Google's Project Ara Could Bring PC-Like Hardware Ecosystem To Phones
An anonymous reader writes "Now that Google's modular phone effort, Project Ara, looks a bit less like vaporware, people are starting to figure out its implications for the future of cellphones. One fascinating possibility is that it could transform the cellphone purchasing process into something resembling desktop computer purchasing. Enthusiasts could search out the individual parts they like the best and assemble them into cellphone Voltron. People who just want a decent phone with no hassle could look at pre-built offerings — and not just from Apple, Samsung, and the like. It could open up a whole new group of phone 'manufacturers.' Of course, this comes with drawbacks, too — if you think fragmentation is bad now, imagine trying to support thousands of different hardware combinations."
It'll be way too expensive to have a build your own phone. And it's not what the majority of consumers want. Apple has proven this time and time again. An unserviceable phone, an unserviceable tablet, and now unserviceable laptops. None of which have the simplest of battery swaps available (hell even the ram is soldered on board now in the MBPs). They've built the largest computer (mostly mobile) empire on hardware that is idiot proof and has no options. This is what consumers want. A build your own phone would be fun, but just not practical and way too expensive.
I'll get slammed, I know. Fanbois unite and all that. Oddly enough I'm typing this on a 5yr old MBP that I will lament when it finally goes. I don't like ios and most likely will never upgrade since I do not like their new models. I know this one has been through a lot with me and performed flawlessly. I've done some hardware hacks and it's been fun.
This is a lousy idea for a smartphone, but it has potential as an industrial automation and robotics controller. Those are built up from lots of little modules, but the mechanical and electrical standards are decades old, and systems are too bulky. Think of this as a replacement for Arduino "shields", too.
How is that going to work with something the size of a phone?
Unless you've got micro-waldos to fiddle with it.
This might have been reasonable years ago, but now phones are getting powerful enough that their hardware is much less of a factor in the user experience. Software is the critical thing on phones now.
Why would anyone want the hassle of piecing together a phone anyway? That sounds like something only /. would be in favor of.
Except for a very few hardcore HW geeks (like me), "modular" PCs are simply not useful. Once a IT department has standardized, they don't change until the vendor stops making the base model, and the PCs are nearly always locked down to simplify support (never mind the stupidity/insanity/bullying by Microsoft that makes many upgrades have to re-authenticate). There's a small market, gamers mostly, that cycle through video cards, and more rarely, HDs/SSDs, but that's about it.
For example, the refurbished desktop (Dell T5400) I'm using for this posting has only the motherboard and CPU left from the minimum-corporate original configuration, but that cost less than a Xeon X5570 and compatible motherboard would have cost me when I bought it, and I've filled every slot, but one.
More likely, end users will rarely change a component, and phone vendors may find modularity useful for prototyping, but they won't bear the cost of the connectors.
It falls into a dozen parts that I can't recover in the dark. Made harder by the fact that the LED light bounced somewhere and is now under someone's foot.
Right now, I just have 3 parts: phone, battery and back cover to worry about.
Slashdot's rate-of-post filter: Preventing you from posting too many great ideas at once.
There will probably be a market for this in the tech enthusiast. But it will be highly unlikely to go mainstream. Mainstream (iphone 5s) is 7.6mm thick and weighs. According to http://motorolaara.com/2013/10... it is probably about 9.3mm - effectively as chunky as a 2 year old device.
What may evolve from this is specialist hardware and specialist configurations.
Some interesting spin-off technologies might be high speed bus interconnects (thunderbolt 2), modular and novel hardware configs (3d scanning - project tango, yotaphone - e-ink backside). Ultimately, enabling technology advances is what google spends it money on these days...
Sounds great but what about the price. For sure, it is not cheap than usual smartphone that we see in the market.
... if they get out of the hardware business and reinvent themselves as a software/content company. If hardware margins diminish, they could still make money on app sales, books, music and movies.
Current market share is what, optimistically, 25% ? That's 3/4 of the market that aren't iTunes customers.
Tie the iOS ROM specifically to an Apple A7 and charge OEMs a fee per CPU/ROM component.
