Slashdot Mirror
Amazon Releases Kindle Source Code
MackieChan writes with a piece of news that slipped past earlier this month: "Barnes & Noble receives a lot of credit from the Slashdot community for standing up to Microsoft and for allowing the Nook to be so easy to root, but perhaps Amazon releasing the source code to the Kindle will help it gain back supporters it lost after remotely removing ebooks."
Well its a nice show of faith by Amazon, let the hacking begin!
Not just the source to the recent Kindle Fire, but code for all of them back to the original. Nice move.
I wonder if they held any bits back?
Learning HOW to think is more important than learning WHAT to think.
Choose one.
Is the sourcecode sufficient to disable Amazon's ability to remotely remove ebooks?
Slashdot social media options: AIM, ICQ, Yahoo, Jabber and Mobile Text. Why no MySpace?
Way to rake up a totally unrelated 2 year old story about removing eBooks. The Kindle has been kicking ass when it comes to eBook reader sales and eBook sales, not losing supporters. If the source code release was done 2 yrs ago, your reasoning might have made sense.
that's the question. otherwise it's only good for finding security flaws/bugs.
world was created 5 seconds before this post as it is.
I know this gem is hidden somewhere in the Kindle source.
10 REM Write I hate Apple to Screen
20 print "I hate Apple"
30 goto 20
40 gosub Kindleforipad
"That's the way to do it" - Punch
The new Nook tablet comes with a locked bootloader, unike the Nook touch.
I downloaded the source for Kindle_src_3.3_611680021.tar.gz (randomly picked).
The contents of their tarball is the below list of files. Which of these sub-tarballs contains the Amazon reader and interface software? Or are they just releasing the bare minimum required by the GPL and keeping their stuff proprietary? Can Kindle owners blank their devices and use the published tarball to restore all functionality?
Put another way: is there a contribution here, or are they just doing what's necessary to avoid getting sued?
Expert in software patents or patent law? Contribute to the ESP wiki!
The remote book removal was 2 years ago, and helped shape Amazon (and much of the mobile tech industry) to be extremely weary of using kill switches. Frankly, I'm glad it happened. It immediately stopped the usual slow creep of increased user control.
I don't think this helps "gain back supporters", but I do think it reinforces Amazon as a company moving in their new direction since then. I like the Amazon model which tries to take the best of Google and the best of Apple, and throw out the worst parts. Tight product integration, but if you want to hack it, why bother stopping you.
I8-D
"are they just doing what's necessary to avoid getting sued"
why should they do more if it isn't required?
It's useless if you intend to hack the kindle...
On the bright side, there seems to be enough stuff in there to port any Linux (including Android) system into the Kindle (or, saying that in another way, all the drivers seem to be there).
On the bad side, no the reader is not there, and you won't be able to remove their capacity of remotely excluding your books (except if you remove the reader). It is also not more than they are required by the (L)GPL, and there is nothing telling if the boot loader will accept a user supplied system, or if you'll need to root it like any other tablet.
Rethinking email
"are they just doing what's necessary to avoid getting sued"
why should they do more if it isn't required?
Doing more than required is usually what it takes to "gain back supporters" like the summary suggests.
Personally I think they should fix the slow page-turns and other problems pointed out by reviewers.
Fulfilling a GPL requirement is fine; releasing extra code is great, but that's not going to help it sell anymore devices, except perhaps to a small segment of customers.
Put another way: is there a contribution here, or are they just doing what's necessary to avoid getting sued?
One of the major lessons of cryptography is that every code is breakable, it's just a matter of how long it takes to break. Releasing a pile of open source is sort of like encryption in the clear, it will take time and effort to decode what has (and possibly has not) been released.
This release of source code should put a good light on Amazon until the Christmas shopping season is over, it will take at least that long for anyone who cares to stir up trouble for them if they haven't released something they should have.
Because if we, as customers, demand that they do, they'll have to. The way to demand it is to state that we dislike the fact they didn't, and go to someone who does, or if that someone doesn't exist, to someone who has provided more sources.
