I've been working on TPMs lately. It turns out that they're moderately awful, but what's significantly more awful is basically all the existing documentation. So here's some of what I've learned, presented in the hope that it saves someone else some amount of misery.
What is a TPM?TPMs are devices that adhere to the Trusted Computing Group's Trusted Platform Module specification. They're typically microcontrollers[1] with a small amount of flash, and attached via either i2c (on embedded devices) or LPC[2] (on PCs). While designed for performing cryptographic tasks, TPMs are not cryptographic accelerators - in almost all situations, carrying out any TPM operations on the CPU instead would be massively faster[3]. So why use a TPM at all?
Keeping secrets with a TPMTPMs can encrypt and decrypt things. They're not terribly fast at doing so, but they have one significant benefit over doing it on the CPU - they can do it with keys that are tied to the TPM. All TPMs have something called a Storage Root Key (or SRK) that's generated when the TPM is initially configured. You can ask the TPM to generate a new keypair, and it'll do so, encrypt them with the SRK (or another key descended from the SRK) and hand it back to you. Other than the SRK (and another key called the Endorsement Key, which we'll get back to later), these keys aren't actually kept on the TPM - the running OS stores them on disk. If the OS wants to encrypt or decrypt something, it loads the key into the TPM and asks it to perform the desired operation. The TPM decrypts the key and then goes to work on the data. For small quantities of data, the secret can even be stored in the TPM's nvram rather than on disk.

All of this means that the keys are tied to a system, which is great for security. An attacker can't obtain the decrypted keys, even if they have a keylogger and full access to your filesystem. If I encrypt my laptop's drive and then encrypt the decryption key with the TPM, stealing my drive won't help even if you have my passphrase - any other TPM simply doesn't have the keys necessary to give you access.

That's fine for keys which are system specific, but what about keys that I might want to use on multiple systems, or keys that I want to carry on using when I need to replace my hardware? Keys can optionally be flagged as migratable, which makes it possible to export them from the TPM and import them to another TPM. This seems like it defeats most of the benefits, but there's a couple of features that improve security here. The first is that you need the TPM ownership password, which is something that's set during initial TPM setup and then not usually used afterwards. An attacker would need to obtain this somehow. The other is that you can set limits on migration when you initially import the key. In this scenario the TPM will only be willing to export the key by encrypting it with a pre-configured public key. If the private half is kept offline, an attacker is still unable to obtain a decrypted copy of the key.
So I just replace the OS with one that steals the secret, right?Say my root filesystem is encrypted with a secret that's stored on the TPM. An attacker can replace my kernel with one that grabs that secret once the TPM's released it. How can I avoid that?

TPMs have a series of Platform Configuration Registers (PCRs) that are used to record system state. These all start off programmed to zero, but applications can extend them at runtime by writing a sha1 hash into them. The new hash is concatenated to the existing PCR value and another sha1 calculated, and then this value is stored in the PCR. The firmware hashes itself and various option ROMs and adds those values to some PCRs, and then grabs the bootloader and hashes that. The bootloader then hashes its configuration and the files it reads before executing them.

This chain of trust means that you can verify that no prior system component has been modified. If an attacker modifies the bootloader then the firmware will calculate a different hash value, and there's no way for the attacker to force that back to the original value. Changing the kernel or the initrd will result in the same problem. Other than replacing the very low level firmware code that controls the root of trust, there's no way an attacker can replace any fundamental system components without changing the hash values.

TPMs support using these hash values to decide whether or not to perform a decryption operation. If an attacker replaces the initrd, the PCRs won't match and the TPM will simply refuse to hand over the secret. You can actually see this in use on Windows devices using Bitlocker - if you do anything that would change the PCR state (like booting into recovery mode), the TPM won't hand over the key and Bitlocker has to prompt for a recovery key. Choosing which PCRs to care about is something of a balancing act. Firmware configuration is typically hashed into PCR 1, so changing any firmware configuration options will change it. If PCR 1 is listed as one of the values that must match in order to release the secret, changing any firmware options will prevent the secret from being released. That's probably overkill. On the other hand, PCR 0 will normally contain the firmware hash itself. Including this means that the user will need to recover after updating their firmware, but failing to include it means that an attacker can subvert the system by replacing the firmware.
What about using TPMs for DRM?In theory you could populate TPMs with DRM keys for media playback, and seal them such that the hardware wouldn't hand them over. In practice this is probably too easily subverted or too user-hostile - changing default boot order in your firmware would result in validation failing, and permitting that would allow fairly straightforward subverted boot processes. You really need a finer grained policy management approach, and that's something that the TPM itself can't support.

This is where Remote Attestation comes in. Rather than keep any secrets on the local TPM, the TPM can assert to a remote site that the system is in a specific state. The remote site can then make a policy determination based on multiple factors and decide whether or not to hand over session decryption keys. The idea here is fairly straightforward. The remote site sends a nonce and a list of PCRs. The TPM generates a blob with the requested PCR values, sticks the nonce on, encrypts it and sends it back to the remote site. The remote site verifies that the reply was encrypted with an actual TPM key, makes sure that the nonce matches and then makes a policy determination based on the PCR state.

But hold on. How does the remote site know that the reply was encrypted with an actual TPM? When TPMs are built, they have something called an Endorsement Key (EK) flashed into them. The idea is that the only way to have a valid EK is to have a TPM, and that the TPM will never release this key to anything else. There's a couple of problems here. The first is that proving you have a valid EK to a remote site involves having a chain of trust between the EK and some globally trusted third party. Most TPMs don't have this - the only ones I know of that do are recent Infineon and STMicro parts. The second is that TPMs only have a single EK, and so any site performing remote attestation can cross-correlate you with any other site. That's a pretty significant privacy concern.

There's a theoretical solution to the privacy issue. TPMs never actually sign PCR quotes with the EK. Instead, TPMs can generate something called an Attestation Identity Key (AIK) and sign it with the EK. The OS can then provide this to a site called a PrivacyCA, which verifies that the AIK is signed by a real EK (and hence a real TPM). When a third party site requests remote attestation, the TPM signs the PCRs with the AIK and the third party site asks the PrivacyCA whether the AIK is real. You can have as many AIKs as you want, so you can provide each service with a different AIK.

As long as the PrivacyCA only keeps track of whether an AIK is valid and not which EK it was signed with, this avoids the privacy concerns - nobody would be able to tell that multiple AIKs came from the same TPM. On the other hand, it makes any PrivacyCA a pretty attractive target. Compromising one would not only allow you to fake up any remote attestation requests, it would let you violate user privacy expectations by seeing that (say) the TPM being used to attest to HolyScriptureVideos.com was also being used to attest to DegradingPornographyInvolvingAnimals.com.

