Shuttleworth Answers FSF Call for Free Software Drivers on Edge 112
WebMink writes "In an interview at OSCON, Mark Shuttleworth of Canonical spoke about the vision behind the Ubuntu Edge phone as a concept device to test features the mobile industry is too conservative to try. Notably, he agreed with the Free Software Foundation's demands that the device should carry no proprietary software and have Free drivers (transcript): '... we'll ship this with Android and Ubuntu, no plans to put proprietary applications on it. We haven't finalized the silicon selection so we're looking at the next generation silicon from all major vendors. I would like to ship it with all Free drivers.'"
Although not a hard promise, it is a promising development.
Re: Subtext is.... (Score:2)
Earning money in the billions, you mean? I'd bet canonical wouldn't mind such a mistake.
Re: (Score:1)
I know a lot of Ubuntu Fanboi's [sic] are foaming at the mouth at the prospect of getting their darling O/S on a phone.
As far as i'm concerned, the boot is on the other foot. I can definitely live without Ubuntu, but the hardware they show in TFA is (IMO) very cool indeed, and I would happily have such a device. Now if only it could run Slackware... :-P
Re: (Score:2)
Canonical blew it w/ Unity on the desktop, but just as Metro is a fine UI for a phone, so is Unity, by the look of it. So Canonical can do well in the phone market, and depending on the hardware vendors willing to sign up w/ them, they may well have a shot
I do see a downside for them, or anyone else who wants to offer an Open/Free phone, at least in the US. The carriers. Since in the US, the phones are subsidized by the carriers, which is why the most expensive of phones can be had pretty cheaply due t
You can't make promises... (Score:2)
Re: (Score:3)
When you know you depend on silicon designed by others. Here's the thing.I bet Canonical would very much rather have everything on that thing be open-source because if something breaks it's way easier to debug than having to bang your head against the wall that a binary blob of anything represents.
It won't happen. Minimally, the SDR (Software Defined Radio) will be required by the FCC, and other similar regulatory agencies around the world, to have a locked down image, or it won't be licensed for use, period. An SDR is defined to be a combination of the software and hardware, and you can't change one or the other without getting the thing relicensed, or requialified for use on the carrier network in the country in question.
My guess is that they will end up with some flavor of Qualcomm Snapdragon, w
Re: (Score:2)
Re: (Score:2)
The problem is that if someone actually manage do to this I belive Canonical is still responsible since they made the hardware.
Re: (Score:2)
Re: (Score:2)
So, if I solder a new crystal into an old CB radio and add a booster, is the manufacturer responsible according to the FCC?
if someone gets out the soldering iron and changes the firmware on any of the thousands of cellphone models out there, is the manufacturer responsible for that?
Re: You can't make promises... (Score:2)
That means they'd better don't license it under GPLv3, not that it can't be open sourced. Granting support/compliancy to a given combination of hardware/software is business-as-usual so I don't really see what your point really is (unless FUD, that is)
Re: (Score:2)
Re: (Score:2)
People running Android and custom ROMs frequently replace the radios. This is probably not quite the same as it's mostly just using radios from other models or with minor changes I believe, but it is done. The equivalent for hardware radios would be like saying you have to release it in a steel box or something. People hack things and have since radios were invented. It's really the responsibility of the person who owns the device not to break the law.
Re: (Score:3)
People running Android and custom ROMs frequently replace the radios. This is probably not quite the same as it's mostly just using radios from other models or with minor changes I believe, but it is done.
I assume you are mostly talking about tablets, not phones, here, since the modules are generally surface mount in phones, due to industrial design requirements that phones be relatively small. It's grey market at best to replace a GSM or CDMA module. WiFi modules, the FCC cares a lot less about. If you replace a GSM or CDMA module with another, however, you have to also replace the binary firmware blob with one whose signature the module can verify when you try to load it. So it's effectively a single p
Re: (Score:2)
he meant the radio firmware. And it's done on cellphones all the time. Often it is an essential step to unlocking the phone.
