Ubuntu Phone Isn't Important Enough To Demand an Open Source Baseband

colinneagle (2544914) writes "Canonical is producing a version of the Ubuntu Linux distribution specifically for smartphones, but Richard Tynan, writing for PrivacyInternational.org, recently pointed out that the baseband in Ubuntu-powered phones will remain proprietary. ... Some have criticized Canonical for missing an opportunity to push for a fully Open Source smartphone, but in order to fix this problem (and open up the code for this super-critical bit of software), we need companies that have a large amount of clout, in the smartphone market, to make it a priority. Canonical (with Ubuntu) just doesn't have that clout yet. They're just now dipping their toes into the smartphone waters. But you know who does have that clout? Google.

Google has made a point of touting Open Source (at least sometimes), and they are the undisputed king of the smartphone operating system world. And yet I hear no big moves by Google to encourage phone manufacturers to utilize Open Source baseband firmware, such as OsmocomBB. So has Canonical missed an opportunity? No. Not yet. If (some may say 'when') Ubuntu gains a critical amount of market share in the phone world, that will be their chance to pressure manufacturers to produce a truly Open Source phone. Until then, Canonical needs to continue to work within the world we have today."

Comments

They're not important even on the desktop

[Anonymous Coward]

If (some may say 'when') Ubuntu gains a critical amount of market share in the phone world, that will be their chance to pressure manufacturers to produce a truly Open Source phone.

Such naivety. Ubuntu never even gained a significant amount of market share in the regular desktop/laptop world. Google, which supposedly uses its own spin of Ubuntu internally, recently released Google Now for Windows and Mac but didn't bother with Linux. The Google Drive client has been out for Windows and Mac for ages and yet a client for Linux is "still in development".

The only people who think it's a case of 'when' Ubuntu gains a critical amount of market share in the phone world are fanboys and kids who are still too idealistic and don't want to face the mountain of evidence which suggests Canonical don't know how to make something make money. They still haven't even turned a profit in close to 10 years for fucks sake! I'm sorry if this sounds like trolling but I'm sick and tired of Linux fans having their heads in the clouds and not being down to earth like I was expecting them to be.

Google? Not very likely

[Inev]

Google, these days, is interested in making as much of the Android ecosystem closed source as possible in order to exert control over it and manufacturers. So I don't see them wanting to open source something important like the baseband firmware. Source: http://arstechnica.com/gadgets...

Re:Google? Not very likely

[AmiMoJo]

Uh, you know Google don't write or control the baseband firmware, right? It is provided by the manufacturer of the system-on-chip that includes the baseband processor. Qualcomm, Samsung, Alcatel, Huawei or whoever they are using.

It's those guys keeping it closed source. They do it because they don't want to give their competitors free hints, they don't want their competitors examining their source code for patent infringements, they don't want black hats examining the code for vulnerabilities and they can't be bothered to tidy it up enough to be publicly presentable.

Re:Google? Not very likely

[Andy Dodd]

Keep in mind that Qualcomm has almost total dominance of the LTE modem market and they want to keep it that way.

Even massive pressure from Google won't work here... Maintaining their lead in baseband chipsets (which is heavily dependent on their modem firmware being as difficult to RE as possible) is EXTREMELY important to Qualcomm. Losing dominance of the LTE market will hurt their cash flow there, and also their ability to keep using it to sell complete SoCs. (It's only recently with Krait that Qualcomm's SoCs were able to stand on their own and obtain design wins without pairing to a Qualcomm modem. The old Scorpion cores in the Snapdragon S3 family kind of sucked.)

Re:Ubuntu isn't important enough for anything

[oodaloop]

Mint probably cares about Ubuntu, since it's a fork. Or was that supposed to be funny?

They need to fix Touch first.

[slackware 3.6]

How can an Ubuntu phone be taken serious when Touch is horribly buggy and hardly runs on anything. I tested it on my Nexus 7 and it was basically unusable it doesn't even have a working memory management. I would probable have better luck with Haiku. O well back to CM, at least it works and has a cool animation at startup. Which is all that matters if you want to look cooler and nerdier that the other guy reading his tablet on the toilet.

Re:They need to fix Touch first.

[drinkypoo]

Uh, it's a VM running in a VM. And it's beta.

So what you're saying is that Ubuntu Phone isn't real enough to be a phone.

You're right, let's just scrap the entire idea because....Google. NO FURTHER QUESTIONS.

I would like to use Android on a PC, but I have the same reservations regarding Android-x86. They keep kicking out new releases, but they never actually finish one. So there's plenty of fail in the Android ecosystem, and everyone [who cares] knows it.

Irrational open source fanboys

[BitZtream]

What part of you thinks a company with half a clue is going to 'open up' their special magic so that companies like Huawei can rape them in the process?