Perhaps unsurprisingly (given the associated liability concerns), there's no public and trusted PrivacyCAs yet, and even if they were (a) many computers are still being sold without TPMs and (b) even those with TPMs often don't have the EK certificate that would be required to make remote attestation possible. So while remote attestation could theoretically be used to impose DRM in a way that would require you to be running a specific OS, practical concerns make it pretty difficult for anyone to deploy that at any point in the near future.
Is this just limited to early OS components?Nope. The Linux kernel has support for measuring each binary run or each module loaded and extending PCRs accordingly. This makes it possible to ensure that the running binaries haven't been modified on disk. There's not a lot of distribution infrastructure for setting this up, but in theory a distribution could deploy an entirely signed userspace and allow the user to opt into only executing correctly signed binaries. Things get more interesting when you add interpreted scripts to the mix, so there's still plenty of work to do there.
So what can I actually use a TPM for?Drive encryption is probably the best example (Bitlocker does it on Windows, and there's a LUKS-based implementation for Linux here) - while in theory you could do things like use your TPM as a factor in two-factor authentication or tie your GPG key to it, there's not a lot of existing infrastructure for handling all of that. For the majority of people, the most useful feature of the TPM is probably the random number generator. rngd has support for pulling numbers out of it and stashing them in /dev/random, and it's probably worth doing that unless you have an Ivy Bridge or other CPU with an RNG.

Things get more interesting in more niche cases. Corporations can bind VPN keys to corporate machines, making it possible to impose varying security policies. Intel use the TPM as part of their anti-theft technology on education-oriented devices like the Classmate. And in the cloud, projects like Trusted Computing Pools use remote attestation to verify that compute nodes are in a known good state before scheduling jobs on them.
Is there a threat to freedom?At the moment, probably not. The lack of any workable general purpose remote attestation makes it difficult for anyone to impose TPM-based restrictions on users, and any local code is obviously under the user's control - got a program that wants to read the PCR state before letting you do something? LD_PRELOAD something that gives it the desired response, or hack it so it ignores failure. It's just far too easy to circumvent.
Summary?TPMs are useful for some very domain-specific applications, drive encryption and random number generation. The current state of technology doesn't make them useful for practical limitations of end-user freedom.

[1] Ranging from 8-bit things that are better suited to driving washing machines, up to full ARM cores
[2] "Low Pin Count", basically ISA without the slots.
[3] Loading a key and decrypting a 5 byte payload takes 1.5 seconds on my laptop's TPM.

comments

Last Saturday, I was invited to talk about Debian to Hackerspace DF, a group that is starting to work at a very nice place together with other collectives, in a quite centric place (Colonia Obrera). I know several of the people in the group (visited them a couple of times in the space's previous incarnation), and wish them great luck in this new hackerspace! Anyway I was invited to give an informal talk about Debian. And of course, I was there. And so was Alfredo, who recorded (most of) it. So, in case you want to see me talking about how Debian works, mostly on a social organization level (but also regarding some technical details). Of course, given the talk was completely informal (it started by me standing there, asking, "OK, any questions?"), I managed to mix up some names and stuff... But I hope that, in the end, the participants understood better what Debian means than when we started. Oh, and by the end of the talk, we were all much happier. Not only because I was about to shut up, but because during my talk, we got notice that Debian 7.0 "Wheezy" was released. Anyway If you want to see me talking for ~1hr, you can download the video or watch it on YouTube.

Ordinarily, I m a big believer that it is important to keep up to date with what every piece of software which competes with yours is doing, to remain educated on the latest concepts. Sometimes, there are concepts that get added which are definitely worth ripping off. We ve ripped off plenty of the better design choices from Windows or Mac OS, over the years, for use in the Free Desktop. So, what about Windows 8, the hip new OS on everyone s lips? Well, here s the thing I ve been using it on and off for a few months now for running legacy apps, and I can t for the life of me find anything worth stealing. Let s take the key change Windows 8 has apps built with a new design paradigm which definitely isn t called Metro. Metro apps don t really have windows in the traditional sense they re more modeled on full-screen apps from smartphones or tablets than on Windows 1.0 -> 7. Which is fine, really, if you re running Windows 8 on a tablet or touchscreen device. But what if you re not? What about the normal PC user? As Microsoft themselves ask:

The answer to that is, well, you sorta don t. Metro apps can exist in three states fullscreen, almost fullscreen, or vertical stripe. You re allowed to have two apps at most at the same time one mostly full screen, and one vertical stripe. So what happens if you try to *use* that? Let s take a fairly common thing I do watch a video and play Minesweeper. In this example, the video player is the current replacement for Windows Media Player, and ships by default. The Minesweeper game isn t installed by default, but is the only Minesweeper game in the Windows 8 app store which is gratis and by Microsoft Game Studios. Here s option A:

And for contrast, here s option B:

Which of these does a better job of letting me play Minesweeper and watch a video at the same time? Oh, here s option C, dumping Microsoft s own software, and using a third-party video player and third party Minesweeper implementation:

It s magical almost as if picking my own window sizes makes the experience better. So, as you can see above, the old OS is still hiding there, in the form of a Windows 8 app called Desktop . Oh, sorry, didn t I say? Metro apps, and non-Metro apps, are segregated. You can run both (the Desktop app can also be almost-fullscreen or a vertical strip), but they get their own lists of apps when multitasking. Compare the list on the left with the list at the bottom:

And it s even more fun for apps like Internet Explorer, which can be started in both modes (and you often need both modes). Oh, and notice how the Ribbon interface from Office 2007 has invaded Explorer, filling the view with large buttons to do things you never want to do under normal circumstances. So, that s a short primer on why Windows 8 is terrible. Is there really nothing here worth stealing? Actually, yes, there is! After much research, I have discovered Windows 8 s shining jewel:

The new Task Manager is lovely. I want it on my Linux systems. But that s it.

(It took me a while to come up with new CDBS packaging series post not because I stopped using CDBS just because I was procrastinating myself as busy) This is the second post in the CDBS packaging series. In this series I'm going to talk about package relationship management. The better example where this feature is useful is packages where build-depends and run-time dependencies overlap. Most of the Perl modules which have test suites have build-depend and run-time dependency intersection. So let me take example of a Perl module. First lets see control file of a Perl package which is not using CDBS and then let me explain how CDBS can help you improve the situation. I choose libxml-libxml-perl lets see part of control file which includes Build-Depends Depends Suggests Recommends. So 2 packages are both in Build-Depends and Depends field

1. libsax-xml-perl
2. libxml-namespacesupport-perl

So in this situation there is a possibility that we miss to add one or both of these packages into Depends field, I'm not saying we will surely miss but we might after all we are all human beings. So how can we improve the situation using CDBS? Let me go through step by step on what we need to do.

1. Create a file called control.in with same above contents but slight modification in Build-Depends and Depends section. I will display the diff below to avoid re-pasting entire file again and again.

1. Next we need to have something like below in the rule files

So basically we moved all Build-Depends and Depends to rules file. The common ones are placed in deps variable and assigned to both Build-Depends and Depends. CDBS uses following variables for package relationship management.

1. CDBS_BUILD_DEPENDS: This variable helps you manage Build-Depends field and all you need to do is in your control.in files Build-Depends field put place holder @cdbs@
2. CDBS_DEPENDS: This field can be used to manage Depends field of binary package for each binary package you need to have one CDBS_DEPENDS_pkgname variable with depends assigned to it. And in your control.in append $ cdbs:Depends to Depends field.
3. CDBS_PROVIDES, CDBS_BREAKS, CDBS_RECOMMENDS, CDBS_PREDEPENDS, CDBS_SUGGESTS, CDBS_REPLACES: all these does the job what you think it does :-).

