- luv-main - luv.asn.au

Re: [luv-main] Microsoft requires Windows 8 logo systems to not bootunsigned OS's
by Andrew McNaughton 29 Sep '11

29 Sep '11

On 23/09/11 14:37, Daniel Pittman wrote: > On Thu, Sep 22, 2011 at 21:11, Jason White <jason(a)jasonjgw.net> wrote: >> Daniel Pittman <daniel(a)rimspace.net> wrote: >>> So, the biggest advantage is that it does work against all those >>> attacks that compromise the kernel and/or drivers to get into the >>> kernel after a restart. Which, indeed, is where many of the "root >>> kit" tools hit, on Windows. >> That's interesting... and the proposed solution just happens to be the one >> which also has potential to disadvantage competitors... > Well, assuming that the vendors are lazy, yeah. Which /probably/th > that the vendors will have an option to turn it off in the > configuration whatever, so you can install whatever on it. Because > Linux is enough of a server OS, these days, that the low margin, high > volume vendors probably don't want to go down the path of cutting off > part of their business, and producing two different EFI > implementations would cut into that. > > Allowing loading another key, probably screwing up any security > assurance for !Win32 systems in the process, feels like the cheap > option to me. I think that's exactly what's needed. People should read up on the secure boot process associated with the TPM chips that are in most recent computers, but typically not turned on. If there's a user process to load a new key into the TPM chip, then I don't see a problem with Windows and other OSes being secured in this way. Rather than boot sector viruses, people should probably be thinking about the requirements for full disk encryption to be effective, which is something we want. This basically requires that any non encrypted boot mechanisms must be secured in some fashion, and that's what the TPM implements these checksums for. The other alternative is implementing hardware encryption inside the disk drive itself (as in the hitachi drive that came with my Lenovo x220). This latter is a more expensive option though, and requires BIOS/UEFI support to ask for the password. I'm not clear on how many systems have that. My x220 is a UEFI based system, the TPM can be enabled and disabled from within the bios, and that's protected by the BIOS password. The system ships with the TPM disabled, and so far I haven't played with it. If it shipped with the TPM enabled, I wouldn't have thought that would prevent me installing Linux. I might have to disable it for a bit while I get things set up. I haven't yet gone very far down this road, but I have briefly looked over a 2008 howto at http://www.grounation.org/index.php?post/2008/07/04/8-how-to-use-a-tpm-with… which includes some details on how one loads a key into the TPM using linux. It's on my to-do list. So far I've just read enough to know that I can do it when I'm ready. So I'm not very worried about booting linux becoming an impossibility. If however hardware manufacturers were to ship systems with TPM enabled and no way to turn it off, or with no way for a user to load their own key, then that would be a different matter. Does the system that Microsoft is mandating necessarily include such user controls in order to be compliant? If not, then that is probably the point we should be aiming any legal efforts at. And perhaps there should be some requirement that if any system is restricted by the keys and ability to change them, such that it is only able to run windows, then that they must advertise that they are *only* compliant with windows 8, rather than just that they *are* compliant with windows 8. I am concerned that it might be harder to boot from a Linux CD in order to install. Not massively harder - it should require only that the TPM be disabled for a while. This will have some impact however on less confident installers. I'd like it to be possible to boot from a CD or USB for anyone who has the BIOS/UEFI password, without making any changes that would affect the subsequent boot. Anti-virus disks are another important one for booting outside the usual OS. I am a bit concerned that I probably won't be able to boot linux from the flash disk on my keyring on most systems in future. It's currently quite convenient. Andrew McNaughton

1 0

IEEE SSIT: Sat 1 Oct 11:40-1:10 Panel: Climate Change and Engineering responsibility
by Rodney Brown 29 Sep '11

29 Sep '11

Melbourne Saturday 1 October: Climate Change and Engineering responsibility 11.40am to 1.10pm, cnr Victoria & Lygon Street, Carlton South A panel at the Climate Change Social Change conference sponsored by University of Melbourne Office for Environmental Programs: http://climatechangesocialchange2011.wordpress.com/. Per day registration $35. Speakers: 1. Dr Michael Arnold, History and Philosophy of Science, University of Melbourne: An introduction to the interaction between technology and society. 2. Reihana Mohideen, electrical engineer and social development specialist advicing on "gender and energy" in energy sector projects in Asia: Practical experiences of engineering in the field, looking at the social impact of renewal energy projects. 3. Dr Greg Adamson, Department of Information Systems, University of Melbourne: Are technology approaches to climate change a distraction from addressing an underlying problem? 4. Sophie McKenzie, School of IT, Deakin University: How engineers can be encouraged to think about the wider context of the work they are undertaking. IEEE Society on Social Implications of Technology (Australian Chapter) [SSIT technical meetings address a wide range of topics on which IEEE as a whole has no specific view. We don't suggest that IEEE members should agree with all the speakers we invite, but we do hope that they are interesting.]

