Led by Motorola’s Advanced Technology and Projects group, Project Ara is developing a free, open hardware platform for creating highly modular smartphones. We want to do for hardware what the Android platform has done for software: create a vibrant third-party developer ecosystem, lower the barriers to entry, increase the pace of innovation, and substantially compress development timelines.
[…]
The design for Project Ara consists of what we call an endoskeleton (endo) and modules. The endo is the structural frame that holds all the modules in place. A module can be anything, from a new application processor to a new display or keyboard, an extra battery, a pulse oximeter – or something not yet thought of!
This could be the biggest innovation in smartphones since, well, the first smartphones appeared on the market. I am incredibly excited about this.
It’s an interesting idea in a “huh, cool if they can do that” way but I’m not sure I need or want (extra connectors = bulkier device) an upgradeable CPU in my phone.
Yeah, there is group of people who understand an appreciate modularity, the ability to tinker, improve, learn and discover new things. And there is group that doesn’t want “bulk”.
I’ll take the bulk not only for the modularity, but so I can also have a device that doesn’t feel like it can double as a meat cleaver and won’t shatter if I squeeze it. I’m tired of feeling like I need a case just for my device to fit comfortably in my hand.
I don’t see how a device made of several, smaller devices has any chance of being more durable — probably easier to repair, but not more durable.
It seems like most phones are moving to a system on chip type setup, which seems like it would eliminate the possibility to upgrade storage RAM or CPU independently. I’m not sure how you would upgrade the storage without disrupting the operating system anyway.
This leaves camera battery and screen, which in itself isn’t a bad idea. Still the tradeoffs overall size and price, may not be worth it.
Most SoCs do not have integrated RAM or flash storage so these could be updated provided there was a common bus (that doesn’t exist) that worked across vendors.
Most SoCs designed for PoP memory.
I think the chip makers can any package that is requested. For example there are flavors of Tegra 4 with PoP memory and without.
This is dictated by characteristics like power consumption, number of memory channels / frequency.
Anyway you need 200+pin connector for memory interface on the back of the phone, instead of 4 shiny pins in Phonebloks concept.
Edited 2013-10-29 23:44 UTC
True, but you’d need more that 4 wires for flash, camera, radio, etc. unless you create a common bus and make all devices “smart” which both increases complexity, cost, and unfortunately overhead.
I like this idea a lot. Interesting that they apparently had the same idea as the creator of PhoneBloks, which also intrigued me.
Hopefully this will reverse the current trend of closed “black box” phones with non-removable batteries and no expandable storage. Oh, and while you’re at it Google/Motorola, PLEASE PLEASE PLEASE offer a QWERTY keyboard for this thing, and I’ll be your first buyer.
Edited 2013-10-29 15:22 UTC
This is in partnership with Phonebloks.
I understand what you are saying but modularity could make the battery part of one of a number of ‘endos’. So, replace the endo as the battery bag degrades. Makes the battery replaceable but may still have expense and complexity/recycle issues.
Also SanDisk now makes WiFi enabled SD card holders. Not enough internal storage and don’t want to make a change to the current state of build? Move data to the external SD card. Alternatively move data in if you are doing that sort of housekeeping. And maybe it can be NFC, low power Bluetooth, a USB cable or some other option yet to be.
No reason there can’t be a QWERTY module. Well, if there is enough demand for them to be made. But don’t worry, I think there will be.
I don’t want to sound like I am dumping on the Phonebloks guy, but most engineers over the age of 40 have had this idea in one form or another.
The problem isn’t ideas, ideas are cheap and easy; the hard part is execution. As others have pointed out, a modular phone is almost certainly going to be heavier and bulkier than the alternative.
It may well be more expensive given the need to make each module its own discreet product — each of which will have to go through FCC & CE certification separately.
There are physical issues, making sure this jigsaw of components can stay together without falling apart, but making it easy enough to take apart. There are bus and bandwidth issues.
And then there is the single biggest issue: Will enough people want it to justify its continued existence? Or will people look at the cost of upgrading their screen and realize it is nearly the same as getting a new subsidized phone.
So I am encouraged that Motorola is giving it a go, they can execute on this kind of technology; but not quite sure the technology is ready for the reality of the market just yet
I am not even an engineer and I am not over 40 and I had similar idea years ago.