Other than CDBS_BUILD_DEPENDS all other variables work using substvars i.e. CDBS will put the respective substitutions in pkgname.substvars file which will be used during deb creation to replace things in control file. So to make CDBS generate new control file run the below command Basically this command needs to be executed before starting build process if you miss your changes will not be reflected into debian/control. Additionally this feature is Maintainer Mode helper tool because Debian policy prohibits change of debian/control during normal package build. So what is the benefit of using this feature of CDBS? I've listed some of them which I felt are obvious.

1. When there is intersection in Build-Depends and Depends this feature from CDBS is helpful. As I shown above put all intersecting dependencies in common variable and appropriately assign them wherever it is required. Thus we can avoid possible missing of some of run-time dependencies due to human error.
2. It is also possible that newer version of package requires specific version of some package (mostly libraries) and we updated build dependencies but we might forget to do the same in Depends, by using this feature we can make sure we will not miss such minute details.

One last thing I want to point out is if you are NMUing a CDBS package

NMUs need not (but are encouraged to) make special use of these tools. In particular, the debian/control.in file can be completely ignored.

Before closing down the post, If you find some mistake in the post please let me know either through comments or through the email. Soon I will be back with new CDBS recipes till then cya.

There was Lenny, there was Squeeze, now there is Wheezy. Another major release of Debian where I had the pleasure to do the ftpmaster work for the release. Like the last times, Mark joined to help with the work. But 2 FTPMasters aren t enough for one Wheezy, or so, so Ansgar had his first run aside from point releases. With the whole load of work the release team did in the past to prepare this, combined with all the changes in dak we had since Squeeze, it turned out to be rather easy for us. Again, a few moments and not a detailed log (times in CEST):

• [10:45] Let s prepare. Cronjobs go off on ftp-master and security, shared tmux session setup, including a live pipe into a file so other team members can follow using less +F -R , a titanpad page opened to share in writing the commands to process later.
• [11:18] One last package update done manually. The latest and greatest of installation-guide, 20130503, really wants to join the rest in wheezy.
• [12:00] While I did a database backup and started preparing the set of commands for later, Mark and Ansgar prepared our code to deal with the changes we are about to make.
• [12:33] I start executing the commands (shell and sql), so somewhen after here we have Squeeze as oldstable (with all its extra suites also renamed), Wheezy as stable (taking its own suites with it, as well as gaining the -updates). And Jessie starts its life.
• [12:45] Heissa, wouldn t be fun if we wouldn t have some bugs, so we just created a load of directories not needed or wanted. Not getting Jessies either. Ohwell.
• [13:00] Jessie gets a copy of Wheezy to start with. Made the mistake to try that with our usual tool, control-suite. Err. Wrong. This is doing a set of checks on each package inserted. Would take ages. While one SQL to bypass the checks was a minute or so.
• [13:xx] Got annoyed by our concept of consistency. Its great. Every tool takes suite names/codenames in a different way. Seperated with spaces, with commas, with codenames, with suitenames, we mix it in all possible combinations.
• [14:00] Ansgar finished updating our security master host to cope with the changes for the release.
• [14:39] Squeeze Release files got regenerated and signed again.
• [14:54] Wheezy Release files got generated and signed.
• [15:05] The cd build machine has a synced mirror tree, so they can start building our CD/DVD images
• [16:25] A manual dinstall finished. And while it had some small problems (including nearly overwriting something inside the wheezy dists/ tree), it went pretty smoothly.
• [21:00] After having kept the archive turned off for the whole day, we have 90 uploads with ~3Gb of uploads waiting. As it is now unlikely that we need larger archive changes, we turn (most of) our cronjobs back on, to get them processed.

Right now (that is, 21:08 CEST) I am mostly waiting for the CD build to finish. Current schedule seems to have that in some two and a half hours, after which we can push the mirrors and our press people announce it. Except if Murphy wants to show up, but lets hope that not. Thanks go out to everyone involved in preparing this release, be it in the past by fixing bugs, uploading packages, doing whatever was needed, as well as doing the work today.

Review: Democracy at Work, by Richard Wolff

Publisher:	Haymarket
Copyright:	2012
ISBN:	1-60846-247-1
Format:	Kindle
Pages:	220

