On Mon, Aug 27, 2012 at 02:50:54PM +1000, Ben Finney wrote:
Adam Bolte <abolte@systemsaviour.com> writes:
On Sun, Aug 26, 2012 at 10:04:05AM +1000, Ben Finney wrote:
First you'd need to show that we're treating graphics firmware specially. I think the same criticisms are applied to vendors who act against freedom in network interface firmware, graphics firmware, radio firmware, etc.
Like yourself, instead of purchasing computer hardware from perhaps a local computer retailer, Ben has elected to import hardware from the US on the basis that these Think Penguin machines are tested to be "compatible" with free software - or at least that was my understanding.
Close. They go to that effort in benefit of customer's software freedom, advertise it as a distinguishing feature, and say what devices are in the machine so we can verify exactly what free drivers will be used. That puts them a huge stretch ahead of most Australian notebook vendors in the advance of software freedom for customers.
Interesting. Out of curiosity, how do you feel Pioneer Computers fair? Say, the DreamBook Ultra Power W11. Product page: http://www.pioneercomputers.com.au/products/info.asp?c1=3&c2=166&id=3437 Customisation page: http://www.pioneercomputers.com.au/products/configure.asp?c1=3&c2=166&id=3437 Like Think Penguin, they don't say exactly what kind of SSDs or HDDs are used... they also don't say exactly what the wireless card is (although as Ubuntu GNU/Linux is an option there's a good chance it's free software compatible - and you could likely just call up and ask about the chipset used - and it's cheap and easy to replace anyway. Pretty much all the major components are known and can easily be researched, so one could have a strong degree of confidence. Out of curiosity, do Think Pentuin guarantee that the firmware updater for the SSDs and other devices are not proprietary? Given that the firmware is proprietary anyway, should this be on our radar?
However, even those machines provide the option of various SSDs, HDD&SSD hybrids (all surely requiring non-free firmware), and even non-free BIOSs.
Which makes them no different from Australian vendors in that regard. On the other hand, I know of no Australian vendor that goes anywhere near close to the level of we-guarantee-it-works-with-only-free-software effort and proactive advertising on the basis of software freedom.
If you can find one which goes *even further* than Think Penguin, ZaReason, Gerlach44, and so on, please let us know. Bonus, of course, if they're Australian.
Shouldn't be too hard to make a few calls or send a few e-mails I'd imagine. Surely local stores that already support GNU/Linux installations would like to sell a few extra units at minimal effort. After all, they would just need to throw in a Trisquel disc, make sure everything appears working (perhaps an application is available that could automate much of this), and label it as being compatible. Heck, I'd volunteer a few hours of my time over a weekend each month or so to let a store know if new models meet our requirements - and then we could just point our fingers to that local store when making recommendations. What are your thoughts on something like this? Is there perhaps a better way that should be considered?
Yet, these issues are rarely given any attention. Instead, most efforts seem to be directed towards network and graphics firmware.
I think that's a function of the long-standing intractability of nVidia and ATI on software freedom, and the rather recent advent of SSDs. I agree with you that they are both important issues for software freedom, and I don't treat either of them as less important than the other.
That's a fair call. Graphics drivers have largely been problematic for free software driver authors and users up until the last few years (and they're still generally problematic when buying the very latest models) so that's probably given them some extra attention.
In the case of my SSD, you may recall that I was the one who raised the problems of SSD non-free firmware updates and HDMI+HDCP, at the Free Software Melbourne meeting where I discussed this machine and my efforts in buying it.
I was unaware of either problem when I was researching the machine, so I deny the charge you're bringing that I treated them as somehow lesser problems.
Fair enough. There's potentially a public awareness issue here. We should research free alternatives to traditional storage devices that are more free, which we can recommend to people. While having a USB hub (preferably transparent with lots of flashing lights) and USB key RAID0 array duck taped to the lid of your laptop might get you strange looks, it could be a valid option for people who want to reduce their reliance on proprietary firmwares (particularly with USB3 now readily available). As a bonus, it's guaranteed to gain you geek points. :)
For the most part however, Intel doesn't issue microcode updates. AMD has only enabled users to update microcode since 2009 (on GNU/Linux systems at least). Out of sight, out of mind?
No – as I said, if the vendor is not in a privileged position to deploy updates to the device's behaviour, then the customer's freedom is significantly more secure. That's good reson to treat it as less of a problem.
Maybe I'm not clear after all on what "privileged position to deploy updates" is. Obviously if we're running a free software distribution, we choose what is deployed to our own systems - proprietary or otherwise. If AMD publishes a new microcode update today on their website, I'm the one who decides if I push it to my device - not AMD. So if that's what you meant, fine. However, I thought what you were trying to say was "Can the vendor issue firmware that adds features or functionality to the device as *they* want, whereas I don't have that control due to the proprietary nature of the firmware?". In that case, if Intel can issue updates that improve clock speed, unlock CPU cores, add or remove a unique CPU identification tag, etc, and I cannot, then obviously Intel *is* in a privileged position in that it could utilise the hardware more than I possibly could due to the proprietary nature of the update.
But do these graphic card firmwares really see proprietary updates from vendors that modify the behaviour in some useful way? Or is this something you are assuming just because a firmware needs to be loaded at boot, and proprietary graphics card drivers (which include the firmware) regularly get updates?
Either the firmware update makes a significant change to the device's behaviour, in which case the customer's freedom to choose to load different firmware is important to their freedom; or it's not significant, in which case it's not important.
I see. You still said "firmware update". I don't think the AMD cards actually get firmware *updates*, as per my previous e-mail. It's just the same firmware which is always loaded. Cheers, Adam