Google's Open Source Mobile Platform

As expected, today Google took the wraps off of the gPhone (as the media have for months been referring to the rumored project). Google is "leading a broad industry alliance to transform mobile phones into powerful mobile computers," and will be licensing its software to all comers on an open source basis under the Apache license. (The Wall Street Journal's Ben Worthen demonstrates a miserable grasp of what "open source" means.) Google's US partners include Nextel and Sprint, but not AT&T nor Verizon. Phones will be available in the second half of 2008 — not the spring as earlier reports had speculated. News.com's analysis warns that Google won't take over the mobile market overnight, though they quote Forrester in the opinion that Google may be one of the three biggest mobile players after several years of shakeout.

Re:Phone or Platform?

[timmarhy]

it's just a platform, google has some serious hardware backers though. from googles point of view it's a great move, all they have to produce is software which they are really good at, and someone else takes the risk of producing the actual phones. It would also open up more places for them to shove adsense, which is their cash maker.

consumers will be the winners because it's a serious competitor to at&t and their outrageous charges, it opens up the possiblity of an adsense supported phone, and because google is doing it microsoft will do it giving us even more competition.

Re:Software Development Skills / Security

[cheater512]

You read the article I take it?

Viruses need to self replicate.
Social Enginnering 'OMG Download this cool app d00dz' doesnt count.

There arent any easy ways to get a phone to send a virus to another phone.
The easiest way is Bluetooth or Wifi and then its still a pain in the ass to make it spread.

With Bluetooth you first need to somehow get another phone to connect to you, without user intervention which is impossible (without flaws in the stack).
Then you need to send data to the other phone in a way which makes it execute the code. Also basically impossible.

Whats the chance of Google's code having fundamental bugs like that? Nil.

it's just incompetence

[m2943]

For the past 3 days I've been trying to modify and mess with my Motorola V3M Razor and it's a glitchy hell to try and do. Any phone that's more open than the current phone Nazis keep them is fine with me.

You simply bought a bad phone. If you want an extensible or modifiable phone, you can already get a Palm, Nokia, or Windows Mobile GSM phone; those are quite extensible. The advantage of Android over those existing systems is that it's probably easier to program because it gives you a full set of desktop APIs.

Re:first psot!!!

[loubs001]

I think thats a bit harsh. He didnt even use the term 'open source', he just said 'open' and clarified what he meant by it, which is that anyone can write software for it, which is true (contrast that to the pre-SDK iPhone). I think his concern is a valid one. You could imagine a malicious application on the phone that uses Bluetooth to detect other phones nearby and spam them with SMS messages or something. But I'm sure google's thought of this and there will be security mechanisms, permissions, signed applications with digital certificates etc. etc.

Boo

[timeOday]

It's still just a client device. Somehow I was hoping for a much bolder stroke from google, like if they'd bought up that new spectrum, thrown in their own fiber backbone, and used it to change the cellco/customer relationship fundamentally. So long as they're working through the same old networks, the US cellphone industry will stay pretty much as-is.

Re:Phone or Platform?

[dwater]

> What sets this platform apart form the rest?

The license, and the license fee. Plus, I'll bet, the development environment.

Re:first psot!!!

[Datamonstar]

No, it's not at all harsh for what is supposed to be a professional writer. He starts off with an idea, a dangerous beginning in the first place, that there should be some sort of software security specifically for interfacing phones and PCs in the office. A good idea (perhaps even a profitable one) and doesn't think it through at all. He starts off, not with the good idea, but with a broad, one-sided assumption that all open applications are prone to security issues simply because they are open. If he were somewhere in the ballpark range of competent he would have reversed the two topics and stated that we need security software for smart phone to PC interfaces and that the result of not developing it could be rogue open applications creating a security nightmare. But he didn't. He speculated on something that went well in hand with his idea, but he didn't have a clue about it worked, and also didn't do any research on it to get more knowledge. He even pretty much says all this (sans admitting that he doesn't know what he's talking about and didn't do any research, but that much is very obvious) in his rehash he added to the article to address the people who e-mailed him about his mistake. The update is almost as large as the article itself. I'd say he pretty much deserves to be criticized on his grasp of Open Source as it is demonstrated by this article.

Re:first psot!!!

[Bent Cardan]

Bent Cardan says: Too bad about the gphone. Why can't they make their operating system open and protected like OS X?