I've been reading (mostly on-line) and thinking quite a bit lately about workplace governance models, economic structure, and why the current organization of the US workplace bothers me so intensely, partly triggered by reading John Kenneth Galbraith's The Affluent Society. The economic monoculture has made that process particularly frustrating. It's rare to find a discussion, even in the context of organizational strategies that are considered radical, that avoids the standard frame of productivity and business value. Most discussion is long on tactics and short on strategy and examination of goals. Wolff's appearance on Moyers & Company was a rare breath of fresh air, enough so that I grabbed his book shortly afterwards. Wolff is a Marxian economist (meaning that he makes use of Marxist analysis of economics and capitalism while separating them from Marxist politics and Marx's advocacy of revolutionary socialism), which for me was part of the interest. Marxist thought of any branch is not common in the United States; we're regularly deprived of several sides in the international conversation on economic models. I was taught Marxist theory in elementary and high school (in retrospect, surprisingly even-handedly and well, despite the biases of my schooling), but none of the later developments of Marxist thought. I think that's a typical experience here; in the United States, Marxism culminated in Mao and Stalin, and no further development of the underlying theories is ever mentioned. Democracy at Work is subtitled A Cure for Capitalism and does indeed advocate a concrete alternative to capitalist business structures. But this is only the last third of the book (and in some ways the least useful, as I'll discuss in a moment). The first two-thirds of the book is basically a remedial education in modern, as opposed to historic, Marxian economics for US readers like myself who have never heard it before, cast in the context of the current financial crisis. This may well be old hat for Europeans, but if you've been wondering what (at least some) modern-day Marxists actually believe, or are saying to yourself "there are modern-day Marxists after the collapse of the Soviet Union?", I recommend this book to your attention. It's an excellent summary, which I read with the delightful feeling of an expanding viewpoint and the discovery of new directions from which to look at a problem. There's quite a bit in this section that's worth thinking about, including another take on the nature of the recent economic collapse and how that fits into a Marxian analysis of capitalist crises. But there was one point in Wolff's explanation that I found particularly helpful. He completely restructured my understanding of the Marxian analysis of worker exploitation and profit allocation. There are two angles of Marxist economic thought and socialist economics that get a great deal of attention, at least in the United States, in history and economics classes: the role (or lack thereof) of markets in price setting, and the ownership of the means of production. Defenders of capitalism like to focus on the former, since it's quite easy to identify the advantages of theoretical free markets in finding ideal prices and balancing supply and demand, whereas central planning of prices and production has resulted in some catastrophic and deadly failures. (Although I will note, with passing interest, that those failures predate large-scale computing, and there are now large corporations that manage budgets larger than some countries via centralized command-and-control economic practices.) Defenders of socialism are more likely to focus on the ownership of the means of production, since it's easy to show prima facie unfairness in owners of capital extracting vast profits without having to do any work themselves, only be lucky enough to start with large quantities of money. Wolff, however, argues that both of these focuses misses a core critique by Marx of the workplace structure in capitalism, and that, by ignoring that critique, supposedly Marxist countries did not create anything that was actually Marxist in implementation. The Soviet Union was just as much a capitalist country as the United States is. It was state capitalism rather than private capitalism, but the core capitalist structure was intact. Wolff arrives at this conclusion, which may be well-trodden ground in parts of the world that include active Marxist thought but which was quite startling to this American, by treating the ownership of capital as a partial distraction. He focuses on a more direct question and practical question: who determines what a worker does on a day-to-day basis and how that product is used? Who determines what profits are collected and how they are spent? In private capitalism, this is done by the owners of the capital: large shareholders, major investors, and the managerial class that they hire. In Soviet state capitalism, this is done by national politicians, bureaucrats, and the managerial class that they hire. In neither model is it done by the workers themselves. The Soviet model gives theoretical ownership of the capital to the workers, but that ownership is diffused, centralized, and politicized, redirected through the mechanisms of the state, and therefore is effectively ignored. Ownership and control is entirely captured by the political class. Both of these systems are capitalist in Wolff's view of Marx: there is a class of owners and managers, who control the terms and nature of work and who allocate the profits, and a class of workers, who have to do what they're told, are not paid full value for their labor, and don't have a say in how the profits their work generates are spent. At the most important level of day-to-day autonomy and empowerment, they are functionally identical. They are both equally hierarchical and exploitative; the only difference is in whether the system is controlled by rich individuals or by well-connected politicians. (And, as any study of modern politics quickly reveals, the distinction between those two groups in most countries is murky at best.) Wolff convincingly recasts modern economic history as a constant pendulum swing between private capitalism and state capitalism. Crises in one system push countries towards the other system; subsequent crises push the country back towards the first system. Regulation grows and shrinks, companies are nationalized and then privatized, but both systems are united in excluding the worker from any meaningful control over their work life. The first two-thirds of the book was full of insights like this for me. I didn't agree with all of it, but all of it was worthwhile and thought-provoking. But I was a bit leery of Wolff's proposed solution. My past experience with critics of capitalism is that the critique is often quite compelling, but the proposed solution is much less believable. And, sadly, that concern was warranted here as well. The core of Wolff's proposal is predictable but possibly sound: a restructuring of the workplace to be radically democratic. The business would be owned entirely and exclusively by the people who work for it, equally regardless of the job of the worker, and the workers would decide democratically or via elected repesentatives from among the workers how to allocate the profits of the business, what standards and business practices should be followed, and how the work is to be done. I was particularly interested to hear that this model (workers' self-directed enterprises) has apparently been successful in Spain in the form of the Mondragon Cooperative. Given all the tricky, small details that have to be resolved in an actual workplace, an existence proof is worth more than pounds of theory. Unfortunately, like a lot of proposed alternatives, Wolff's description of WSDEs is quite fuzzy and involves a lot of hand-waving. I was never able to build, from this book, a coherent and complete mental model of how such a workplace would function. Wolff tends to surround every specific in a halo of contingencies, possibilities, and alternative models, and is maddeningly nonspecific on such practical matters as how line management would work, how such a business would do financial planning or project approval, how competing interests in different parts of the organization would be balanced, and other practical governance matters that fill my work life. Maybe the answer to all of that is just "democracy," but I'm dubious. Democracy has a number of well-known flaws that I thought weren't adequately addressed. For example, democracies are often quite happy to further and reinforce existing prejudice (such as sexism or racism), and are prone to yielding control to the most charismatic. Democracies also have an informed voter problem, which seems like it would be particularly acute if democracy is going to make detailed business decisions. And, for larger organizations, control by pre-existing money could re-enter the equation via propaganda and campaigning around votes. Some of this gap in the book could be addressed via a more in-depth look at how Mondragon and any other real-life examples work, but that is sadly missing here. (I am interested enough now, though, that I'd read a good popular treatment of the history and methodology of Mondragon, although I don't think I'm up to working through an academic study.) The hierarchical, dictatorial management structure imposed by capitalism is so awful that WSDEs don't have a particularly high bar to meet to be fairer and more empowering than what we have today. The question, rather, is would they function sufficiently well that the business would be able to make effective decisions, and that's unclear to me from this book. This is, as Wolff spends some time discussing, particularly difficult when in direct competition with capitalist enterprises. This sort of endeavor will probably trade some degree of economic efficiency and raw marketplace power for improvements to fairness and empowerment, but that means it's going to require support from the surrounding society, which is a huge obstacle. I doubt there's a free lunch here: true fairness and equity also leading to improved economic efficiency in a capitalist context is a nice dream, but not horribly practical. Wolff also seems to suffer from something else I associate with Marxist thought: a preferential focus on a very narrowly-defined type of productive worker, apparently left over from Marx's original critique in the context of industrialization. Wolff inserts a very odd and quite awkward distinction in his WSDE model between workers who directly produce the product that's sold, and who in his model therefore produce the profits of the business, and all other workers in the business. He then gives special privileges to the former group to decide how much profit to return in worker compensation and how much to use for other purposes, thus making the supporting workers second-class citizens within this supposedly equal workplace. Speaking as someone who works in IT, and hence would be classified by Wolff into the supporting rather than directly productive category, I do not find this division at all convincing, and Wolff never provides a coherent explanation for why he introduced it. He only says that it's necessary for the governance of the business to not be exploitative, which seems to assume that there is a special economic role played by the workers who work directly on a saleable product. Maybe there is some analysis that could convince me of this, but, if so, it's not present in this book. It struck me as a recipe for continuing the exploitation of the most invisible and powerless workers in capitalism: janitors, groundskeepers, and other low-paid service jobs. I wish the whole book were as insightful and pointed as the first two-thirds, but alas I found the WSDE discussion to be somewhat muddled and utopian. "How do we get there from here" is always the hardest part of this type of discussion, and Wolff has no special skill in that department. But, despite that, I got a lot of fascinating ideas and new conceptual frameworks out of this book, and I'm tempted to read it again. I suspect some of this, similar to my discovery of promises and infinite streams in programming, is filling in of odd gaps in my personal education rather than a discovery of unusual, new ideas. But if you too have gotten your political education within the US capitalism ber alles bubble, this book may fill in similar gaps in your knowledge. If so, it's a very rewarding experience. If you're curious about a preview of Wolff's perspective without paying for the book, I recommend watching the first episode of Moyers & Company in which he appeared. Wolff is a clear and engaging speaker, and his interview provides a good feel for his discussion style and his general perspective. Rating: 8 out of 10

For (at least) the third time, I managed to register for debconf before registration was actually open. Oops. I found out that unfortunately, it's not quite certain yet that there will actually be a debcamp this year and if there is going to be a debcamp, it won't be a full week. Pity. At any rate, I'll be there the whole time, whatever the duration of debcamp. Since Vaumarcus is closer to Mechelen than Edinburgh (by about 250km), this is going to be the closest debconf for me, ever. And if I could go to Banja Luka by car, I can certainly go to Vaumarcus by car. Anyone care to join me?