Edited 2013-10-30 00:14 UTC
I’d go for such a platform if it just were a modular radio package. One for each carrier. Then an unlocked phone would truly be unlocked.
You mean like replacing a SIM card in a non-US phone?
I think he means the ability to use the same phone but switch between CDMA and GSM carriers, which is lower level than just switching SIM cards between GSM carriers.
There are already phones that can do that. See:
Lenono a600e – dual sim CDMA + GSM: http://www.shopclues.com/lenovo-a600e-cdma–gsm-dual-sim-android-mo…
You don’t even need to switch between them, you can connect to both networks at the same time. And yes, even CDMA phones can use SIM cards.
Edited 2013-10-30 18:36 UTC
I’m well aware of that, having owned a Sprint iPhone 4S until recently. My post was more of a clarification of what the GP said.
In 95% of the world it already is.
Open hardware as in open drivers and specs? That would be very welcome.
Unfortunately what Android did to hardware on the other hand, was “Android only hardware” so far. So getting really open and serious options will be a breakthrough.
Edited 2013-10-29 16:27 UTC
shmerl,
Making the hardware open is a neat idea, however to be honest I’m more interested in having a fully open software stack. I think artificial software restrictions have done more to discourage independent innovation than the lack of hardware configurations.
I expect the modular hardware would be limited to a niche market, but as long as it runs a fully open software stack, then I’m not complaining.
Well, fully open stack depends on hardware to be open (for open drivers), as in providing hardware documentation which enables writing drivers for it. Otherwise you’ll end up with closed blobs and Android-like mess which limits the choice of operating systems you can make for it (which in turn limits the rest of the software stack). Unless of course you can reverse engineer all the closed components.
Edited 2013-10-29 19:15 UTC
shmerl,
Oh I hear you. And even if you can reverse engineer all the drivers/hardware for your device, then what? After all that hard work you will likely end up on your own little software island because no one else can actually run your code unless they go out of their way to buy the exact same device as you. And even if you accept this limitation, there’s no guarantee the hardware will even be available to buy once your software becomes useful.
I did some pretty awesome things with a tiny buffalo linkstation NAS drive, creating a very nice little VOIP PBX for my house. I was very happy with it, but when Buffalo obsoleted the product and released a new model, I had to throw away most of my earlier work and restart from the ground up. I decided then and there that I wouldn’t ever become dependent upon proprietary ARM hardware again until this got sorted out. Of course we now know that it didn’t get sorted out and it may never get sorted out because manufacturers would rather we use their proprietary software anyways.
http://blog.gsmarena.com/phonebloks-is-the-modular-phone-concept-yo…
The video is here
http://youtu.be/oDAw7vW7H0c
Sort of Lego for tech. People have also been talking about this approach in relation to high end DSLRs as well – why upgrade the body to get a new sensor?
They are cooperating with the Phoneblocks team.
Won’t all the excessive connectors add weight? How will each connector reroute the correct signal to the right place?
Nope. They are weightless.
I love the idea though. I’m envisioning multiple battery modules and other combinations for rugged outdoor long term use.
That will be entailed in endoskeleton design(s).
I am interested in this from the customization perspective (as a geek, of course).
But the environmental aspect is perhaps even more important. It is just so wasteful to dispose of equipment when most of it is perfectly functional still.
It is an uphill battle, but I hope it works, and I hope it gets applied to other products also.
The idea sounds great on paper – add, upgrade or remove parts of your phone (or tablet eventually?) to match the specs *you* want.
There’s so many downers though that you can’t see it succeeding:
* It will inevitably cost a lot more than a standard phone, especially if only Motorola make them – there has to be take up from most major manufacturers (and full interoperability, which is highly unlikely) for this to work and I don’t see that happening.
* It will be heavier than a standard phone.
* Will it survive a drop on a hard floor without all the modules flying off in all directions?
* Can you put the modules in *any* order you like (useful if the previous point ends up a jigsaw puzzle 🙂 ) – that would require connectors on all four sides and connector covers/fasteners where appropriate.