A source code release would be good for the customers, for the community and for the general progress of mobile devices such as ebook readers. That's why we should insist that source code should be released.
It's just a minimal GPL drop. No application level source. Unlike (for example) Netgear or Linksys, they don't even provide the object code and build tools to let you build your own usable device ROM image from a combination of proprietary and OSS components.
Although ... but
Does using a double comparison mean there is not comparison between the two?
Remember the open source software phones that would brick if users tampered with the firmware? It *all* has to be open to be really useful.
You can get the full Kobo/Touch source code if you e-mail their support staff and wait two to four weeks.
And then you can't really do jack-squat with it.
Which is infuriating because the features most people want to steal from the Kindle are amazingly easy to implement in the Qt environment the Kobo uses.
Yeah, doesn't look to me like they released any of the proprietary stuff that makes it a Kindle. It's not surprising, really. GPL doesn't require it, and this is what makes their software what it is, so from their perspective there's no reason to do so.
Still, as has been pointed out, ideally this is all you'll need to write your own custom OS for Kindles.
Hail Eris, full of mischief...
E pluribus sanguinem
Does anyone care to take this source code and produce a largely "stock" OS but add in the sorely lacking ability to be able to natively read the EPUB format of ebooks?
I'm sure thousands of Kindle owners would be eternally grateful.
They don't need to gain back supporters.
The kinds of people who care about a source code release don't care about remote book wiping because their primary goal is likely CM7/CM9.
Amazon needs to do NOTHING to win these people over, since B&N drove them away with the Nook Tablet (locked bootloader).
retrorocket.o not found, launch anyway?
Comment removed based on user account deletion
So does this mean that there will be kindle knockoffs in the next few months? Also is this going to cause more splintering of the Android ecosystem? Meaning Kindle phones and devices and forget about Google's advancements?
Last time I looked at the code it was just things like busybox.tar.gz and kernel.tar.gz, just to comply with the GPL. You can download all that lot from Sourceforge FFS.
Its not like you can actually compile you own Kindle OS from what they're distributing, there's no Makefile or documentation on how all the bits glue together.
Its certainly not like Android where you can compile your own phone OS (if you have the proprietary blobs for gfx/gps etc.)
#include <sig.h>
Hacks to get a simple (root) shell prompt to the Kindle are actually a lot more useful than this source drop. Google is your friend. Use at your own risk. Slippery when wet.
cpghost at Cordula's Web.
I've had a copy of 1984 (not from Amazon) on my Kindle for a year now. They haven't deleted it.
I don't read your sig. Why are you reading mine?
Amen to that... ended up adding a 3rd party PDF reader from the app store
Nevermind PDF what about epub? The lack of support for epub is my main reason for not buying a Kindle...well that and the fact that the only way to get them in Canada is to order them from the US, paying import duties exchange commission etc. and ending up with an unsupported device.
but perhaps Amazon releasing the source code to the Kindle will help it gain back supporters it lost after remotely removing ebooks.
You are talking about one of the hottest products on the market.
"Winning back" the geek is not all that important.
From the GPL: "a) You must cause the modified files to carry prominent notices stating that you changed the files and the date of any change." So it is required for them to list their changes.
i like the "might win back the customers amazon lost from removing ebooks remotely" how many customers they lost? 5 out of millions? for better or for worse most of us don't care...
They can remove only books with their DRM (that means, stuff you brought from Amazon). Also, they only removed one version, of the several they have available.
I'd buy a Kindle to use as a tablet if it is easy to hack and cheap enough (and meets the specs I need, and if they deliver it here). I'll certanly hack it if I buy one, and won't buy DRMed books (Kindle or not). Thus Amazon probably isn't amazed by the idea of selling it to me. I can see why they'd ignore the entire demographics that thinks like me.
Rethinking email
You don't know what you're talking about.
> One of the major lessons of cryptography is that every code is breakable,
> it's just a matter of how long it takes to break.
Err, I don't think you were paying attention in your crypto class.
