Ask Hackaday: Does Project Ara Solve The Phonebloks’ Problems?

Our tips line is blowing up again, this time directing us to Motorola’s Project Ara: a phone with modular components that plug into a base “endoskeleton.” If you missed the news coverage strewn across the web and you are doing a double-take, that’s because Project Ara is frighteningly similar to the (presumed vaporware) Phonebloks concept from a few weeks ago. Phonebloks was the subject of our last “Ask Hackaday” article, generating hundreds of comments ranging from those defending the concept to those furiously opposed to it.

There’s a conspiracy theory circulating that suggests Motorola released the Phonebloks concept as a viral marketing scheme to generate hype before revealing the official product line. We suspect it’s a bit less conniving. As [jorde] explained on Hacker News, an Israeli startup, Modu, had developed a similar modular cell phone several years ago, and Google bought the patents in May of 2011. A few months later, Google bought something else: Motorola. It seems likely that Project Ara is merely a resurrected and revised Modu, and Motorola conveniently announced it in the wake of Phonebloks’ popularity. Regardless, Motorola has announced that they have partnered with Phonebloks’ creator Dave Hakkens .

So what’s different? Phonebloks was met with cries of “vaporware!” and fervent arguments raising concerns about unavoidable hardware limitations. Motorola claims their goal is:

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.

Unlike Project Ara, Phonebloks didn’t consider open-source hardware (Wayback Machine link), and Motorola makes an interesting argument here: that advances in 3D printing indicate an evolving “open hardware ecosystem,” and the next era of phone development may rest in the hands of your average hacker or a small startup company. Some speculate that the Ara will be similar to the relationship between a PC and its peripherals: Motorola provides the essential guts while giving you some slots for attaching additional components. Let us know in the comments what you think about Project Ara: is it just more vaporware, or a watered-down but plausible alternative to Phonebloks?  And, perhaps most important: do you, as a hacker, want a phone that supports open hardware and lets you plug in “peripherals?” The Phonebloks website has since changed to reflect the partnership with Motorola, and includes a new video that you can watch below.

86 thoughts on “Ask Hackaday: Does Project Ara Solve The Phonebloks’ Problems?

  1. Vaporware? The thunderclap campaign has spammed my facebook news feed no less than 30 times yesterday alone.

    Back to ara. It solved some of the issues of phonebloks in my opinion. Phoneblocks you had the problem of an infinitely configurable backplane able to accomadate varying sized modules regardless of location or type. Ara however it seems they have taken the approach of 3 module sizes and they have to match up with a corresponding socket. I still think there will be technical issues, but they can probably be overcome more easily than the original phonebloks proposal.
    Although, bottom of the image, alternate backplane shown.

    If motorola can demo a working prototype I will be interested. Open source is even better.

  2. Cellphones are so tightly integrated that separating the functions would make them consume more power and more space. That’s the reason why you can fit ten laptops inside a single desktop PC, and why the only thing you can swap out in your laptop is the memories.

    But I digress, let them show what is possible.

    1. Well, you can still swap the hard-drive and a few years ago (not too many!) you could even swap the CPU… it was not for the faint of heart, but it was definitely possible.

      We used to carry good ol’ Nokia phones (3310 anyone? I even had a 5110…) in our pockets, I think we can give up a bit with the “it has to be thin like a hair or it won’t fit my pocket” madness.

          1. Well, the Galaxy S4 is 136.6×69.8×7.9 mm while the 3310 is 113×48×22 mm so it’s about three times as thick.

            Now imaigine if the Galaxy S4 was almost an inch thick.

      1. For what it’s worth, I upgraded the memory, hard drive, AND processor in the laptop that I bought about a year ago. It’s not that difficult if it’s a well designed laptop (this one has a single door that opens to allow access to all of those components) and it was definitely worth the effort in each case.

        If laptop CPU makers would stop switching out pin configurations (and memory controllers in the case of Intel) then we could very easily hold onto our laptops for a much longer time.

      2. You can still cpu swap many laptops. Not all are using soldered on BGA chips. Clevo, some Dells, some ASUS, MSI…etc. I only buy whitebooks because it is the only way to not pay for Windows; and then I usually get a cpu cheaper on ebay than it would be to buy from a custom notebook seller.

    2. Your laptop example actually demonstrates that interchangeability and portability don’t mix very well.

      Introducing interchangeable parts in the phone’s architecture will be paid with bulk volume.