Dear all, Registration is now open for DebConf13, which will take place in Le Camp near Vaumarcus, Switzerland, from Sunday 11 August to Saturday 17 August 2013. The registration deadline for those who wish to make sponsorship applications is Wednesday 15 May 2013. After this date, you can still register, but you will need to pay for accommodation and food. Registration If you want to attend DebConf13, go to the Registration page and follow the procedure outlined there. Please even read the website page if you attended previous conferences, because it explains some important changes with respect to previous years. DebCamp We are planning to prepend the current DebConf week with additional DebCamp days, but we need some more time to sort out the details. We also need more money to be able to fully sponsor DebCamp for everyone or to extend it to more than a few days. If you would like this to happen, please contact us and help find additional sponsorship. We will send another announcement about DebCamp as soon as registration for DebCamp opens. Thank you DebConf13 thanks its two platinum sponsors, Google and the Matanel Foundation, as well as all other sponsors. We look forward to welcoming you to Le Camp in August! The DebConf team

The DebConf13 matching fund announced previously is now complete. The matching sponsor, Brandorr Group will match the collected amount of 2513 USD, resulting in 5026 USD of sponsoring money for DebConf13, for the benefit of all attendees. Despite raising only slightly more than half of the expected money, this first experience of raising funds through this matching fund mechanism can be considered successful.

DebConf is very important to the health of Debian, and Debian is very important to our team, so we re doing what we can to help. In this case we really wanted to see if we could provide an incentive for individual users of Debian to donate to the project. It seems like it worked well, and we d love to do something like this again in the future.

Brian Gupta, Brandorr Group

DebConf13 continues to welcome donations from individuals and is also seeking further sponsors. Registration opening Registration for DebConf13 will be opening shortly: stay tuned! We look forward to welcoming you to Vaumarcus!

Review: The System of the World, by Neal Stephenson

Series:	The Baroque Cycle #3
Publisher:	William Morrow
Copyright:	2004
ISBN:	0-06-052387-5
Format:	Hardcover
Pages:	892

This is the third book of the three-volume Baroque Cycle. I think you could, if you really wanted, read it without reading the previous volumes; Stephenson is certainly long-winded enough that you can pick up most of what's going on while you read. It's been a year since I read the second volume, and I only resorted to Wikipedia a couple of times to remember plot elements (and mostly from the first book). However, I wouldn't recommended starting here. Many of the character relationships, and most of the underpinning of the plot, is established in the previous volumes and given more significance by them. You would also miss The Confusion, which is the best book of the series, although none of this series rises to the level at which I'd recommend it except under specific circumstances. Quicksilver establishes the characters of Daniel Waterhouse, a fictional Puritan whose family was close to Cromwell and who became a friend to Isaac Newton in the days following the Restoration; Jack Shaftoe, a vagabond who wanders Europe in a sequence of improbable adventures; and Eliza, who becomes a friend to Leibniz and a spy for William of Orange. The Waterhouse sections are prominent in Quicksilver: full of the early history of the Royal Society, alchemy, and a small amount of politics. Of those three characters, Eliza is by far the most interesting, which meant that I was delighted when The Confusion dropped Waterhouse almost entirely and mixed Eliza's further story with more improbable but entertaining sea adventures of Jack Shaftoe. You will immediately sense my root problem with The System of the World when you hear that it is almost entirely about Daniel Waterhouse. While Eliza and Jack both appear, they play supporting roles at best, and Eliza's wonderful sharp intelligence and pragmatic survival skills are left out almost entirely. Instead, this is a novel about Waterhouse's return to England after spending quite a bit of time in the American colonies working on calculating machines. He is almost immediately entangled in dangerous politics from multiple directions: the precarious national politics in England near the end of the reign of Queen Anne, Isaac Newton's attempts to maintain the currency of England as Master of the Mint, and a bombing attempt that may have been aimed at him, may have been aimed at Newton, and may have been aimed at someone else entirely. Much of the book consists of an extended investigation of this bombing plot, skullduggery involving counterfeiters, and attempts to use the currency and the Mint as part of the political conflict between Whigs and Tories, mixed in with attempts to construct a very early computer (this is Stephenson, after all). Leibniz and Eliza come into this only as confidants of the Hanoverians. All this may sound exciting, and there are parts of it that hold the attention. But this book sprawls as badly as Quicksilver did. There's just too much detail without either enough plot or enough clarity. Stephenson tries to make you feel, smell, and hear the streets of London and the concerns of an idiosyncratic group of semi-nobles during one of the more interesting junctures of British history, but he does that by nearly drowning you in it, and without providing enough high-level guidance. For most of the book, I felt like I was being given a tour of a house on my hands and knees with a magnifying glass. It's a bad sign when the reader of a historical novel is regularly resorting to Wikipedia, not to follow interesting tangents of supporting material, but to try to get a basic sense of the players and the politics involved because the author never explains them clearly. If you're more familiar with the details of British history than I am, and can more easily follow the casual intermixing of two or three forms of address for the same historical figure, you may not have that problem. But I think other structural issues remain, and one of the largest is Waterhouse himself. Jack Shaftoe, and particularly Eliza, are more interesting characters because they're characters. They're not always particularly believable, but they attack the world with panache and are constantly squirming into the center of things. Stephenson's portrayals of Newton, Leibniz, the Duke of Marlborough, Sophia of Hanover, Peter the Great, and the other historical figures who show up here are interesting for different reasons: Stephenson has history to draw on and elaborate, and it's fascinating to meet those people from a different angle than dry lists of accomplishments. History has a way of providing random details that are too bizarre to make up; Isaac Newton, for example, actually did disguise himself to infiltrate London criminal society in pursuit of counterfeiters while he was Master of the Mint! Waterhouse, for me, has none of these advantages. He is an invented character in whom I have no pre-existing interest. He drifts through events largely through personal connections, all of which seem to be almost accidental. He's welcome in the councils of the Royal Society because he's apparently a scientist, but the amount of actual science we see him doing is quite limited. His nonconformist background allies him squarely with the Whigs, but his actual position on religious matters seems much less set than the others around him. What he seems to want, more than anything else, is to help Leibniz in the development of a computer and to reconcile Newton and Leibniz. And he's not particularly effective at either. In short, he has little in the way of memorable character or dynamism, despite being the primary viewpoint character, and seems to exist mostly to know everyone and be everywhere that's important to the story. He feels like an authorial insertion more than a character. It's quite easy to believe that Stephenson himself would have loved to be in exactly the role and situation that Waterhouse finds himself in throughout the book, in the middle of the councils of the wise and powerful, in just the right position to watch the events of history. I can sympathize, but it doesn't make for engrossing reading. Novels live and die by the strength of their characters, particularly their protagonists; I want more than just a neutral viewpoint. The third major structural problem that I had with this book is that I think Stephenson buries his lede. After finishing it, I think this is a book with a point, a central premise around which all the events of the story turn, and which is the philosophical culmination of The Baroque Cycle as a whole. But Stephenson seems oddly unwilling to state that premise outright until the very end of the book. For the first half, one could be forgiven in thinking this is a story about alchemy and the oddly heavy gold that's been a part of the story since The Confusion, or perhaps about foundational but forgotten work on computation that preceded Babbage by a century. But those all turn out to be side stories, sometimes even without a proper conclusion. I appreciate honoring the intelligence of the reader, and I presume that Stephenson would like to guide the reader through the same process of realization that the characters go through, but I think he takes this much too far and fails to make the realization clear. I'll therefore state what I believe is the premise outright, since I think it's a stronger book with this idea in mind: The System of the World is a continuation of the transformational economics shown in The Confusion into the realm of politics. Specifically, it's about the replacement of people with systems, about the journey towards Parliamentary supremacy, central banking, and the persistent state, and about the application of scientific principles of consistency and reproducibility to politics and economics (however fitfully and arbitrarily). Quicksilver was about the rise of science; The Confusion was, in retrospect, about the rise of economics; and The System of the World tries to be about the rise of technocratic modern politics, barely perceptible among the squabbles between Tories and Whigs. I think that's a fascinating premise, and I would have loved to read a book that tackles it head-on. That's a concept that is much more familiar from the late 19th and early 20th centuries in the context of Marxism, early socialism, technological utopianism, and similar attempted applications of scientific analysis to political and human behavior for the betterment of human civilization. Shifting that 200 years earlier and looking at a similar question from the perspective of the giants of the Enlightenment feels full of of potential. There are moments when I think Stephenson captures the sense of a seismic shift in how economies are run, knowledge is established, and civilizations are knit together. But, most of the time, it just isn't clear. There's so much other stuff in this book, and in the whole series: so many false starts, digressions, abandoned plots, discarded characters, and awkward attempts at romance (as much as I like the characters, Stephenson's portrayal of the relationship between Eliza and Jack is simply ridiculous and not particularly funny) that the whole weight of the edifice crushes what I think is the core concept. Stephenson is never going to be sparse. When you start a Stephenson novel, you know it's going to be full of chunks of partly digested encyclopedia and random research findings that may have nothing to do with the plot. But his best books (Snow Crash, The Diamond Age, even Cryptonomicon) have an underlying structure off of which all of those digressions are hung. You can see the bones beneath the flesh, and the creature they create is one you want to get to know. I'm not sure there are any bones here, and that may be the peril, for Stephenson, of writing historical fiction. I wonder if he felt that the structure of history would provide enough structure by itself that he could wrap a few plots around the outside of it and call it good. If so, it didn't work, at least for me. A lot of things happen. Some of them are even exciting and tense. A lot of people meet, interact, and show off their views of the world. A great deal of history, research, and sense of place is described in painstaking detail. But at the end of the book, I felt like I had to reach for some sort of point and try to retrofit it to the story. Lots happened, but there wasn't a novel. And that makes it quite hard to get enthused by the book. If you adored Quicksilver, I suspect you will also like this. I think they're the most similar. If, like I did, you thought The Confusion was a significant step up in enjoyment in the series and were hoping the trend will continue, I'm sad to report that it didn't. If you were considering whether to read the whole series and were waiting to see what I thought of the end, my advice is to give The Baroque Cycle a pass unless you absolutely love Stephenson's digressions, don't care if they're about history instead of current technology, and cannot live without 3,000 pages of them. It's not that they're bad books, but they're very long books, they take a significant investment of time and attention, and I think that, for most readers, there are other books that would repay a similar investment with more enjoyment. Rating: 5 out of 10