NO smartphone maker gives enough of a shit about Ubuntu for them to have ANY chance at ALL of opening the baseband.

You guys live in a really silly fantasy world where people seem to hurt themselves to benefit you, that doesn't actually happen and you have provided absolutely no reason they should actually open the baseband. You're a statistically insignificant group of people demanding something silly that no one else on the planet gives a shit about because you think no one else's way is possibly acceptable, you're way or the high way ...

Theres no compelling reason for any one to open their baseband.

Demanding OSS firmware is fucking silly anyway, the chips aren't open, you can't actually see whats going on in them and you never will because again, theres NO REASON to compel them to do so.

Your monitor/display ... not OSS. Your computer hardware ... not OSS pretty much the only OSS in your life is the tiny little bit you run on your desktop, which is itself in no way what so ever open ...

You guys pick some random silly shit to rant about and miss far bigger issues. It really makes you look silly and unaware of what you're talking about when you rant on about OSS baseband on a proprietary chip.

Re:Irrational open source fanboys

[TubeSteak]

[...] theres NO REASON to compel them to do so.

How about to make sure there isn't a backdoor in the baseband software?
https://www.fsf.org/blogs/community/replicant-developers-find-and-close-samsung-galaxy-backdoor

The NSA's activities should have us rushing to audit and open as much as possible.
"Trust us" isn't a viable business model anymore.

Re:Irrational open source fanboys

[exomondo]

He may be condescending and dismissive but I do see his point, this same formula has been repeated for years (first in desktops) with almost no successful results because it is flawed. Open source rarely gets used by end users because it's open source, it gets used because it is a more compelling product for one reason or another. Saying "you should open this up" inevitably leads to the question of "why" and if you cannot answer that then obviously it won't be opened up.

Re:Irrational open source fanboys

[gravious]

Irrational? No. Maybe you could argue for misguided. Or optimistic. Or deluded. But irrational? No, it's _entirely_ rational to desire this state of affairs.

Regarding smartphone manufacturers having a clue. Way back in the day before Linux people made the same arguments about servers and desktops and laptops. Before Android people made the same argument about smartphones. Heck now there's even an open hardware server consortium. These processes always seem impossible right up until the point they become reality and then they were the most obvious thing in the world.

Smartphone makers employ engineers and engineers give enough of a 'shit' about Ubuntu, ergo smartphone makers give enough of a 'shit' about Ubuntu.

This 'silly fantasy world' you speak of. It's called mutual cooperation. It's called trust. It's called sharing the load. It's called socially responsible. It's called not being hostile to your users and customers. The only hurt I see is the current situation. A mirage of short-term gains for longer-term whatever the opposite of benefits is (what's the opposite of benefits again? aaargh, stupid brain).

Statistically insignificant? 75% of smartphones run a more or less free and open-source OS. Oh, there's that word silly again. And shit. And fucking silly. And ... Oh dear I'm having a hard time digesting your argument (you do have an argument, don't you) what with all the childish insults you're dropping. Tell you what, when you grow up and want to have an adult conversation with us I'm all ears, until then any chance you could stay away from that keyboard of yours and not waste our time. You come across like you're aged 14 and so do the people who upvoted you.

Re:Irrational open source fanboys

[peppepz]

because they can't just hard code that right into the chip and never let you see it ...

No, because we would see either the software interfacing with the hard-coded backdoor, or some undocumented hardware means of communication coming out from the chip, and we'd start asking questions.

So if I just embed my code into the processor itself, you won't bitch.

Thats just silly.

Embedding code in (readonly or flash) ROMs is actually preferred from Stallman's point of view, because it allows the hardware to work out-of-the-box when using free software to control that hardware. Binary firmware is problematic for free software operating systems, not because free software enthusiasts have some maniac obsession about not running binaries that they haven't compiled themselves, but because the copyright holders of the firmware binary blobs often attach very restrictive licensing conditions to them, making them very hard or impossible to redistribute.

Open mouth, Insert foot.

[westlake]

Since Mint has solidly dethroned it, who cares about Ubuntu?

Next time, don't link to a page that debunks your own argument.

The DistroWatch Page Hit Ranking statistics are a light-hearted way of measuring the popularity of Linux distributions and other free operating systems among the visitors of this website. They correlate neither to usage nor to quality and should not be used to measure the market share of distributions. They simply show the number of times a distribution page on DistroWatch.com was accessed each day, nothing more.

DistroWatch Page Hit Ranking

What an open source baseband can be.

[queazocotal]

Open source basebands cannot, legally, in most parts of the world be up-datable by the user, which removes most of the interest.

There are several good reasons for this.
Radio is a shared resource. Cellphones only work as well as they do as the towers arrange it so that no cellphone is transmitting on top of another one.