      But on the other side: if the market for these modules takes off we might see them used in other projects. If the volume is high enough we might get good GPS/GSM/WiFi modules for a reasonable price. I can dream, can’t I?

    3. It’s not that I disagree. But I figure we’ll see more and more peripherals integrated into a single die for cellphones…. if you have on master chip and few external components it means smaller PCB, faster and cheaper assembly, and I think that’s the keystone to this concept.

      Our hands aren’t getting any smaller so the outline of the phone is pretty much set. If you can shrink all of the driving electronics it does make more room for a modular backplane design. I’m just surprised that we’re already there — would have thought this was years away still.

      1. Current cell phone and tablet designs are shrinking the electronics to make room for a larger battery while maintaining the same external dimensions. Silicon is getting smaller, but batteries (for a given capacity) aren’t. The time is quickly approaching where there will be no point in making cell phone chips and PCBs smaller, because the device is already 99% battery.

  3. What’s going to happen is, you may end up buying the base by itself, and for a CDMA/4G LTE module, have to go to [verizon/other CDMA carrier here] to pick this up and drop it in your phone. there’s going to be a big conglomerate over how much they can charge you for a bitty piece of plastic and board that does nothing without the shell.

    1. Speaking of radios, how would they solve the problem of needing an antenna that is larger than the module itself to get good reception? Usually the antenna is integrated as a part of the phone’s chassis or circuit board, and changing the radio requires changing the antenna to match.

      1. They solved that problem year ago. They just stoped using antennas. Ok, they have those little fractal chip antenna things but come on. My old phones 10 years ago with actual antennas 3 or 4 inches long that slid out got way better reception than anything produced today does.

        1. That wasn’t exactly because longer antenna = better reception. Antenna size requirement scales with wavelength. As an example, that’s why you didn’t always want your television’s “rabbit ears” to be fully extended, shorter would be better for some channels.

          1. A full wavelenght antenna for 2 GHz is still 15 centimeters or six inches, and the LTE/4G networks may use longer wavelenghts near 800/900 MHz for increased range with a full wavelenght of over a foot.

            The antenna has to be in resonance with the frequency it intends to transmit to work efficiently, so your physical (or electrical) antenna lenght must be an even fraction of the wavelenght, and the shorter the fraction the less it couples with the radio wave. Any piece of wire can be made to radiate, but at a huge loss.

        2. The antennas on the Motorola Micro TAC series were all fake, just a bit of plastic rod. Dunno if the Star TAC antennas were fake.

          An updated Star TAC that’s identical to the original on the outside, even to being able to used the old batteries and new LiPo ones (easy, a second contact set for LiPo), but with the inside completely covered by two high pixel density touch screens would be a most awesome phone, and still smaller than many later clamshell phones.

  4. I’d go a bit further: cellphones should have embraced some modularity from day 1. Not as far reaching as this, but something along the line of an SPI connector with half a cubic cm of space next to it. /hindsight I know, they had to squeeze out as much space as possible.

    But more on topic: open interface (i.e. not a new proprietary connector & SW interface) is probably more important than open hardware blocks.

  5. The problem is that I see no *significant* market for a modular phone. Phonebloks garnered great public and media support for its supposed green credentials, but the argument – in my mind – doesn’t stand. By far the largest proportion of waste is not breakages or failed components; it’s obsolescence. Obsolescence which occurs across almost all the major components; processors, memory, radios, screens and cameras are all routinely improved and people want to upgrade.

    On the argument of repairability; the Phonebloks concept says, for example; if the camera breaks, buy a new camera and clip it in. The fallacy is that most people don’t (catastrophically) break these individual parts of their phones and those that do, by and large either break the screen or take them for a swim, neither of which are benefitted by the Phonebloks concept. Replacing dud components in current phones, whilst a PITA or beyond the skills of most people, is solved by taking it to a local repair shop. Very few people bin their smartphones due to a cracked screen these days. It was actually a much bigger problem when phones were cheaper (or with current budget phones), since in a consumers eyes, the low cost makes them disposable. i.e. it’s cheaper and easier to go out and buy a new phone.

    Finally, on the subject of customisation in the vein of DIY desktop PCs; both make no sense for most people. I’m pretty sure that almost all DIY PCs are high end gaming or workstation type rigs, which I’d guess are below 2% of total sales. I’d guess that the market for such a phone would be the same, in reality.