Re:Software Development Skills / Security

[timmarhy]

"instead hiring and training those who will tow the party line"

What a load of spin.

and your trying to tag google as irresponsible because people might catch trojans? welcome to the fucking world. nothing on earth is going to prevent people willingly running annakornakova.jpg.exe.

to suggest google is somehow negligent in this area is too retarded for words.

Won't take over the market overnight?

[rastoboy29]

I don't think they'll take over the mobile market overnight, either.  I think it'll take at least a few weeks.

After all, who would want an open standard phone where you can install your own software and not be charged a buck for a text message or a ringtone?  Who the fuck would want that?

Still moronic.

[SanityInAnarchy]

First, it's a blatant rewrite in at least a few places -- while I saw what he was trying to say, maybe, it was worded so horribly wrong that I'd be amazed if it wasn't intentional.

"Google's operating system is open, meaning anyone can write software for it."

Yeah, that's not at all implying that it's about an open platform (vs iPhones locked down one), and not about an open source platform.

But more importantly, he's assuming that cell phone viruses are somehow new with this phone, and that they will somehow cause problems for a corporate network, and that the way to deal with it is anti-virus.

This is wrong on all counts. Cell phone (and mobile) viruses are not new, though they've never been widespread. They generally don't jump to desktop machines -- the corporate network should be safe. And generally, no one's stupid enough to run anti-virus software on Linux, and very few on the Mac -- even on Windows, the usefulness of anti-virus is questionable.

So, your IT guy might freak out -- but really, you've got a much higher risk of getting hit by some road warrior bringing his laptop back into the company network (from Starbucks or whatever).

So that's two for two. Spam him again. Any chance he'll write an update that isn't pure bullshit?

Sounds vaguely familiar

[tm2b]

I dunno... it really strikes me a lot like a number of the software standard alliances that Sun and the other Unix vendors tried to put together or participate in over the years. They always full apart because nobody's interests aligned in any lasting way and everybody had a bad case of NIH ("Not Invented Here").

I'm not saying none have worked, but I am asking honestly - how many technology projects with even half as many partners have actually succeeded in producing a stable platform? It seems to me that the truly successful open source projects have always been independent of any corporate interests - Linux succeeded in making a standard Unix-like platform where years of Dec / Sun / IBM / HP alliances failed and the business interests that have been successful with Linux have done so by learning to support efforts where there was already community leadership instead of trying to dictate a direction to the platform. Netscape did okay, I guess, but that wasn't a big business alliance and hasn't exactly been an exemplar of efficient platform production.

I'm just not seeing that this is a big deal, except that everybody thought that something much more exciting was actually going to happen.

Re:Still moronic.

[emurphy42]

But more importantly, he's assuming that cell phone viruses are somehow new with this phone, and that they will somehow cause problems for a corporate network, and that the way to deal with it is anti-virus.

This is wrong on all counts. Cell phone (and mobile) viruses are not new, though they've never been widespread. They generally don't jump to desktop machines -- the corporate network should be safe. And generally, no one's stupid enough to run anti-virus software on Linux, and very few on the Mac -- even on Windows, the usefulness of anti-virus is questionable.

Things might change if this platform becomes ubiquitous. I'm not saying it's likely, mind you, and anyway the same arguments could be applied to the iPhone SDK (once the bad guys yoink themselves a copy of those dev tools).

Re:Phone or Platform?

[davester666]

It's not the 'openness' of the platform that matters. It's the openness of the end-product, that is delivered to the customer [namely you and me] that matters [well, to you and me]. And that's an issue that's pretty much independent of what OS the phone is running. Particularly in the US.

And Sprint being part of this 'group' means nothing w.r.t. how open the shipping product will be.

The US wireless carriers will fight tooth and nail to NOT be treated as what they are: wireless service providers.

On the other hand, if anything this could make customer demand for 'openness' more difficult, because this fractures the market for developers a bit more. Now, to develop a ubiquitous app, you need to support another platform. One that with the source available, developers can't necessarily count on a given set of API's even being available on a 'googleos' phone...

I think it'll still take quite a while before the US wireless carriers permit much advancement. Even Apple had to deliberately cripple iTunes support on the iPhone so you can't use it over your "unlimited data plan" EDGE connection.

Re:Won't take over the market overnight?