This is my monthly summary of my free software related activities. If you re among the people who made a donation to support my work (102.70 , thanks everybody!), then you can learn how I spent your money. Otherwise it s just an interesting status update on my various projects. Debian France Work on Galette. I spent quite some time on Debian France s galette installation (the web application handling its member database), first converting its Postgres database to UTF-8, then upgrading to 0.7.4 while working-around many known problems. I also created Debian packages of three Galette plugins that we have been using (galette-plugin-paypal, galette-plugin-admintools, galette-plugin-fullcard). But every time I use galette, I tend to find something to report. This month I filed 5 tickets:

• #588: galette should offer a way to send a test mail while setting up the mail notifications
• #589: CSV export page contains an invalid download link
• #590: confirmation page of a successful PayPal payment contains empty fields
• #591: problem with the selection of recipients of a mailing
• #595: galette should put a proper recipient in the To: field of automatically generated mails.

I tested quite some fixes prepared by the upstream author (3 of the above bugs are already fixed) and this lead to the 0.7.4.1 bugfix release. Preliminary work on new bylaws. I have setup a git repository to make it easier to collaborate on new versions of our bylaws and internal rules. The goal is to make Debian France a trusted organization of Debian and to update everything to be compliant with the association 1901 law (we currently have a special statute reserved to associations from Alsace/Moselle). Kali Linux Improve accessibility support in Debian Wheezy. Offensive Security wanted Kali Linux to be fully accessible to disabled people. Since Wheezy was suffering from some serious regressions in that area, we hired Emilio Pozuelo Monfort to fix #680636 and #689559 in gdm3. On my side, I updated debian-installer s finish-install to correctly pre-configure the system when you make an install with speech synthesis (patch submitted in #705599).
Thanks to accommodating release managers, this work has already been integrated in Wheezy and won t have to wait the first point release. Fix bugs in Debian s live desktop installer. We also wanted to enable the desktop installer in the Kali live DVD. While our first tries a few months ago failed, this time it worked almost out of the box (thanks to Ben Armstrong who fixed it). I still identified a few issues that I fixed in debian-installer-launcher s git repository. Packaging and misc Debian work

• I reviewed the work of Charles Plessy who drafted an important update of the Debian Policy to document dpkg triggers (see #582109)
• I reviewed the libwebsockets package prepared by Peter Pentchev (ITP 697671)
• I discovered Tanglu and joined their mailing list because I want to watch its evolution (and maybe use it as a test-bed for some future infrastructure developments).
• I reviewed and committed a patch of Robert Spencer on debian-cd (see #703431).
• I packaged version 3.3 of cpputest (in experimental). I tested a new upstream snapshot converted to autotools.

I also spent a significant number of hours to answer questions of students who want to participate in Google s summer of code and who are interested by the rewrite of the Package Tracking System with Python and Django. Some of the discussions happened on debian-qa@lists.debian.org. Thanks See you next month for a new summary of my activities.

One comment Liked this article? Click here. My blog is Flattr-enabled.

Shared Memory Fences In our last adventure, dri3k first steps, one of the future work items was to deal with synchronization between the direct rendering application and the X server. DRI2 handles this by performing a round trip each time the application starts using a buffer that was being used by the X server. As DRI3 manages buffer allocation within the application, there s really no reason to talk to the server, so this implicit serialization point just isn t available to us. As I mentioned last time, James Jones and Aaron Plattner added an explicit GPU serialization system to the Sync extension. These SyncFences serializing rendering between two X clients, but within the server there are hooks provided for the driver to use hardware-specific serialization primitives. The existing Linux DRM interfaces queue rendering to the GPU in the order requests are made to the kernel, so we don t need the ability to serialize within the GPU, we just need to serialize requests to the kernel. Simple CPU-based serialization gating access to the GPU will suffice here, at least for the current set of drivers. GPU access which is not mediated by the kernel will presumably require serialization that involves the GPU itself. We ll leave that for a future adventure though; the goal today is to build something that works with the current Linux DRM interfaces. SyncFence Semantics The semantics required by SyncFences is for multiple clients to block on a fence which a single client then triggers. All of the blocked clients start executing requests immediately after the trigger fires. There are four basic operations on SyncFences:

• Trigger. Mark the fence as ready and wake up all waiting clients
• Await. Block until the fence is ready.
• Query. Retrieve the current state of the fence.
• Reset. Unset the fence; future Await requests will block.