Drivers? Develop an iOS shim over whatever Google is proposing for Android. Better yet, support os independent drivers e.g. efi bytecode.
I have a cellphone. It is an Applie iPhone 3GS I got two years ago as a free upgrade from my flip phone.It works fine. My wife and daughter have Samsung Galaxy 3. They work. We can call each other and text each other, and if we have time, wecan play games on them and occasionally listen to music. Will this new device help that? Not really, not for FREE, which is what we paid for our phones. Ara is a solution to a problem we don't (and no one I know) has.
Shoes for Industry. Shoes for the Dead.
What exactly would you replace the MacBook with though? I mean, they're certainly not perfect but I'd argue they're still one of the best high end laptops available right now. I mean, especially since it seems most other laptops are effectively rebadged Compals.
Something that already exists on the PC. You can trivially boot up any operating system you want on any PC and the basic things like the display and the input devices will just work.
This is because the PC platform not only has certain basic hardware components standardised, but also because there are interfaces to enumerate the hardware you have. Once you have your kernel in memory and running, it can simply look for the hardware and access the hardware accordingly.
On ARM there is no such thing as a PCI bus. Therefore your kernel needs to be compiled for the very device you want to use it for. You cannot just compile in the most common ethernet controllers into your kernel and expect it to choose the right one. This may work, but very likely your first driver will try to probe blindly for its device, crashing your system, before the second one even has a chance to run.
This is why there are movements to create a common hardware interface, one where you just have a single operating system image running on a huge variation of hardware just like on the PC. Unfortunately the business model of ARM doesn't help here. ARM licenses its cores to many SoC manufacturers. Each one of them hopes to lock in its customers making it deliberately hard to switch to any of their competitors. A common interface would sweep away the borders. You could switch from manufacturer A to manufacturer B just like you can switch from a Dell PC to an HP one.
With Ara, a dead battery in the middle of a day trip doesn’t set off a frantic search for someone with a charger. Instead, you pop in a spare.
Revolutionary.
From TFS: "Now that Google's modular phone effort, Project Ara, looks a bit less like vaporware"
Wait... what hallucinogenics is "anymous reader" overdosing on to come to the conclusion that Project Ara "looks a bit less like vaporware"? It's nothing but a bunch of sketches, pretty graphics, cheap models, and vague design concepts. It's practically the very effin' definition of vaporware.
It's not going to be LIKE the PC ecosystem; it IS the PC ecosystem, just with new players. 10 years from now your Replicators, laptops & server clusters are all going to be sporting Ara-derived chassis. You heard it here first, kids. Why deal with Intel & IBM's bullshit when you have a architecture-agnostic interface ready to go? A computer is the result of an accumulation of standards & as someone that's taken a decades-long generalistic approach to the industry, this IS the new standard. Maybe not in its initial public incarnation, but 2.0'll hit it of just like Android 2.0 took over mobile software.
The magnets exert 30 N (6.74 lb) in lock state. It won't be falling apart.
Or just use a damn case, like you should for any expensive phone.
TFA doesn't explain what would motivate entrepreneurs to invest in this concept. Suppose you have a great idea for a module- are you willing to design & fabricate it for an unknown number of buyers? Not as easy as selling an app for a known market of millions.
...omphaloskepsis often...
Not what the masses want.
It's REALLY not what the masses want when you consider drivers have to me made and maintained for each module... nothing like driver conflicts in your phone! I don't know that even many geeks want to go back to that model.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
> We all have shit phones and we're happy, why does anyone need anything else
You wouldn't have felt the need to point that out if you genuinely believed it.
And listening to music is a solution to a problem that no one has, but the music industry is worth billions of dollars a year. What does that tell you?
https://phonebloks.com/en/goals
http://www.bbc.com/news/magazine-24490331
Is google now just copying without even attributing?
The big manufacturers like Samsung, HTC, et al. will probably continue to sell one-off phone models like they do now. "People who just want a decent phone" is the majority of the smartphone market, and the big companies probably won't want to lose their current model of selling you a new phone every two years.
That said, I would love to be proven wrong—all it takes is for one of them to start doing it and the rest will likely follow suit.
regular man whom love computer (Also, fuck beta).
