’s technology is impressive. The physical hardware isn’t bad, although the screen should be bigger: there isn’t enough space before I have to scroll. But until iTunes for the Kindle comes along and whisks the searching and sorting problems away, the Kindle is effectively crippled by software.

I’m sure the omission of iTunes-for-the-Kindle is intentional on Amazon’s part: what they really want you to do is pay them money every time you buy a book or convert a .pdf. That’s okay but seems penny-wise and pound-foolish; think of Scott Adams’ complaint about bad user interfaces. At the end he asks, “What is your biggest interface peeve?” I now have one.

In other news, Apple released the iPad not long ago, which virtually every media outlet on the planet has covered. Megan McArdle says of it:

I’m still unsure how the iPad gets around the core problem: it doesn’t replace anything. Buying an iPhone let me take my phone, my camera, and my iPod out of the briefcase. Buying a Kindle let me remove a newspaper, several books, and some documents I have on PDF.

You can see similar comments here.

But if the iPad software is sufficiently better than the Kindle, users might end up chiefly with it.
One should read this article from Paul Buchheit’s blog, in which he notes the three reasons why the original iPod succeeded where others didn’t. It was:

1) small enough to fit in your pocket, 2) had enough storage to hold many hours of music and 3) easy to sync with your Mac (most hardware companies can’t make software, so I bet the others got this wrong).

Emphasis added. The weird thing is that Amazon is getting this wrong right now. Syncing the Kindle to my computer is cumbersome; there isn’t a good program for organizing my books and .pdfs. Charlie Stross writes about why he, a self-described UNIX bigot, uses a couple of Macs, instead of cheaper Linux boxes:

The reason I choose to pay through the nose for my computers is very simple: unlike just about every other manufacturer in the business, Apple appreciate the importance of good industrial design.

(Note: he’s British, which explains the “Apple appreciate” rather than “Apple appreciates.” The Brits think of corporations as plural, we think of them as singular. What would Steven Pinker say?)

I would also add that Apple has fewer and different hassles than running Linux boxes, which I say as someone who had periodic problems with audio drivers and other things in the ~2001 – 2003 range before I gave up. But the Kindle’s hassles are reminiscent of a product that should be better than it is. I’ve drifted somewhat from the main point regarding the Kindle, but the device is one of these “close, but still wrong” items that is somewhat frustrating, much like Linux, the last Volvo I drove, the Ikea desks I’ve seen, and chairs that unsuccessfully mimic the Aeron.

The Kindle is very, very good for English majors who get assigned a lot of pre-1923 fiction (which they can get free online) or for people who like reading from that era and do so voluminously. For the rest of us, it lacks, especially in the nonfiction department, where it’s hard to skip from section to section quickly.

Reading fiction on it is a substantially better experience because I seldom skip long sections in novels—it’s pretty hard to decide an entire chapter should be skipped, since that chapter will usually contribute something important to the story (and, if it doesn’t, the novel isn’t very good). In addition, novels are relatively unlikely to have research citations, which are sometimes important in evaluating nonfiction, especially if that nonfiction makes extensive or dubious claims. Right now, the small amount of nonfiction I’ve got doesn’t come with footnote hyperlinks. It shouldn’t be all that hard to create a style named, say, footnote with an automated number linking it to a later number so that one can jump freely back and forth between them. But that’s rare in the books I’ve read.

Amazon has released a kindle Software Development Kit (SDK), which might improve some of its current problems. But until it solves the “organizing home” problem that iTunes does so well, it’s not going to be a tremendously useful device for me and many other heavy readers who need some way of getting articles to and from the device. That’s a huge missing feature that Instapaper (somewhat) solves, but not well enough. The Kindle is an “almost” device, like many of the “almost” mp3 players before the iPod. But I don’t think almost is enough.