SyncFences are the same as Events as provided by Python and other systems. Of course all of the names have been changed to keep things interesting. I ll call them Fences here, to be consistent with the current X usage. Using Pthread Primitives One fact about pthreads that I recently learned is that the synchronization primitives (mutexes, barriers and semaphores) are actually supposed to work across process boundaries, if those objects are in shared memory mapped by each process. That seemed like a great simplification for this project; allocate a page of shared memory, map into the X server and direct rendering application and use the existing pthreads APIs. Alas, the pthread objects are architecture specific. I m pretty sure that when that spec was written, no-one ever thought of running multiple architectures within the same memory space. I went and looked at the code to check, and found that each of these objects has a different size and structure on x86 and x86_64 architectures. That makes it pretty hard to use this API within X as we often have both 32- and 64- bit applications talking to the same (presumably 64-bit) X server. As a last resort, I read through a bunch of articles on using futexes directly within applications and decided that it was probably possible to implement what I needed in an architecture-independent fashion. Futexes Linux Futexes live in this strange limbo of being a not-quite-public kernel interface. Glibc uses them internally to implement locking primitives, but it doesn t export any direct interface to the system call. Certainly they re easy to use incorrectly, but it s unusual in the Linux space to have our fundamental tools locked away for our own safety . Fortunately, we can still get at futexes by creating our own syscall wrappers.

static inline long sys_futex(void *addr1, int op, int val1,
                 struct timespec *timeout, void *addr2, int val3)
 
    return syscall(SYS_futex, addr1, op, val1, timeout, addr2, val3);
 

For this little exercise, I created two simple wrappers, one to block on a futex:

static inline int futex_wait(int32_t *addr, int32_t value)  
    return sys_futex(addr, FUTEX_WAIT, value, NULL, NULL, 0);
 

and one to wake up all futex waiters:

static inline int futex_wake(int32_t *addr)  
    return sys_futex(addr, FUTEX_WAKE, MAXINT, NULL, NULL, 0);
 

Atomic Memory Operations I need atomic memory operations to keep separate cores from seeing different values of the fence value, GCC defines a few such primitives and I picked _syncboolcompareandswap and _syncvalcompareandswap. I also need fetch and store operations that the compiler won t shuffle around:

#define barrier() __asm__ __volatile__("": : :"memory")
static inline void atomic_store(int32_t *f, int32_t v)
 
    barrier();
    *f = v;
    barrier();
 
static inline int32_t atomic_fetch(int32_t *a)
 
    int32_t v;
    barrier();
    v = *a;
    barrier();
    return v;
 

If your machine doesn t make these two operations atomic, then you would redefine these as needed. Futex-based Fences These wake-all semantics of Fences greatly simplify reasoning about the operation as there s no need to ensure that only a single thread runs past Await, the only requirement is that no threads pass the Await operation until the fence is triggered. A Fence is defined by a single 32-bit integer which can take one of three values:

• 0 - The fence is not triggered, and there are no waiters.
• 1 - The fence is triggered (there can be no waiters at this point).
• -1 - The fence is not triggered, and there are waiters (one or more).

With those, I built the fence operations as follows. Here s Await:

int fence_await(int32_t *f)
 
    while (__sync_val_compare_and_swap(f, 0, -1) != 1)  
        if (futex_wait(f, -1))  
            if (errno != EWOULDBLOCK)
                return -1;
         
     
    return 0;
 

The basic requirement that the thread not run until the fence is triggered is met by fetching the current value of the fence and comparing it with 1. Until it is signaled, that comparison will return false. The compareandswap operation makes sure the fence is -1 before the thread calls futex_wait, either it was already -1 in the case where there were other waiters, or it was 0 before and is now -1 in the case where there were no waiters before. This needs to be an atomic operation so that the fence value will be seen as -1 by the trigger operation if there are any threads in the syscall. The futex_wait call will return once the value is no longer -1, it also ensures that the thread won t block if the trigger occurs between the swap and the syscall. Here s the Trigger function:

int fence_trigger(int32_t *f)
 
    if (__sync_val_compare_and_swap(f, 0, 1) == -1)  
        atomic_store(f, 1);
        if (futex_wake(f) < 0)
            return -1;
     
    return 0;
 

The atomic compareandswap operation will make sure that no Await thread swaps the 0 for a -1 while the trigger is changing the value from 0 to 1; either the Await switches from 0 to -1 or the Trigger switches from 0 to 1. If the value before the compareandswap was -1, then there may be threads waiting on the Fence. An atomic store, constructed with two memory barriers and a regular store operation, to mark the Fence triggered is followed by the futex_wake call to unblock all Awaiting threads. The Query function is just an atomic fetch:

int fence_query(int32_t *f)
 
    return atomic_fetch(f) == 1;
 

Reset requires a compareandswap so that it doesn t disturb things if the fence has already been reset and there are threads waiting on it:

void fence_reset(int32_t *f)
 
    __sync_bool_compare_and_swap(f, 1, 0);
 

A Request for Review Ok, so we ve all tried to create synchronization primitives only to find that our obvious implementations were full of holes. I d love to hear from you if you ve identified any problems in the above code, or if you can figure out how to use the existing glibc primitives for this operation.

Wow, my first "Debian day" in 2013, but it's a bit of a misnomer because I didn't do any Debian work tonight. I spent some time poking at geary, Yorba's new email client which looks promising. I haven't managed to get it working much, though. Someone packaged an old version in Debian but it wouldn't work with either my home or my work IMAP/SMTP servers for various reasons. I did get a git checkout to build in March or April but that stopped working on Debian when they dropped "Precise" support (It seems they were backporting various GIR/Webkit bits and pieces into their own code and didn't want to carry that around any more). They've currently got a fundraiser going, they're aiming for $100k and (at the time of writing) need half of that with only 10 hours to go. Recently I've also been trying to fix a rockbox bug on Sansa Fuze v1 which really cripples writing onto the Fuze. The original firmware does not support microsd capacities >32G which rules it out for me. Sadly the v1 version of the Fuze is uncommon enough that I don't think this bug is getting much attention. My investigations have been limited to fairly dumb bisect-like approaches, combined with lots of writing onto a microsd card (which will probably be killed dead by all this). My initial attempts at a bisect have failed as it turns out the problem is the inverse of a regression: it bizarrely seems as if the bug was fixed in a version-branch, rather than stopped working in one, so it has never been squashed in the main branch at all. Furthermore, the commit that seems to fix the problem is a totally benign version bump, and almost certainly hasn't fixed the bug. So with sufficient further testing I'll probably just prove that this bug has never been fixed. It might be time to switch tack and look into diagnosing the bug rather than trying to side-step it. Or perhaps just spend 15 on a Sansa Clip Zip and forget all about the Fuze v1. Either way I hope someone releases a 128G microsd card this year! Every now and then I wonder whether it would make sense to package rockbox in Debian in some way. Probably not.