Re: (Score:2)
The FCC aren't acting like fucking savages, having certified SDR equipment--with a modifiable software software component--since 2004.
http://transition.fcc.gov/pshs/techtopics/techtopics4.html [fcc.gov]
http://hraunfoss.fcc.gov/edocs_public/attachmatch/DOC-254463A1.doc [fcc.gov]
Re: (Score:2)
I take that back. On further consideration (of about 10 seconds) it looks like you're right--the combination of hardware and software is what's certified.
Comment removed (Score:5, Informative)
Re: (Score:2)
Remember even the OLPC 1 couldn't get 100% free because they couldn't find a BIOS and wireless chipset at the time that wasn't proprietary and that was with X86, with ARM its even worse.
Significant deployments of OLPC 1 outside of Spanish speaking South America can be counted on the fingers of one hand. Participating countries [wikipedia.org] OLPC was a product of the Western media lab and it had problems "free" software and hardware could not solve.
Re: (Score:2)
Re: (Score:2)
Well, they could, instead of ARM, use a CPU like OpenRISC and get around the proprietary aspect of the CPU. Hell, if they wanted, they could make an FPGA based GPU, as you mentioned, and have all the patented operations implemented in a HAL on top of things, so that the hardware runs free of violating any patents, and the software that runs is written from scratch, and neatly sidesteps patents. Put it under a GPL3 license, and fight that battle there. The main issue I see here is having a semi company th
Re: (Score:2)
Re: (Score:2)
And as my grandpa used to say "Girls want ponies, people in hell want ice water, I want a million dollars...that don't mean any of us are gonna get it".
Unless they are gonna kickstarter the chips in the thing it'll be DAMN hard to make it FOSS, simply because the ones making the GPUs, wireless, etc, are about the most proprietary lot on the planet. Hell I don't even think you CAN make a FOSS GPU as everything from texture compression on up is patented up the ass, I know there was a project to make one using an FPGA but I never heard any more about it, probably ran into the legal minefield and ran aground.
Basically, it ran out of money; the main contributors didn't have as much time available any more and making an ASIC is expensive. Some prototype boards were manufactured, and the employer of the main developer (who allowed him to use their tools, and work on it some during office hours) made a commercial product based on the design. It never got to producing a consumer video card though. I see now that Kickstarter actually existed in 2010, but I don't think anyone of us had ever heard of it, and I don't th
Re: (Score:2)
Making an ASIC? Isn't that something you only do when there are volume ramps, and you want to cut costs by volumes? It doesn't make sense to make ASICs when your market is not up as yet - vendors start w/ FPGAs, get them to ramp (and in the process, check out if any bugs are discovered while out there) and then freeze them on ASICs.
Well, this couldn't be a hobbyist's project, a company would have to throw all its weight behind it. The large ones would probably be patent owners, and not want to put thei
Re: (Score:2)
So? Why do we need a GPU? The goal is fully open har
Re: (Score:2)
Just hit $8M at the fundraiser (Score:2)
Re: (Score:1)
so crowd funding is only for poor people?
Re: (Score:2)
No, it is for stupid people.
Counterproductive Fixation (Score:2)
Re: (Score:1)
The things that are needed from hardware are specifications for the interfaces and/or free drivers, and possibly the keys that unlock any binary signing functions. If you have free drivers to hardware, it's not hard (relatively speaking) to modify the drivers to work on a different OS because you can get the specification from studying the drivers.
I wouldn't expect it to happen. (Score:2)
You'd like wifi right? While there are wifi adapters that have free /open drivers, not many are in the ultra-low-power-cost SystemOnChip wifi adapters. Likewise the drivers for the telco data side of things are unlikely to be open/free, especially for sprint and Verizon in the US, can't speak for overseas.
I'm pretty sure that mark would like them to be free too. That doesn't mean that it's going to, or is likely to, happen.
Firmware vs. Software (Score:2)
Again many accept without discussion that Firmware is proprietary but at the same time they demand the software to be open, I am not so much against having a few proprietary blobs for drivers and things like the SDR, all depending on a well defined interface with said blob or firmware.
There were great hopes around the Nokia N900 development even though it too had it's closed sections.