* Will all modules be completely interchangeable and multipliable (e.g. 2 screens, 2 RAM modules, 2 CPU modules etc.)?
To sum it up – nice idea, too many hurdles to succeed.
I don’t see a big problem with heavier and price. That just depends on 2 things:
devices and parts keep getting cheaper and smaller.
Which seems to be working just fine.
Massproduction.
That last part depends on how many people would want (and can pay for) such a device.
We don’t need to go all the way to lego. Just don’t glue everything down and use impossibly small connectors. Just don’t hide the battery. Just don’t avoid interoperability standards.
Google would make a bigger impact just by putting the SD card slot back into their nexus phones
The idea is to make it more PC like.
You guys want to run <insert niche os> on your phone, project like this are likely to be your only hope of having something that is nearly fully functional.
I LOVE this idea! What the smartphone and tblet market need is a modular platform, much like the ATX PC became. To be able to swap out CPU’s, RAM, storage, screens, batteries etc will make the smartphone and tablet landscapes so much more interesting. I don’t want to fork out £500 a time whenever i want more RAM, better graphics or more processing power. What i’d really like is to be able to swap out my CPU for a better one instead of just chucking out a perfectly good phone. Until someone comes out with a modular smartphone, i’ll never be satisfied…
Of course, only a small fraction of the PC-using population actually use that modularity. Most just buy a PC from Dell (etc), use it until it breaks, then buy a new one. You or I might value the ability to open up the box and put in some more memory or another disk, but most people wouldn’t even notice if the case was welded shut…
It’s a good idea, but as long as planned obsolescence is the engine in our consuming lifestyle, I doubt it will ever be a success. Companies want us to throw away perfectly usable components/products so they can sell us new ones.
Just look at the PC market. Before it was more normal to build your own PC and install the software you wanted, from the distributor you preferred. Now people buy locked down devices where you can’t even replace the battery, with locked down ecosystems and non-standarized accessories. The market is going that way, and it makes them a lot of profit.
It’s never been normal to build your own PC. Sure, it’s an option, but only for the kind of person who’s actually capable of building their own PC – someone who knows what a motherboard and RAM are, and can operate a screwdriver without injuring themselves.
Most PC users do not fall into that category… they buy pre-assembled machines from Dell or a local store, then replace them with a new one when they eventually die.
Note that I wrote “more normal”. English is not my first language, so it’s probably not the best way to say what I wanted to say.
The good thing about the blocks and Motorola project is the module parts really are blocks. So it might end up being a lot easier to replace parts on such a device ?
Nobody will buy it. That’s already been for PC, nobody really want a bulky PC and nobody really care about upgrading just a part of it. Yeah, someone will want this, but it is just a zero point something percentage of people out there. It’s like changeable batteries, nobody really care if their smartphones battery can be changed. If you say differently you just don’t live in the real world
Computing will be slates (small, medium and big), wireless network and cloud. Even in servers the trend is “everything on chip / on board”. Want to upgrade, then change the blade.
Frankly, I’m not so sure about this stuff. Most professional uses of computing are related to much more complex input then slates can provide. Typing documents on touch screen? Photoshop on touchscreen? CAD on touchscreen? Just no way.
See, nothing of the slate is really new. It just got hyped. I owned a PalmOS 5 device, and I’m not exactly sure whether I can’t name any single advancement of the current tablets compared to what that device allowed (discarding natural course of technology advancement, of course).
The cloud thing is just again an overblown client-server architecture, this time with a huge single ugly API, which is more of regression than advancement, and more vocal evangelists. Back in the day the concept was killed by increasing processing power of hardware; these days it still increases, and other factors – eg. privacy, long-term availability – are in play. I doubt that cloud computing would survive a major increase in end-users’ bandwidth for instance.
Both technologies are hyped, which instigates their adoption rate, but they were already tried more then once, so their dominance really doesn’t feel imminent. The only real change is that back in the day these technologies were targeting people who knew their uses, and now they are applied to general consumers.
You don’t get it then. Cloud itself has nothing to do with client server arch specifically.
* If I wanna scale my database to meet demand, I just select a dropdown on Amazon ECS. I don’t need to do anything else.
* If I want redundancy across Amazon data-centres, I just push the instance onto a different data-centre done.