    As much as I’d quite like to tinker with a modular phone, I don’t see it making any significant waves at all. I hope they prove me wrong.

    1. I like how Android has approached this type of problem. A large majority of regular phone users don’t care about root, or ROMs, or open-source. Nevertheless, Android has delivered a competitive product that appeals to run-of-the-mill consumers while also keeping the things the hackers and power users want.

    2. I agree – and disagree. By using the term “mobile phone” you’re limiting your options…
      Instead, think of a tablet, like the iPad mini. Make every part of it expandable and adjustable. Finally, add one more, purely optional module: the mobile phone transceiver.

      Instead of turning Bluetooth off to save power, simply remove the module. Ditto for WiFi, or GPS…

      You want to take your technology into a restricted area? They exclude transmitters and cameras? Pop out the phone & camera sections, the rest should STILL work.

      Finally, I’ll be interested in the “development” setup mentioned as coming from Motorola. There are many dev boards available [hackaday once did a write-up on them], it would be a dream to be able to hook an FPGA or Raspberry Pi [for instance] up to a compact power supply, display, and maybe even a keypad.

      1. Alan, what part do you disagree on?

        I didn’t limit the options to a mobile phone…the Phonebloks concept did, for the purposes of discussion.

        I do agree that there could be a reasonable market for a high end modular tablet (it could be quite large), with functionality you describe given by specialist/high end hardware modules. It’s the same in any market – the market for customisation is at the high end for specialist applications.

        I don’t get the ‘remove the module to save power’ idea…it’s far easier to just turn it off in software…

        1. I think obsolescence will be solved with the modular idea. As explained by an Ara project member, the fact that components could be swapped out with a newer technology module. Sumple as that. Every slots in the endoskeleton has a power supply and other connections that are needed to make any type of connection that is needed. You fail to realize this is the 21St century and everything we dreamed about when we were just kids are all becoming a reality one by one. Were talking about Google here, it has billions and billions in its disposal to further advance the technology needed to make this happen. Advances in manufacturing, the advent of 3D Printing on a commercial scale, makes production of precise and compact and complex components a thing of the past. Have you been living in the 80’s? Move on… And I also disagree with you comparing a modular phone with a “gaming rig” which only a handful of people do. Your analogy is just plain wrong. You’re supposed to be comparing what the PC did with the desktop computer industry. The modular phone wasn’t just a way to pimp a phone. It is way for people to customize their phones, according to their needs. While pimping it is possible, it wasnt the real reason. Theres a big difference…

          1. Thanks for reminding me about the century. I design products every day and I know very well what most people actually care about and spend their money on. And it’s not on components of a product; it’s the benefits that a finished product brings. I’ve failed to realise nothing that you have said. Markets are driven by demand, not by trying to sell stuff that is clever or neat but that has no utility. The part that you’re missing is that you are not representative of what most other people want. You’ve also demonstrated that you do not understand practicality, cost, needs or desires.

            “You’re supposed to be comparing what the PC did with the desktop computer industry”. Oh wow, thanks for telling me what I should be using for my analogies. Donkey. My analogy is perfect for the comparison. Desktop PCs started as being modular, and the vast majority of the market has moved away from modularity towards products that are less modular, physically smaller/more portable and lower cost. Mobile phones have always been non-modular, and there’s no *significant* demand for them not to be. Since you disagree, give me a specific example where a good proportion of the general public would get any significant benefit from, and be prepared to pay for, some kind of custom plug in module for their phone or tablet? What significant market is there for customisable desktop PCs that are not being used for niche, high-end applications such as CAD workstations or gaming rigs? Even then, a company could make a limited range of CAD workstations or gaming rigs. The customisations themselves are totally unnecessary. Internal customisation is almost always only a novelty. Apple have shown that they are one of the first companies to realise this can work for a high-end machine with the development of the new Mac Pro. High speed data connections give all the external customisation that most users want or need. Prove me wrong.

  6. There’s too many interconnects in this kind of industry for products. There’s no real “unified” standard to base your interconnects on.

    Displays for instance have eDP, LVDS (and the variations of LVDS), FPDLink, and so on. Hardware interconnects could be i2c, SPI, UART, embedded PCI products…

    … so I’m stuck with vendor lock-in at this point to be able to use specific products with my device.

      1. That XKCD precisely describes what happened to USB. In the beginning there were two connectors, Type A and Type B. Now how many are there?