4 3

stop/start info in /etc/init.d
by James Harper 29 Sep '11

29 Sep '11

I want to make sure xend stops after corosync etc so that the cluster can take care of migrating the domains off of the machine when it reboots, but I don't want to modify the init.d files directly in case they get overwritten on an upgrade. Is there a place I can put this information to make sure the scripts get processed in the right order without actually changing the scripts themselves? Thanks James

3 2

Re: [luv-main] stop/start info in /etc/init.d
by James Harper 29 Sep '11

29 Sep '11

Hmmm... haven't tested yet but I think I can put my new headers in /etc/insserv/overrides/ > -----Original Message----- > From: luv-main-bounces(a)lists.luv.asn.au [mailto:luv-main- > bounces(a)lists.luv.asn.au] On Behalf Of James Harper > Sent: Thursday, 29 September 2011 11:34 AM > To: luv-main(a)lists.luv.asn.au > Subject: [luv-main] stop/start info in /etc/init.d > > I want to make sure xend stops after corosync etc so that the cluster > can take care of migrating the domains off of the machine when it > reboots, but I don't want to modify the init.d files directly in case > they get overwritten on an upgrade. > > Is there a place I can put this information to make sure the scripts get > processed in the right order without actually changing the scripts > themselves? > > Thanks > > James > _______________________________________________ > luv-main mailing list > luv-main(a)lists.luv.asn.au > http://lists.luv.asn.au/listinfo/luv-main

2 1

Re: [luv-main] Microsoft requires Windows 8 logo systems to not bootunsigned OS's
by Andrew Voumard 29 Sep '11

29 Sep '11

FYI, slashdot has picked up the issue: Australian Users Petitioning Against Windows 8 Secure Boot http://yro.slashdot.org/story/11/09/28/1311220/australian-users-petitioning… Andrew At 06:07 PM 28/09/2011, Luke Martinez wrote: >lol, There is *no* discussion i can find going on at linux-aus >except that a small thread on the post that I was reposting everywhere. :) > > >> Linux Australia have put a more positive spin on it.. > >> > >> > <http://www.zdnet.com.au/linux-aus-may-have-a-case-on-win-8-accc-339323233.h…>http://www.zdnet.com.au/linux-aus-may-have-a-case-on-win-8-accc-339323233.h… > > >> >on another note, it seems that ZDnet contacted Linux Australia >president John Ferlito. Hopefully they can continue to report on >this and raise awareness. > > Luke > >_______________________________________________ >luv-main mailing list >luv-main(a)lists.luv.asn.au >http://lists.luv.asn.au/listinfo/luv-main

1 0

Re: [luv-main] Microsoft requires Windows 8 logo systems to not bootunsigned OS's
by Luke Martinez 28 Sep '11

28 Sep '11

lol, There is *no* discussion i can find going on at linux-aus except that a small thread on the post that I was reposting everywhere. :) >> Linux Australia have put a more positive spin on it.. >>> >> >>> >> >>> http://www.zdnet.com.au/linux-aus-may-have-a-case-on-win-8-accc-339323233.h… >>> >> >> on another note, it seems that ZDnet contacted Linux Australia president John Ferlito. Hopefully they can continue to report on this and raise awareness. Luke

1 0

Re: [luv-main] Microsoft requires Windows 8 logo systems to not bootunsigned OS's
by stripes theotoky 28 Sep '11