[imemyself]

There are already plenty of phones that you can install your own software on and not have to pay for ringtones (I can't believe that people are stupid enough to do that). For example, I have a Treo (Windows Mobile) - I can install any software I want on it, and easily create my own Compact .NET Framework apps for it if I want to - it doesn't have to be signed by the carrier or anything. I believe I can use any MP3 file as a ringtone, though I just use one of the MIDI's that came with it. Song ringtones annoy the hell out of me. Text messages aren't free - but that obviously has nothing to do with the phone and isn't going to ever happen. You'll always be paying a service provider for text messages - whether its per text, for unlimited text messages, or bundled in with some plan.

If Google is really successful it'll be because they are able to lower the price of smartphones from several hundred dollars to where the cheap toy phones (that don't let you install software/ringtones/etc) currently are. While I do not know how much of the cost of smartphones is for the OS, I highly doubt that a free OS will make smartphones that much cheaper. Maybe they'll subsidize some of the cost through AdSense or something, though I personally would hate to have a phone that forced me to look at ads.

More competition is a good thing though, at the very least it'll hopefully drive prices down a bit.

Re:Still moronic.

[jrumney]

Symbian and Windows Mobile are already "open" enough platforms to expose companies to the theoretical threats this journalist brings up. Open source has nothing to do with this at all, it is an open to developers issue.

Phone hardware vs phone software

[simong]

I have been thinking about the use of Linux on smartphones and one of the conclusions that I came to, rightly or wrongly, is that there is a major licensing problem in the interface between the GPL software in GNU/Linux and the hardware and software employed in a telephony module, to the point that there is a fear that GPL software touching a telephony module would cause the telephony software to become unacceptably open, either from the point of view of business or regulatory authorities, and this is why there is no POTS option for Nokia's Internet Tablet range, and indeed why the iPhone is locked down. OpenMoko has broken this taboo, and will be a major advance in opening the telephony market *if* it passes FCA and European certification - there is no guarantee of this.
To this end I believe that the Google telephony platform will, in its early stages at least, be a GNU/Linux OS running on an ARM processor or similar with a closed interface to the telephony systems, and with Google Gears and a Java for Mobile Telephony, which may or not be the current Mobile Java, as the developer interfaces. There would still be no direct access to the phone module, and only the only open network access would be over wi-fi unless Google manages to obtain its own pieces of the spectrum across the world or can form deals with phone providers... hmm, does that sound familiar?
Right now in the UK for example, I can only see one provider even considering allowing the sort of access that Google would want, and that's the one that has no long distance infrastructure of its own and has just introduced a Skype phone that works over its network, partially to reduce its interconnect costs.
Then again, as most European 3G licences will be about halfway through their life when the OS becomes available, and with the licence holders finally coming to terms with the fact that uptake is being delivered by access to data rather than blocky film clips, the promise of a share of Google's revenues might be enough to encourage the phone providers to open up - a little at least.
This is all empirical but it's what the current state of telephony looks like from the view of an interested spectator. Feel free to correct me.

The problem is the interface

[Colin Smith]

I Not that I think Windows Mobile is the best thing since sliced bread, performance/power wise it's way lacking compared to Symbian, but nevertheless, it is a really nice platform.

You can't just drop a PC style interface onto a mobile, as Qtopia and Windows mobile try to do. It doesn't work. It sucks. There isn't the screen space to waste the way they do, there isn't a keyboard there isn't a mouse.

Symbian and the iPhone are successful because they don't try to fit an inappropriate interface to the devices.

Obligatory OpenMoko disclaimer; sure OpenMoko may be the shit, but the device simple doesn't fit my hardware needs. It's so horribly two years ago.

It's something which has the potential to revolutionise particularly business applications and processes.

English, caffeinemessiah, can you read it?

[Timothy Brownawell]

If anyone was interested in Ben Worthen's moronic grasp of open-source, its pasted below. E-mail your tirades to [e-mail], of which Ben Worthen is the lead writer, and ask him about how he got his job in the first place.

Information-technology departments will ban employees from connecting phones that run Google's operating system to their computers or the corporate network. The reason is that Google's operating system is open, meaning anyone can write software for it. That includes bad guys, who will [...]

He isn't talking about open source. He's talking about it being an open platform like Windows or BSD instead of locked down like game consoles are and the iPhone tried to be. Is the difference really that incomprehensible?