1. Cryptography is mainly concerned with ciphers, not codes. "Climb Mount Fuji" is a code.
2. One-time pads, used correctly, can never be "broken" because there are an infinite number of possible decryptions.
Finally - maybe I can finally get it to display a proper 24hr based clock iso of that retarded AM/PM shit...
The source code for the latest Kindle might be news, but Amazon's been posting that part of the source code for a while now. I purchased (and returned, heh, but probably not for reasons you'd guess) a few months ago and had noticed the source code back then.
They did by giving you the source. You can diff those packages against originals for a list of changes.
Persistent Volume manager for Kubernetes - https://github.com/dwimsey/openshift-pvmanager
In particular, any library can be wrapped by command lines or pipes -- why is it that a library that has been wrapped in this way suddenly "mere aggregation"?
First let's get some basic things out of the way. The output of a program is generally not a derivative work of the program itself. This means you can't GPL the output of a program unless the program copies itself into its output. If two independent programs are connected through a pipe or socket, one program's output is the other's input. If the application-level protocol over such a pipe or socket is documented in plain language, it's no different from a non-free web browser connecting to a GPL HTTP server, or running a non-free program in a GPL terminal emulator, or vice versa. The GPL program exposes an interface that the non-free program can call, and vice versa.
I am unaware of any cases that have been tried, but I'd imagine the judge would distinguish between a larger work from an aggregate by looking at the degree of coupling. If the output of one program is defined simply as "that which the other program accepts", then the programs might as well be one work. But if the format of the data sent over the interface is clean enough that a programmer can develop a replacement for the program on either side of the interface, then the programs are more likely to be seen as independent. Furthermore, Ubuntu includes a few non-free device drivers and the non-free MP3 decoder, yet Canonical hasn't been sued.
There's been lots of lawsuits around BusyBox. I don't to what extent they require source releases.
As I understand it, all these have been about providing the "complete corresponding source code" to BusyBox itself, not the source code to other non-free components of a system such as decoders for patented audio and video codecs.
You could also read about Jin vs IChessU
We learn nothing from this lawsuit because it was settled out of court. Chess engines are probably one of the easiest things to make a low-coupling interface for, as algebraic notation appears to predate the GPL itself.
When you connect to a server, the GPL bits remain on the server. When you distribute the GPL bits along with your bits, then the terms of the GPL come into play.
If I sold you a computer with a copy of Windows, a copy of Apache configured to serve only to localhost, and a copy of MediaWiki, would I be breaking the law? MediaWiki is GPL software, and it communicates with the Internet Explorer component of Windows through a socket.
But when you distribute the same sections as part of a whole which is a work based on the Program, the distribution of the whole must be on the terms of this License
This hinges on the definition of "work based on the Program": "either the Program or any derivative work under copyright law: that is to say, a work containing the Program or a portion of it, either verbatim or with modifications". United States copyright law makes a distinction between a derivative work and a collective work, and I take "mere aggregation" to refer to such a collective work. Furthermore, the GPL refers to "mere aggregation [...] on a volume of a storage or distribution medium" (my emphasis). So the interpretation of "work based on the Program" in light of "mere aggregation" can still apply to distribution.
If C doesn't work when you take away A, then you have created a work based on A
GIMP for Windows doesn't work when I take away Windows. So is GIMP for Windows based on Windows in a copyright law sense? If I sell you a computer that has a copy of Windows and a copy of GIMP installed, along with an archive of the GIMP source tree, have I broken the law? Or let's try another tack: What if C starts working again when I take away A and add V? For example, Modplug Tracker for Windows doesn't work when I take away Windows, but it does work when I take away Windows and add Linux, X, and Wine, after a bit of tweaking of the buffer settings. I'd say the works are separate contributions to a collective work when one can be freely replaced with a complete workalike, and that's easiest to demonstrate when the coupling is narrowed to well-defined message passing.
That hardly counts as a prominent notice.
Is your analysis still valid of GPLv3 with its explicit reference to coupling ("intimate data communication or control flow") and its more precise definition of an "aggregate"?
The result of demanding an apology is uniformly poor. You asked for one, I gave one, and you disagreed with its form. I'm done too. Neither of us has won.