What I DO wa
Yeah yeah, bullshit someone else (Score:2)
There are 19 days left, if you don't signup before that deadline you lost your chance. NOBODY else will try AND NOBODY else will come even close to giving you what you want. Get over your sense of entitlement that EVERYTHING need to be just the way you want it. This is the real world and your mommy ain't around to cut the crusts of your bread.
If this project does not exist, the message is clear, the linux world is to divided for anyone to cater for because no matter what you do, they always want more and n
Re: (Score:2)
Mark's remarks from the original article get very close to my and many other's wishes but more clarification is needed.
Re: (Score:2)
Get over your sense of entitlement that EVERYTHING need to be just the way you want it.
How is it a sense of entitlement to voice your opinion or vote with your wallet? No one's holding a gun to anyone's head. The word "entitlement" just seems to be a buzzword these days.
Re: (Score:2)
I think my message is very clear, I am prepared to spend good money for a a phone with a regular, thus open, GNU/Linux system and that's much more than a Linux kernel with Android or so.
If a company is brave enough to sell such hardware they can do it either with or without an OS, if the HW is wo
Due Diligence (Score:3)
Does anyone know where duedil.com gets its Canonical data from?
If I am going to fork over $800, I want to perform at least some due diligence. Is Canonical simply going to use my money to pay downs its current liabilities, which were recently about 19,000,000 GBP higher than its current assets?
https://www.duedil.com/company/06870835/canonical-group-limited [duedil.com]
I hope they'll launch it in Europe (Score:2)
I would vote for such a development with my wallet.
Then learn to read (Score:4, Informative)
Re: (Score:2)
Thanks for the info!
What this sounds like to me... (Score:3)
It sounds like the Ubuntu Mobile people are saying "hey, we want to ship this with no binary blobs but we recognize that in order to get certain features such as a cellular modem or a 3D-capable GPU we may have no choice but to go with a binary blob if we cant find hardware that is 100% open"
Google apps (Score:1)
That means no Google apps by default. I'm guessing they'll manage it the same way as flash and mp3 support on Ubuntu desktop -- offer a way to install the troublesome applications quickly and with no fuss.
May not happen whether he wants it or not (Score:2)
more empty words from canonical (Score:1)
From TFA: "An open letter from the Free Software Foundation asked if Canonical intended to make all the software in Edge free -- with emphasis on device drivers. The response: Of course, as far as that's commercially realistic."
Surely, nobody in this crowd, or at the FSF, is the slightest bit fooled by this.
Re: (Score:2, Informative)
When the article says "free", it means free as in "free speech", not in "free beer".
Re: (Score:2)
I am a big critic of RMS/FSF, but here, 'free' means freedom to do things w/ your own toy. It means 'free' as in being able to pull in apps from anywhere, not just a single store. It means 'free' as in having drivers include their source, so that one who knows can check on backdoors and the like. It means 'free' as not being tied to any single hardware platform, so that if a phone can be unlocked, it can be installed on anything - be it a Nokia Lumia, Samsung Galaxy, even an iPhone.
I think Ubuntu has i
Re:LET US DO EVERYTHING - FOR FREE !! (Score:4, Informative)
And... (Score:5, Interesting)
Re: (Score:1)
Re: (Score:3)
Re: (Score:1)
Re: (Score:2)
While amusing and worth thinking about, it's unlikely to be a practical concern in this case. I'll one up you and say you have to trust your brain first. Fnord.
Re: (Score:1)
Re: (Score:2)
Sure, right up until your calls/data hit your ISP.
Re: (Score:2)
Assuming that you can trust all that NSA (SELinux) code that's in Linux. Fun fact: RHEL is preferred Linux operating system of surveillance state! Look it up.
well, you can always review the code if you want/need to.
Re: (Score:2)
Or remove SELinux.
Re: (Score:2)
Re: (Score:2)
Assuming you have the expertise and resources to vet every single commit made to the source tree. The fact that the code could be audited doesn't mean that it is, and therefore is no protection against a well-funded state-sponsored attack to insert subtle weaknesses in the code that could be exploited by a sophisticated methodology known only to the organization that created the patches.