        The Micro B is not much thinner than Mini B but it is much harder to get plugged in, in the dark.

        It’s easy to tell that engineers without a clue about things like ease of use designed USB connectors. In the development lab, people plug things in straight and carefully so as not to damage expensive prototypes. What they needed to do was put some jacks on the back of a box then try to plug those #%@^ cables in while standing in front of the box without being able to see the jacks.

        Perhaps then USB jacks wouldn’t have sharp edges to catch the corners of the plugs and would have some type of surround with a smooth and keyed entry to guide the plug in, making it easy to tell when the plug is the wrong way around. Or how about contacts on top and bottom of the jack *or* the plug, with whichever had the 8 contacts wired so it wouldn’t matter which way it’s plugged in. (Apple finally got that clue for Thunderbolt, and likely has a patent on the concept.)

        There is a top and a bottom to a Type A USB jack, but from the many upside down ports on the fronts of desktops (and some laptops) it’s obvious there’s a lot of hardware designers who either don’t know that or don’t care.

        The engineers should tell the designers to put the ports on the right way up, and the designers should’ve told the engineers how to make it easier to plug the cords in.

        1. There’s only 4 connections in a USB plug. Why didn’t they use a 2.5mm 4 pin jack like Nokia’s and Xbox 360’s headsets? They’re smooth and round, so you can plug them in with any orientation without looking.

  7. The issue that I see that I haven’t seen anyone else address is this: If this idea takes off, a lot of people will be spending a lot of time and effort developing for it. Which is great. But I wonder how much inovation will be lost because we are all then married to the physicality thus presented.
    Remember those smartphone concepts that were basically just a rolled-up display that you unrolled to look at and use? Or the one that was basically just a little cube that projected its screen onto the table in front of you?
    The smartphone has converged on a unified form factor: a thin slab of material of which one side is a touchscreen.
    Ara/phoneblok refines and extends that paradigm. I suspect the next leap forward will depart from it.

    1. ” If this idea takes off, a lot of people will be spending a lot of time and effort developing for it. Which is great. But I wonder how much inovation will be lost because we are all then married to the physicality thus presented.”

      Umm… What is so special about project Ara in that regard? Anything that takes off means that people will spend time developing for it and thus less on other things. One could make the same comment about every new toy, gadget and tool that is ever invented!

      Is your point that developing this idea takes time away from developing those other concepts you mentioned? I’m pretty sure that if anybody id developing those things at Google/Motorola they are a small team dedicated to far out longshots and not a part of this.

      The likely alternative to developer time/resources being spent on Ara is developer time spent on ‘yet another smartphone’ ie, just the same old thing but with a bit more memory, a bit faster processor and a bit higher resolution screen/camera and based on a newer version of stock Android. In other words, exactly what we always see every 6 months or so.

  8. I suppose this could work. However there is not a lot of real estate to work with in something as small as a smart phone, unless you have one of the numerous, gawdy “phablets” people are using these days, and even then the desntiy of parts crammed into those things is high. The problem with the concept is that you must take up even more space for connectors that can withstand more than just a few connection cycles and mounting hardware that will withstand the same. I guess the idea that an LTE modual will be the same size as the next gen LTE-“X” will be the same size and snap into the same spot on whatever ends up being the backplane takes care of some of this, but thats negating the ever shrinking size of parts. Also there is a push towards all-in-one chips that can handle all of a phones functions with very few support parts. All you needs is a screen, a camera, a battery and a processor these days, everything else is baked int othe processor itself. So you can get a new camera, due to development times of hardware, everything else on your phone will be obsolete as well, so you might as well scrap it all and start over. The cost of the phone due to all the extra needs for interchangable parts will not make it anymore cost effective. Its a wash imho.

    1. I don’t want my phone to get any smaller. Really, I don’t. Instead I want to see this ‘how thin can you go fad die a quick yet painful death!!! Getting smaller in any other dimension doesn’t even make any sense, it would mean less screen.

      I have a Droid Bionic with the standard battery now, not exactly top of the line in thinness and yet it is more than thin enough. I like it but will probably like it even more when I get the extended battery and it gets thicker. I recently had a Galaxy SIII but that didn’t last long, it died (problem not related to size). That was definitely too thin. It only lasted me about a week, maybe I would have gotten used to it but I found it clumsy when using game apps that required quick ‘button’ pushes. I also always felt like I was going to break it.