So enthusiasts can build liquid-cooled cellphones that are overclocked? With windowed cases and cabling with LED racer lights running up and down them? What are the graphic card options?
WTF? Where is the geek spirit in this /. crowd? When a manufacturer releases a phone with battery soldered, everyone's complaining. When a laptop manufacturer releases a laptop that you can't upgrade, complaining again. Now that people are putting effort to allow you to custom your mobile device till your heart bleeds, you are complaining again.
I had enough of phones that I have to throw away because of one very small, and not even the most important, component went bad, and I can't do anything. And it's not worth repairing coz the repair cost is almost as high, or even higher, than buying a new phone. What a fucking waste of resources.
Give me this modular design anyday. I've been waiting for someone to do this for laptop and mobile phone for a decade. Can't come soon enough.
Just release the design, release the interface, make it so open that anyone on the planet can manufacture components without huge license cost, and let the market decide. I'm sure there will a lot of entrepreneurial folks who will set up shop to assemble this into a nice package for your customization. Just like the PC era. Bring it on. There will be a lot of new applications. Talk about wearables? Wait till you have all these components that you can assemble the way you like it.
If it runs some version of desktop Linux, this could be an awesome desktop. With a built-in UPS and a backup 3G network connection. You'd never shut it down because it's so focused on power savings that it's not really worth it. It would run off a standard USB charger.
It would be great if you could upgrade the CPU and 3D graphics to something tablet or desktop-ish. I could envision a chassis/case that has the standard Project Ara backplane, but mounted below a fan. Bonus points if you can overclock the CPU.
8 of 13 people found this answer helpful. Did you?
From 2006: "With laptops becoming more modular, and the use of mini PCI or PCI express cards for most of the components, are we going to start to see more third party upgrade options for laptops ? [...] Are we going to soon be able to easily upgrade the processors in the laptops as well?"
Don't get me wrong, I'd love to be able to plug a logic analyzer or ham radio module into my smartphone, but I can't see why the mass market would tolerate the extra cost, weight and failure modes of a modular phone.
"If you think fragmentation is bad now, imagine trying to support thousands of different hardware combinations"
Yeah, because we had it way easier when there were a myriad of PC hardware platforms than we do now, right?
I want Universal Firmware - or at least universal to the CPU architecture (ARM, MIPS etc.). It could be supported right now if Android were to make Kernel Device Trees required or if we had *gasp* discoverable busses as we do on the PC.
Then we would be able to have Android or Linux distributions for mobile like we have Linux distributions for the PC. I could buy all kinds of interesting devices from China, and know that I will be able to upgrade them with my favourite "distro" without having to hack about with some guy's weekend project on the XDA developers forum.
If you'd gotten your head out of your own ass for like 20 seconds you could easily read up on the article and realise that google fully intends to make this as open as possible. The only thing afaik that they are keeping for themselves is the endo. Everything else is open and license free. I'm not even sure they can enforce android although if you want to join their ecosystem of stores and whatnot im pretty sure android is the only way. I think that Googles entire goal with this is to break manufacturers monopoly on handsets and get android under more central control as no one will be running samsung-android or htc-android on these things
From system point of view there wouldn't be any complications with different hardware combinations. Linux handled the multi-core transition seemlessly, I have no doubt it will handle the Phonebloks too. However I think that there is a way to go for the Apps and the android.
Except those were rack upon rack of glowing hot vacuum tubes all connected by point-to-point wiring with cabinets of core and drum memory alongside. Very much individual components, with the vacuum tubes having to be replaced at least one a day. These were then replaced with racks of transistor logic, and then IC logic, all on little replaceable cards. All designed for easy maintenance and in the hope that customers would upgrade their systems.
Once again Google proves that it's not a hardware company - they just have too much money to throw around. A modular phone is the dumbest idea I've ever heard of. Despite what they may claim, it'll be larger and more fragile that a non-mobile phone with the same capabilities.
As for Google hardware in general, remember how well they did w/ Motorola Mobility? It also makes me sad to think of how many robotics companies they've bought. Robots are a mix of hardware and software, and Google will never get it right. Doesn't Sergey have enough toys already?