The modem hardware is quite capable in most cases of transmitting right over the top of other transmissions. The worst case would be a free app turning up that gave free data transfer between nearby phones. And did this by ignoring the towers, and going direct.
This has the potential to knock off dozens of calls from the network per user, some of which may be emergency calls.

FCC/... approvals are inherently with a given software version of the modem - most of the behaviour of the modem is set by software - and changing that software without approval will void the approval of the phone.

In some countries, there is actual specific legislation.
If your open-source baseband could change the IMEI, then once you have been informed that this has been done, you are actually committing an offence if you continue to sell the phone which enables the user to do this in the UK.

Re:What an open source baseband can be.

[drinkypoo]

Open source basebands cannot, legally, in most parts of the world be up-datable by the user, which removes most of the interest.

But they could use signature verification to [more or less] prevent people from updating it while still releasing the source code. As long as you could compile it (though not load it) you could verify that it was the software used on the device.

Re:What an open source baseband can be.

[VortexCortex]

Spectrum is a public resource. Your comment perfectly illustrates why we need segments of various classes of signal free and open to the public. No restrictions like on family-band where store and forward packet radio is illegal.

I don't see what the huge concern is. Really, for less that $20 you can create a small tunable jammer. If people want to do the jamming they can already. The assumption that folks won't play by the rules is idiotic considering the existence of short-wave two-way radios. Who knows what innovation we could have if tinkerers were allowed to play. Perhaps a world wide distributed store and forward self organizing spread spectrum multi-power level mesh network with data deduplication (infohashes for resources) which inherently has low latency, free collocation, and anonymity built in because you get most of your data from your neighbors or their neighbors instead of re-sending from the source -- Essentially a terrestrial version of NASA's Space Internet (Delay Tolerant Network) is possible.

We have the technology to create a network where you only pay for the device then become a node in a network: No monthly fees, bigger and more expensive node, faster your cross country connection. We have the technology to automatically frequency hop and reduce or increase power so that channels can be reused over short ranges. We have the technology for point to point line of sight beams. We hobbyists have organized complex information networks like Fidonet before, and were such system allowed, non profit groups could handle coordination and management of local line-of-sight networking. Folks that say it's impossible have never tried, and are likely ignorant of HAM radio operations. Cell phones are proof of the viability.

Given the existence of legally purchasable capacitors, transistors and wires, the issue isn't that software defined radio could possibly stomp on other people's signals -- Hell, a fist or bat could potentially injure people, but we don't lop off hands and outlaw ball games. The issue is that software defined radio threatens to destroy the need for carriers altogether. That's a good thing for the consumers, and hence why it isn't happening: The FCC and equivalent bodies operate in the best interest of the corporations not the people.

Re:What an open source baseband can be.

[VikingNation]

Yes. I have heard about reverse engineering on posts on the Internet. I also have read that many base band processors are implemented in ASICs (QUALCOMM) which are difficult to hack. The base band processors are done with ASICs because they are ultra-low power. A software based band processor will eat your battery very quickly.
So there are reasons, besides hacking prevention, to guide one to make hardware based units.

Re:What an open source baseband can be.

[Viol8]

"Spectrum is a public resource."

So is air and water, but you can't just pollute them as you see fit. Rules exist for everyones benefit, they're not there just to piss you off personally. Get over it.

Re:What an open source baseband can be.

[AmiMoJo]

Really, for less that $20 you can create a small tunable jammer.

Yes, but in most places jammers are illegal. The point of the law is not to alter reality so they no longer exist, it is to punish people who abuse a shared resource by operating them.

Regulatory issues

[Todd Knarr]

Most of it's probably regulatory issues. Anything that transmits radio has to be set up so it can't go outside the FCC-set limits (eg. stays within maximum allowable power for a given channel, stays only on assigned channels, etc. etc.). That used to be handled in hardware, but these days it's cheaper to use generic hardware that'll transmit at any power on any channel and then impose the limitations within the baseband code. And it's cheaper to allow updating of the baseband than it is to replace phones to fix problems in the baseband code. That combination means that open-source baseband would allow you to re-flash a baseband that'd go outside regulatory limits, which'd be a no-no. Combine with a legal environment where the phone manufacturer, not the consumer, would be the one sued (because they've got deep pockets and the consumers don't) and you can see why we have the situation we have.

I'll pass.

[Drunkulus]

A phone from Canonical? This is the company who brings us the buggiest linux distribution in history, the company whose own forums got hacked and were down for two weeks while they tried to restore a backup. They have never made money and are completely dependent on the continued financial support of their self appointed benevolent dictator for life. With tens of millions in personal losses so far, and years past his own deadline for Canonical to break even, do you really trust that they will be around? I hear all the time that Ubuntu is good for linux. I suppose it is, the same way that factory farming is good for chickens.