The Debian Edu / Skolelinux project is still going strong and made its first Wheezy based release today. This is the release announcement: New features for Debian Edu ~7.0.0 alpha0 released 2013-04-26 This is the release notes for for Debian Edu / Skolelinux ~7.0.0 edu alpha0, based on Debian with codename "Wheezy". About Debian Edu and Skolelinux Debian Edu, also known as Skolelinux, is a Linux distribution based on Debian providing an out-of-the box environment of a completely configured school network. Immediatly after installation a school server running all services needed for a school network is set up just waiting for users and machines being added via GOsa , a comfortable Web-UI. A netbooting environment is prepared using PXE, so after initial installation of the main server from CD, DVD or USB stick all other machines can be installed via the network. This is the first test release based on Wheezy (which currently is not released yet). Basically this is an updated and slightly improved version compared to the Squeeze release. Software updates

• Everything which is new in Debian Wheezy, eg:
  • Linux kernel 3.2.x
  • Desktop environments KDE "Plasma" 4.8.4, GNOME 3.4, and LXDE 4 (KDE is installed by default; to choose GNOME or LXDE: see manual.)
  • Web browser Iceweasel 10 ESR
  • LibreOffice 3.5.4
  • LTSP 5.4.2
  • GOsa 2.7.4
  • CUPS print system 1.5.3
  • Educational toolbox GCompris 12.01
  • Music creator Rosegarden 12.04
  • Image editor Gimp 2.8.2
  • Virtual universe Celestia 1.6.1
  • Virtual stargazer Stellarium 0.11.3
  • Scratch visual programming environment 1.4.0.6
  • New version of debian-installer from Debian Wheezy, see installation manual for more details.
  • Debian Wheezy includes about 37000 packages available for installation.
  • More information about Debian Wheezy 7.0 is provided in the release notes and the installation manual.

Documentation

• The (English) Debian Edu Wheezy Manual is fully translated to German, French, Italian and Danish. Partly translated versions exist for Norwegian Bokmal and Spanish.

LDAP related changes

• Slight changes to some objects and acls to have more types to choose from when adding systems in GOsa. Now systems can be of type server, workstation, printer, terminal or netdevice.

Other changes

• LTSP clients start as diskless workstation / thin client can be configured via command line argument -- or individually adding an entry in lts.conf or LDAP.
• GOsa gui: Now some options that seemed to be available, but are non functional, are greyed out (or are not clickable). Some tabs are completely hidden to the end user, others even to the GOsa admin.

Regressions

• No mass import of user account data in GOsa (ldif or csv) available yet.

No updated artwork

• Updated artwork which is visible during installation, in the login screen and as desktop wallpaper is still missing or the same as we had for our Squeeze based release.

Where to get it To download the multiarch netinstall CD release you can use

• ftp://ftp.skolelinux.org/skolelinux-cd/wheezy/
• http://ftp.skolelinux.org/skolelinux-cd/wheezy/
• rsync -avzP ftp.skolelinux.org::skolelinux-cd/wheezy/

The MD5SUM of this image is: c5e773ddafdaa4f48c409c682f598b6c The SHA1SUM of this image is: 25934fabb9b7d20235499a0a51f08ce6c54215f2 How to report bugs http://wiki.debian.org/DebianEdu/HowTo/ReportBugs

Online tech news site Ars Technica (which I recommend, by the way) recently reviewed the Dell XPS 13 Developer Edition. Its unique feature: it ships with Ubuntu Linux as the default operating system. This preload deal had a few unique properties:

• It s from a major system vendor, not a no-name or third-party integrator.
• It s a desktop-oriented product, not a server.
• Most notably, the vendor actually put effort into making it work well.

That last point deserves some explanation. A few vendors have grabbed a Windows computer they sell and allowed the option to preload Linux on it, but without support; you re on your own if it doesn t work in some way, which is likely. Essentially, they save you the time of wiping Windows off the box and doing a fresh install, but not much more. But this laptop comes out of Dell s Project Sputnik, a project to put out Linux machines for developers with a DevOps flavor, and they felt the machine had to work as well as their regular products. So they actually put effort and testing into getting the laptop to run Ubuntu well, with all the drivers configured properly and tweaked to support the machine s quirks, just like they do for Windows. And so, the review is surprised to learn that Ubuntu on the XPS 13, well, just works! It s even in the title of the review. Here s reviewer Lee Hutchinson s observations:

I ve struggled before with using Linux as my full-time operating environment both at work and at home. I did it for years at work, but it was never quite as easy as I wanted it to be on an older Dell laptop, keeping dual monitor support working correctly across updates required endless fiddling with xorg.conf, and whether or not it was Nvidia s fault was totally irrelevant to swearing, cursing Past Lee, trying desperately to get his monitors to display images so he could make his 10am conference call without having to resort to running the meeting on the small laptop screen.

And thence comes the astonishment: on this Linux laptop, everything just works. Most of the review is spent on the kinds of hardware features that distinguish this from other laptops: the keyboard is like this, the screen is that resolution, it has this CPU and this much RAM and so on. Some space is devoted to impressions of the default Ubuntu 12.04 install, and some space is given to the special DevOps software, which helps the developer reproduce the software environment on the laptop when deploying apps. But before all that, Hutchinson has to put in a dig:

It s an impressive achievement, and it s also a sad comment on the overall viability of Linux as a consumer-facing operating system for normal people. I don t think anyone is arguing that Linux hasn t earned its place in the data center it most certainly has but there s no way I d feel comfy installing even newbie-friendly Ubuntu or Mint on my parents computers. The XPS 13 DE shows the level of functionality and polish possible with extra effort, and that effort and polish together means this kind of Linux integration is something we won t see very often outside of boutique OEMs.

Of course, Windows is actually worse than Linux on the hardware front when you don t get it pre-installed. Imagine if more vendors put as much effort into preinstalled Linux as they did into preinstalled Windows. In that alternate reality, I imagine people would react more like this:

Isn t that what you re looking for in a mainstream product? Rick chided. In 1996 it was: Wow look at this, I got Linux running on xxxxxxxx. Even in 2006 that was at times an accomplishment When was the last time you turned on an Apple or Windows machine and marveled that it just worked? It should be boring.

Which was, of course, the reaction Hutchinson got when discussing the review with a Linux-using friend. With Microsoft being less of a friend to the hardware vendors every day, here s a case study more of them should be paying attention to.

By now, Debian ships quite a lot of Haskell packages (~600). Because of GHC's ABI volatility, whenever we upload a new version of a library, we have to rebuild all libraries that depend on that. In particular, if we upload a new version of the compiler itself, we have to rebuild all Haskell library packages. So we have to rebuild stuff a lot. Luckily, Debian has a decent autobuilding setup so that I just need to tell it what to rebuild, and the rest happens automatically (including figuring out the actual order to build things).
I was curious how much we use the buildd system compared to other packages, and also how long the builders are busy building Haskell packages. All the data is in a postgresql database on buildd.debian.org, so with some python and javascript code, I can visualize this. The graphs show the number of all uploads by autobuilder on the amd64 architecture, with haskell uploads specially marked, and the second graph does the same for the build time. You can select time ranges and get aggregate statistics for that time span. During the last four days a complete rebuild was happening, due to the upload of GHC 7.6.3. During these 2 days and 18 hours building 537 packages took 48 hours of build time and produced 15kg of CO₂. That is 94% of all uploads and 91% the total build time. The numbers are lower for the whole of last year: 52% of uploads, 31% of build time and 57kg of CO₂. (The CO₂ numbers are very rough estimates.)
Note that amd64 is a bit special, as most packages are uploaded on this architecture by the developers, so no automatic builds are happening. On other architectures have, every upload of a (arch:any) package is built, so the share of Haskell packages will be lower. Unfortunately, at the moment the database does not provide me with a table across all architectures (and I was too lazy to make it configurable yet).