I wouldn't use my Kindle half as much if it weren't for KOReader [0]. Ironically enough for a dedicated reading device, Amazon's built-in reader app pales in comparison to this third-party tool. The killer feature for me is on-the-fly column splitting and text reflow, with the ability to flip to the original page view by tapping a corner -- this is critical for reading academic papers, which tend to be two-column PDFs. It also features contrast adjustment, more fonts, stylesheets, wireless syncing with Calibre, and support for many more file formats including ePub.<p>There's also a Gargoyle [1] port for interactive fiction on the go. It's less practical due to the input lag on the Kindle keyboard, but I still pull it out every now and again.<p>[0] <a href="https://github.com/koreader/koreader" rel="nofollow">https://github.com/koreader/koreader</a><p>[1] <a href="http://www.fabiszewski.net/kindle-gargoyle/" rel="nofollow">http://www.fabiszewski.net/kindle-gargoyle/</a>
I did some playing with a Kindle 3 a few years back --- I was writing programs that integrated into the native UI. I built an app which was a Javascript interpreter bolted onto a VT52 terminal emulator. You could type in programs and run them! Using the K3's fiddly little keyboard! Um, awesome. <a href="http://cowlark.com/kindle/javascript.html" rel="nofollow">http://cowlark.com/kindle/javascript.html</a><p>This was on the 3.1 firmware, so it's likely all completely obsolete on modern devices.<p>...the 3.1 firmware was terrible. It was all Java based, but Java 1.4. No generics! No autoboxing! No foreach! People forget just how <i>awful</i> early versions of Java were in comparison to what we have today. I ended up building a toolchain using RetroWeaver to convert modern Java bytecode into something that would run on the Kindle.<p>Also, the firmware was based on the Personal Basis Profile 1.1. Think back, way into the past, before there were smartphones and Android and iOS... back to the heyday of the downloadable Java applet for your T9-based phone. Yup, that. Kindle apps were midlets, and anyone who remembers writing programs for midlets will be shuddering by now.<p>And it gets worse! The Kindle ran the entire UI, third-party applications included, in a single Java VM. It was as fragile as hell, and it tended to silt up with un-garbagecollectable data until it crashed and rebooted. If you left a thread running on application exit, it would crash and reboot. If your app hung you had to power cycle the device. I believe that the reason why Amazon never really opened up the Kindle to large-scale third party apps was mainly embarrassment.<p>Good times. Good times...
Not sure if you are the author or just sharing. If you authored this I would highly recommend mentioning which "kindle" this is applicable to as the first topic. There are multiple generations of kindle e-readers and kindle tablets. It's not readily apparent up front as to which this is applicable to.
The more I see writeups like this, the more I wonder if the effort being laid out by the people doing the work is compensated appropriately.<p>I'm not sure what Amazon pays for identifying a security flaw, but I imagine it's somewhere between $5 and $15k.<p>Having success monthly might yield reasonable compensation, but companies only pay when a flaw is identified, which means you don't get paid for your work, you get paid for your successful work. And you don't get to define what is successful, nor is there usually a clear definition of what successful actually means.<p>I understand that many people do this to get a job in security / security research, but it just seems like the effort-to-payoff ratio still favors people using their found exploits for evil dramatically.<p>There really should be a different pricing model around security exploits - one that encourages responsible disclosure more heavily.
How do you get the background to pursue this sort of thing?<p>I've programmed and used Linux for a little while and I've done some simple things in assembly language (although not in much depth), but all the technical things past the <i></i>CVE-2013-2842<i></i> section are impenetrable to me.
I was recently researching ebook readers and found Kobo devices way better than Kindle (and cheaper too). Especially for somebody who is a power user of devices. Without getting into specifities, I found that in general Kobo is more open.
Btw here is a webservice that converts and sends ePub's to your kindle, without having to use Calibre yourself <a href="http://www.sendepubtokindle.com" rel="nofollow">http://www.sendepubtokindle.com</a>
Maybe that'll make Amazon update their Kindle's experimental browser. It has an underestimated potential.<p>Will keep an eye on this, even if am now using a Kobo H20.