And much less as soon as they move even slightly apart, like from a shock. That's the thing with magnets, they stay in place until given a tug or shock. Apple's MagSafe connector uses magnets exactly because it comes off so easy when tugged.
Let me just say, I'm certain that I don't want to have to choose every component of my phone, at any level of granularity less than "the whole phone", and then assemble it, do maintenance on it, troubleshoot why some piece of third-party software isn't working with my particular mix of phone parts.
You're imagining a system where everything 'just works' for a gigantic ecosystem that somehow increases your choices and simultaneously decreases the cost to get exactly the options YOU want. It's not going to happen.
I know manufacturers are looking for the next big gravy train they can count on to pad out revenues and guarantee executive profits.
But I see this as being niche at best, and completely undesirable at worst.
It's my freaking phone. I don't want to be swapping out video cards and tweaking it.
I, for one, will not be interested in this. And I predict a very tiny amount of people ever will.
Lost at C:>. Found at C.
"Not really, not for FREE which is what we paid for our phones" + 2 year agreement to pay $X/mo or should you really want to cancel, a termination fee that makes up the difference. Either way you end up paying to own a device that is likely tethered to the particular carrier. You don't own it.
Was it a case of the mass market not tolerating modular laptops? Or was it a case of manufacturers seeing they can make more $$ by sticking with a handful of designs each year?
I'm completely spit-balling here, but what if each component needing drivers brought their own?
Yes, just like each computer peripheral you bought used to come with a CD - that you had to toss and download an updated version of because your OS had changed since the CD was made... which would remain true of the OS vis-a-vis whatever driver shipped on the physical node.
So it would be no different, you'd be constantly managing drivers.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
I like the hard keyboard on my droid4. I'm really terrible with the onscreen touchboards. Really, really terrible. Which makes onscreen touchboards not good for work emails, posting in web forums, etc. The hard keyboard makes me able to type anything at all. As the market seems to be moving away from built-in keyboards, a optional, modular hard keyboard, would be great for those of us that need them, without forcing them onto the larger market public, or to be stuck with an otherwise tremendously mediocre smartphone just to get the only one with a keyboard on it.
I myself am a big fan of modular design. I've even long wished for a very modular laptop, such that I could pick my case, pick my motherboard, pick my GPU, etc. like people like me do when building desktops or towers. Some laptop equivalent of the ATX standard...
I think that Googles entire goal with this is to break manufacturers monopoly on handsets and get android under more central control as no one will be running samsung-android or htc-android on these things
Eh, I wouldn't underestimate manufacturers' ability to attempt to control them. As long as the bus allows it, all Samsung has to do is sell Samsung-branded modules that only work (or only fully work) with other Samsung-branded modules.
Think about the heyday of the PC market - sure, you had vendors selling complete systems, but you could also buy individual components to upgrade, and even build complete systems from scratch yourself. If this tech allows the mobile phone market to follow the pattern of the explosion of the "IBM-compatible" PC market, it could lead to real customer choice and innovation, as long as the general public avoids too much vendor lock-in.
Further, the Android app market already has to deal with the large variation in devices, software versions, screen sizes, etc.; so I don't think this would change it that much. PC software has pretty much always had to list minimum hardware requirements, so I don't see this as being much different, with apps requiring minimums for Android version, memory, processor power, etc.
my, your, his/her/its, our, your, their
I'm, you're, he's/she's/it's, we're, you're, they're
I would imagine that the increased expertise you see among users of the OS X version of your app would not correlate well with users of an iOS app. OS X and iOS are very different market segments.
Simple physical factors argue against a modular phone. The sockets and connectors that would enable swapping plug-ins take up more space (and cost more) than just soldering to a PCB. Pieces held together by them aren't as rigid or reliable.
Bad idea. And for what? So a few tech weenies can have something to brag about over a beer?
WIth Google proposing to come out with a phone, is there any one realizing that one next step could be the Google Bank. Google may create a bitcoin type of operation or a virutal bank in the cloud, (line ING).
Leslie Satenstein Montreal Quebec Canada