This idea was recently covered by Poul-Henning Kamp (FreeBSD dev) in his essay, "More Encryption Is Not the Solution"
Sure, but at least you have that option, you can outsource the expertise if you really want to. With proprietary software, you stuck with the binary as your only option, even when you have the resource to review the code
Re: (Score:1)
The objecting of the FSF isn't forcing developers to share improvements, the objective is promoting user freedom. If developers do not distribute software, they aren't required to share anything.
Re: (Score:2, Insightful)
Only because they can't be. Don't confuse what a license doesn't do with with what it can't do.
RMS's motivation was always to get access to other people's work. It angered him that he was denied source to software that wasn't his but was free to use on a device he didn't pay for. The GPL IS about forcing others to "share", there are simply limits to its reach.
Now, it would seem that "user freedom" is somehow nicer that "forcing developers" but it is not. GPLv3 came about not because GPLv2 software isn'
Re:LET US DO EVERYTHING - FOR FREE !! (Score:5, Insightful)
Again the tired old debate about "which is more free", GPL2 vs GPL3, GPL vs MIT etc. I'm amazed at how people keep falling for linguistic traps. "Freedom" isn't subject to gradation on a linear scale, necessarily marred by increased regulation nor evenly distributed. As a concept, it's as ill-defined as "love", so arguing about what license is "more free" doesn't make a lot of sense, unless you also fall for the cultural trap that "freedom" is the main moral goal in everything and a necessary attribute/buzzword for garnering support regardless of the issue at hand.
Having said that, I believe the GPL is better because it guarantees the possibility of forking.
Re: (Score:2)
Again the tired old debate about "which is more free",
It never gets tired. Oh and vi > emacs, FVWM > whatever you use, Picard > Kirk and Arch > Ubuntu. Any others? :)
Re: (Score:2)
Re: (Score:1)
Re: (Score:2)
I'm amazed at how people keep falling for linguistic traps.
It's not a trap. It's a direct response to 1984-style propaganda.
unless you also fall for the cultural trap that "freedom" is the main moral goal in everything and a necessary attribute/buzzword for garnering support regardless of the issue at hand
Who chose the name "freedom" to describe being forced to release source code? Stallman did. What Stallman is advocating is more in line with consumer protection laws than freedom.
Re:LET US DO EVERYTHING - FOR FREE !! (Score:5, Insightful)
Can't believe this anti-GPL flamebait/troll has been modded up.
It's simple: If you don't like GPL (2 or 3 or whatever), don't use/modify/distribute the software, fuck off and write your own.
Those of us who think that GPL (2 and 3) is a guarantee of the freedoms we are interested in will use/modify/distribute GPL software.
Basically, you think the freedom to restrict others freedoms is essential. And that's fine. But don't try to pretend your view is more pro-freedom.
Re: (Score:1)
RMS's motivation is that the USERS (and their community) have control over their own technology; the point of the FSF freedom is promoting freedom for the USER; the point of USER freedom is that the USERS control their own computing which means the developers must not restrict the users by withholding source code (or hardware lockout keys) or restrict users to distribute the software. The balance of power is promoted as the USER who controls the software. The printer that he had was university property and
Re: (Score:2)
As far has him wanting people to call linux OS GNU/Linux that's not going to happen. however i think the proper name should actually be just simply "GNU" since that is the operating system and Linux is jus
Re: (Score:2)
GPLv3 came about not because GPLv2 software isn't free but because RMS wanted to further leverage the license to restrict what hardware vendors do with your property.
Fixed that for ya'.
Re: (Score:2)
How is that even the case? TiVo makes a set top box where the flash is locked after the firmware is installed, but as per GPL2, TiVo publishes the sources - not that it's of any use to anybody since it can't be altered. But TiVo made this before selling it to the content provider: it's usually the content provider who includes this in their package. So you get the thing already with the flash locked. On a rare occasion that the content provider might want to upgrade something on the STB, they access you
Re: (Score:2)
How is that even the case? TiVo makes a set top box where the flash is locked after the firmware is installed, but as per GPL2, TiVo publishes the sources - not that it's of any use to anybody since it can't be altered.