* If I want to send email, I just call the REST API. Job done.
Sorry working with cloud shit (I been mostly Amazon and Azure) has been a pretty awesome as a developer. I can use whatever tech I want, scale it to suit and do it with minimum fuss.
Traditional server infrastructure can’t do that.
Not if you are talking about things that are called “cloud services” when they are really websites or web services …. well that is a different thing entirely.
Edited 2013-10-30 21:15 UTC
Photoshop or CAD could be awesome on a touchscreen… (a digital form of drafting table)
That is what you think, I wouldn’t be so sure.
Do you remember the story of the mainframe and the PC ? This could happen again.
A quick back of the envelope calculation tells me that in 15 years a device the size of a smartphone can have the same processing power, memory capability and storage capability as the IBM Watson has now.
Sure, average Joe just wants a pretty tweetbook machine but there’s gaming/enthusiast market where “bulky” and “upgrading” are still doing very much OK.
Gee a modular phone…. yawn. The only thing Motorola make that was interesting was the Atrix, and they completely stuffed the concept up. A computer phone is still a dream not reality. Needs 2-4gb of ram to become a reality, 1 gb of ram let me open 3 firefox tabs and then firefox crashed on my lapdock! I don’t want a phone that comes apart because it means it’ll probably come apart in my pocket or when I’m trying to use it. It’s just a damn shame that Motorola rather than fixing their broken but awesome computer phone concept, are planning to rush another bad idea to market.
Edited 2013-10-30 04:58 UTC
If that phone falls on the ground, it should break into 8 pieces that scatters all around
Because nobody upgrades their phone purely for fashion reasons, nope.
This will never work, ever. It is an idea enamoured by bedroom designers and is absolutely detached from the reality of execution.
I don’t know why you think this has a lot to do with fashion.
In no way do I agree with the above quote. If anything it may become a sizable innovation for handheld devices in general, but to say this is a smartphone innovation is laughable to me. I do see potential for other types devices that could benefit from the customization – things such as metering, measuring, various types of data collecting & processing, etc. This project may have a future as a platform for creating multi-tools, but as a smartphone?…. Nah, I won’t hold my breath.
That was the first thing I thought of when I saw the original Phonebloks announcement a while back. I’m working on bringing my workplace into the modern age of inventory/warehouse management, and I’ve been looking into various data acquisition tools. We have a few barcode laser wands, a camera based 2D wand, and a few Datalogic Kyman laser PDAs, but I’d be happier with modular camera based units that can be repaired or upgraded for far less than buying an entire new unit. We’ve experimented with using our cellphones in the warehouse, but there’s too much danger of damaging them and driving costs up even further, not to mention the clumsiness of using a consumer oriented multimedia device for a narrowly focused task.
Ohh, yes, I can see that.
The smartphone is driving down the cost of all the sensor parts and other stuff that powers smartphones, if you can buy better reusable parts this could really help that.
(apologies if someone’s already said exactly this above – don’t have time to read the whole thread at the moment)
My personal favourite idea going back some time -no doubt others have had similar many times- is to basically have all the 3g/4g+ radios/chips plus the sim slot and a little memory and small usb ‘network’ cpu live in what is essentially a micro-usb thumb drive 4mm x 12mm x 15mm long (could that do it?) and simply recess the standard micro usb slot on most phones by a hole of those dimensions..
If all Windows Phone, iPhones and Androids operated like this (but were still fully functional wifi only mini tablets without said micro usb cellular thumb drive) – then I’d almost certainly have one of each.. as they have different strengths. …There’s no way however, that I’m up for swapping sims and thereby losing lots functionality in the de-simmed phone (or tbh inactivating the phone in most cases), or paying for 3 mobile contracts..
as for modular screens, camera units, main cpus
…..it doesn’t sound like it can be done both ‘slickly’ AND ‘cheaply’ / ‘affordably’ .
I’d prefer just my modular ‘ipod’-plus-‘plugin-tiny-iphone-unit’ idea.
There is a good reason that mobile phones aren’t modular, it’s called integration. Stuffing everything but the kitchen sink in two or three chips saves space, energy and money. Modular design is even dying in the PC market, why should it make any sense in phones?