28 Sep '11

Appears that Competition commission are useless as usual. Long response that says nothing and shows that they don't even understand the issue. Funny they have written this rubbish to a Professor of Economics, if any of my students had this little understanding I'd fail them out of hand. ----------------------------------------------------------------------------------------------------------- Dear Professor Poyogo-Theotoky, Thank you for your email of 26 September 2011 to the Australian Competition & Consumer Commission (the ACCC) regarding the Microsoft’s upcoming Windows 8 operating system. Your reference number for this matter is 1154451. The role of the ACCC is to ensure compliance with the Competition and Consumer Act 2010 (Cth) (the Act), which is designed to encourage fair trading and discourage anti-competitive conduct through a specific set of competition and consumer protection rules. One of the Act’s aims is to foster fair markets, that is, markets where normal competition can continue without being hindered by unfair and illegal market practices. Such illegal market practices include price-fixing, market-sharing, resale price maintenance, misuse of market power and certain forms of exclusive dealings/boycotts. Section 47 of the Act prohibits exclusive dealing. Broadly speaking, exclusive dealing occurs when one person trading with another imposes some restrictions on the other’s freedom to choose with whom, in what, or where they deal. Exclusive dealing is only a breach of the Act where the conduct has the purpose, effect or likely effect of substantially lessening competition in the market. In an assessment of the effect of the conduct on competition, it is not enough merely to show that an individual business has been damaged. The wider market for the particular product or service must be considered. The situation you described may raise issues of exclusive dealing, but it is unclear from the details provided whether it would be likely to meet the competition test described. In assessing any complaint, staff of the ACCC would generally determine whether or not the matter falls within the jurisdiction of the Act, whether or not there appears to have been a breach of the Act, and if so, whether the impact of the conduct is so serious and widespread that it is appropriate that the ACCC should take some action. It should first be noted that in general, investigations are conducted confidentially and the ACCC does not comment on matters it may be investigating. Further, complainants will only be contacted by the ACCC where clarification or additional information is sought. It is important to note that the ACCC cannot pursue all the complaints it receives. While all complaints are carefully considered, the ACCC must exercise its discretion to direct resources to the investigation and resolution of matters that provide the greatest overall benefit for consumers and businesses. The ACCC’s Compliance and Enforcement policy describes in more detail how this discretion is exercised. This policy, which is available on the ACCC’s website [www.accc.gov.au] lists a number of factors that are weighed including whether conduct raises national or international issues, involves significant consumer detriment or a blatant disregard of the law. The Act also allows an affected party to take their own legal action for a breach of the Act. You may wish to seek legal advice on the possibility of taking your own action in this circumstance. Thank you for contacting the ACCC with your concerns. I trust this information is of use. Yours sincerely, Katy ACCC Infocentre Ph: 1300 302 502

6 8

OT Python
by Roger 28 Sep '11

28 Sep '11

My daughter teaches python 3 in her software development classes. The school laptops are fixed on windows 7 (no flames please she has no choice, the students all use linux but the school will not). She wants to teach her students to use a GUI developer like Glade but has not found a way to install into her school laptop. Also it seems that gtk and glade are reliant on python 2.6 and 2.7 Does anyone know of modern tutorials that she can use to install this or if not has anyone solved the install problems. Help appreciated thanks Roger

7 10

Re: [luv-main] Purchasing a new laptop
by Russell Coker 28 Sep '11

28 Sep '11

On Tue, 27 Sep 2011, Daniel Pittman <daniel(a)rimspace.net> wrote: > Yes, but do you care? The day to day performance difference > approximates zero, and it has done for several years, for most > practical purposes. You can tell the difference when it comes to > compiling software in languages with highly efficient compilers (eg: > not, generally, C or C++), and when doing extremely CPU intensive > operations (3D rendering, encoding), but desktop stuff? The main performance issue I have on DESKTOP tasks (as opposed to compiling, video processing, etc) is web browsing. For that the difference between a Pentium-M 1.7GHz (like a P3) and a dual-core 64bit CPU wasn't that obvious - Mozilla is slow everywhere. Supporting more than 4G of RAM is a real performance benefit for modern desktop software, something that is a problem with my latest desktop system that is limited to 3300M of RAM (nasty Intel). For my desktop stuff moving from Mozilla to Chrome was a much better performance boost than moving from 32bit to dual-core 64bit. -- My Main Blog http://etbe.coker.com.au/ My Documents Blog http://doc.coker.com.au/

5 15

Re: [luv-main] Microsoft requires Windows 8 logo systems to not bootunsigned OS's
by Rick Moen 28 Sep '11

28 Sep '11

Quoting Daniel Pittman (daniel(a)rimspace.net) > They note the specific part of their role that this hits, what the > larger conditions are, and identify that there might be an issue under > those terms. > > They are not, obviously, leaping directly into an investigation today, > but that seems reasonable: the issue came to light weeks ago, the > product is not on the market, and generally governments (for better or > worse) prefer a hands-off approach to allow businesses to sort this > out. > > So, you win: they agree that there might be a restriction on > competition in the market, and that might deserve investigation later, > when the details are clear. Moreover, they impliedly invite someone like Prof. Poyogo-Theotoky to detail for them how the substantial lessening of competition is occurring. (Obviously, such an explanation will be much more effective after MS-Window 8 ships and specific real-world OEM details can be cited.)

1 0