Exactly the problem GPL3 was intended to address.
But TiVo made this before selling it to the content provider: it's usually the content provider who includes this in their package. So you get the thing already with the flash locked. On a rare occasion that the content provider might want to upgrade something on the STB, they access your box, do it and then resume operations. Most people neither know nor care. Those who do probably got MythTV or something of that sort.
Public outrage against TiVo-ization predates the integration of TiVo into cable/satellite boxes. I had the original TiVo. It was marketed direct to consumers. No "content provider" involved. It was a stand-alone box. They didn't start integrating it into cable/satellite boxes until years later. It was my hardware, not their hardware. The device phoned-home (quite literally, through the integrated modem) every couple of days to check for new listings and s
Re: (Score:2)
Well, the nice thing is that if you don't want to use GPLed software, you don't have to.
Sounds like there are people on both sides of the equation who want to "access other people's work". The difference may be that the GPL facilitates that access in both directions.
Re: (Score:2)
Yes, it's about forcing developers to share and share alike. How is RMS more equal? Is there a secret directive in the GPL that says RMS doesn't harve to share? Or perhaps that robocop can't arrest him?
Re: (Score:1)
Re: (Score:2)
Your post is not accurate in any sense. The GPL exists to stop software written as an open collaboration being privatised and turned into close software. GPLv3 exists because people were cheating with GPLv2 software.
Re: (Score:2)
It sounds more like a attack on RMS than on the license.
Re: (Score:3)
I make GPL software. I want to help my neighbors, even you. You may not want to, or you may have other reasons for not making GPL software. More power to you. I respect your right, and I hope you respect my right to publish my software under the license I choose.
You can take my GPL software and use it freely and copy it to all the computers at your office. You can make changes to my software and you are not forced to publish the changes as long as you keep the changes within your
Re: (Score:2)
You miss a major problem w/ the GPL, which I've described a number of times. That it's a terrible shared-source license.
Here is a hypothetical. I wrote an HDL CAD software, which I sell to a number of my customers, who are in the business of designing ICs. I decide that I want to share my code downstream, so that the engineers of my customers know what I've done, and that my designers don't have to spend lots of time with them: maybe, my customer support people do. As a result of my sharing my work, m
Re: (Score:2)
If I don't accept your license, I will not use your code. If you don't accept GPL, nobody forces you, or your clients, to use my code. As I said I am sure there are many commercial options.
About your example. Assuming that you used my GPL software, you would be just one of my client
Re: (Score:2)
Actually, GPL doesn't allow me to forbid them from re-distributing - it simply allows them to not re-distribute. As you said, I'd not use GPL. Or if I had to combine it w/ yours, I'd put other terms & conditions in which you could redistribute it, such as sell it at prices similar to mine, we agree on how to split the market, etc. But no way would I be okay w/ you taking my changes and just giving them away, since it would undercut me. But such an agreement/arrangement would have to happen outside t
Re: (Score:1)
No, "free" in this context usually means copyrighted and protected under a license like the GPL (or something similar). The "free" that the FSF endorses is actually more restrictive than public domain, with the objective of forcing developers to share their improvements on the code.
And yet, when compared to my prospects for distributing "improvements" on Windows or OS X (which would get me sued out of existence by Microsoft or Apple, respectively), the GPL seems downright generous.
Re: (Score:2)
Which isn't a bad thing, they want to give users freedom even if that means developers might have slightly less freedom.
Also it does not mean you have to share your changes with the world, just the people you give (or sell) your program or code to.
It really isn't that bad.
Re: (Score:2)
You may not have to share your changes w/ the world, but if the people you gave/sold to want to do it, there is absolutely nothing you can do to stop them.
Other than not putting it under GPL in the first place
Re: (Score:2)
Free does not require public domain.
Re: (Score:1)
By this logic, you abstain entirely from GPL'd software. Such as, for instance, "ls" from GNU coreutils.
But good for you, standing up so resolutely for your principles.