Re:I'll pass.

[wisnoskij]

If Google had not turned completely incompetent, and have shown their complete inability to produce a even half decent smartphone OS, I would agree.

but at this point we need as many competitors as possible.

It's not that easy.

[dos1]

If the open source baseband was even remotely feasible to do, open projects like Openmoko, OpenPhoenux (GTA04, Neo900) together with OsmocomBB would already come up with 100% open GSM device. The people working on those project dream to be able to do that, but they simply can't. OsmocomBB is practically a research project, as there are no practical use-cases for it to "normal user" (in most countries it's illegal to use modem with OsmocomBB on it unless you're operating it with your own BTS-lab network you got permission to set up for development or research purposes), and it only operates on very old devices with TI Calypso, as basically all of more modern basebands are cryptographically signed (TI Calypso was also supposed to be, but for some unknown reason that feature was disabled, probably due to misconfiguration at the factory - this is the only reason OsmocomBB was possible at all).

Unless we do lots of legal lobbying and raise much more resources than a company like Canonical has (trust me, building proper 4G modem is awfully hard and expensive. You have to comply to several thousands pages of protocol documentation and pass many certifications. Canonical probably could would be able to afford producing Ubuntu Edge, but they certainly won't be able to afford the modem development), it's much more helpful to look at projects like Neo900 ( http://neo900.org/ ) which aim for the best possible separation between APE and the baseband with built-in monitoring in case you suspect modem might be doing something malicious. In my opinion, this is the proper step forward the truly free mobile devices in our pockets, not shouting and demanding open basebands (even if we all, including Neo900 developers, dream about them).

Re:It's not that easy.

[dos1]

Also, "The choice of Canonical to use a binary only baseband is even more disappointing when Osmocom have already produced a functional open-source GSM baseband for the Calypso chipset. One must wonder why was this not adopted or improved upon by the talented individuals at Canonical, especially given the previous enthusiasm for open-source philosophy."

The reason is simple. They didn't want to limit their capabilities to 2G EDGE. I suppose that the target niche that could accept such limitation to gain some freedom already has their Freerunners in their pockets.

I so much want a free device and I completely agree with the spirit of the article, but it's unfortunately so damn wrong at technical level it hurts :(

Open source baseband will never happen

[jonwil]

Even if Samsung (the biggest phone maker in the world) decided it wanted open source baseband for its phones, it wouldn't happen.
It wont happen because of:
1.NDAs and secret stuff. (last I checked, protocols like GSM still contain stuff you cant officially get without NDAs, also the makers of the cellular radio hardware would never give away the secrets of their cellular radio hardware to their competitors)
2.Patents (with all the patents applying to cellular technology, any source that was made available would be examined by an army of patent lawyers looking for violations, also some of the license agreements tied to cellular standards probably specifically prohibit sharing source with anyone who hasn't also signed a patent agreement)
3.Carriers (no carrier is going to want a baseband that could be changed because it could be changed in ways that harm their networks (maybe not intentionally but it could still happen)
4.Regulations (FCC and other regulators have strict rules about how cellphones are allowed to operate and I doubt they would allow a phone with an open source baseband to get approval because such a phone could be modified to violate the rules)

Re:Not so open source friendly

[shutdown -p now]

The protocol is not XMPP anymore, not since Talk was phased out in favor of Hangouts. It was pretty big news item a year ago, did you miss it? It even made the EFF chime in to complain.

Coincidentally, a mere week before that happened, Microsoft added Google Talk support to outlook.com webmail (which already supported FB chat and Live/Skype). Needless to say, said support became effectively dysfunctional for anyone who "upgraded" from Talk to Hangouts.

Re:Not so open source friendly

[jopsen]

Coincidentally, a mere week before that happened, Microsoft added Google Talk support to outlook.com webmail (which already supported FB chat and Live/Skype). Needless to say, said support became effectively dysfunctional for anyone who "upgraded" from Talk to Hangouts.

Yeah, Google really took a page from the Microsoft playbook there...

I seriously doubt they're going to get us anything open sourced. Google is starting to look more like Microsoft in the 90'ties.

This move is particularly sad, because Google went with XMPP because they didn't have a customer base and needed others to open up and integrate. And now that Microsoft plays ball, Google just kicks it off the field.


On-topic, open source baseband isn't so important. It's not really something that very hackable anyways. Nor should it be hackable, just imagine teenager bringing down the GSM network by playing around with their firm ware. That is not a good thing.
Nevertheless, Mozilla with Firefox OS might eventually be in a position to pressure manufacturers at some point. I know they should love to, but there is still some market to grow before they have enough leverage.