      I do not have big fingers and I am NOT one of those people who always complain about the little buttons on things. I used to do actual coding on the keyboard of my Sharp Zaurus!

      What I really hate about thin phones.. USB micro connectors. I haven’t broken one phone-side yet but my wife went through a stack of the things, she kept breaking the sockets. We both have gone through piles of cables. I think my phones have only lasted because I really baby them. No connector so thin can possibly not suck.

      I want a thicker phone with something like an actual USB-B socket but with USB-OTG support. Better yet, give me an actual USB-B socket AND a separate USB-A one for host support. Make sure the phone body is no thicker than the connectors though. It’s not that I’m even picky about that, I just want the case to support the connectors rather than leave it to the PCB. I could probably go up to about 3/4″ and not mind the bulk. I’d even go for a full inch but I had better be getting some kind of feature out of it.

      1. Something like the size of the Palm LifeDrive, but with a flush mounted, high resolution screen covering the whole front, with one each full size type A and B USB ports (ruggedized, waterproof style). With current technology, at least 50% of the volume could be devoted to battery and there’d be plenty of room left over for a peripheral bay with a couple of connectors. How closely tied to the PC architecture is ExpressCard?

        That would (with drivers written for the phone’s OS) open up a large number of peripherals that already exist. Want eSATA or Firewire on the phone? Just plug it in.

        Give up the bulky USB connectors for the mini versions and the phone could be slimmer while still supporting two side by side sockets for the narrow ExpressCard format.

        Something I don’t much care for on many Android 4.x.x devices is they have a permanently installed, non upgradeable “SD Card” where most apps get installed. Apparently Android 4.x.x automatically installs any app that doesn’t have to be installed into main RAM to the ersatz “SD Card”. These devices usually have an “External SD” where large cards can be plugged in but without some tricky feats of hacking, apps cannot be installed nor can things like where the camera saves photos or book reader apps and others store there files.

        I have a Motorola Photon Q (silly name, it’s a completely different phone from the Photon, it’s really a Droid 4 with a smidge faster CPU, on Sprint instead of Verizon) with exactly that setup. Soon as the warranty is up I’m either using Motochopper or just going with Motorola’s own bootloader unlock (which voids the warranty) and seeing what can be done to fix that.

        1. There’s no hacking required to get apps on the removeable SD card now. Since 4.0 it’s an option in the stock rom. Just go Settings->Applications, then find the app you wish to move, and press the big “move to SD card” button. I don’t know of a way to chose the location of app data though.

          That’s awful that they can void your warranty for rooting the phone. Here in Australia they can’t void a warranty unless they can prove that the modification could have affected the particular thing that required warranty service (so a loose USB socket must be still covered because a ROM couldn’t make that happen). If they try to stiff you on your warranty cover, you talk to the relevent ombudsman and they cop a $5,000 fine from our government. You can pull then same trick with your telco if they bill you for things that the costs were not specifically explained to you (like SMS delivery reports).

    1. Yeah, it might make it possible for some sort of universal standard like the IBM PC to emerge. However in the case of IBM it was mostly an accident, and I don’t think that commoditization of hardware with the accompanying reduction in margins is something that manufacturers are looking forward to so there will probably be a lot of opposition.

      1. All the north American CDMA phone companies will be against it. They refuse to adopt the RUIM (CDMA version of GSM’s SIM) so they can control which phones may be used. The GSM companies, all you need is a SIM and any GSM phone that’s not carrier locked. You can have a dozen different phones to swap a SIM amongst.

        Some of the newer phones sold by CDMA companies, like Sprint’s Photon Q, have a SIM chip soldered inside (I assume that’s for LTE support). Could have been a socket for a normal micro or nano SIM, some people have removed the soldered chip, soldered in a socket then cloned the original chip to a standard one – or just pop in any ordinary GSM carrier’s SIM to use the phone on AT&T, T-Mobile and others.

  9. If the hardware is open enough then the potential this has for us as makers would be huge! These modular parts wouldn’t be that much different from the single board computers and accessories we all like to play with but they would be mass produced on a totally different level.

    I do think it will be harder to catch on though, especially in the wider non-maker market as it will likely be a little more bulky. I really don’t get the whole slim phone craze anyway. Come on, I carried this > 1/2 lb beast for years and it never bothered me a bit. Of course I didn’t try to put it in a pocket, it was on my belt or clipped to my pants. I know that is out these days. Whatever! I carry my wallet and my keys in my front pocket. Where else would my phone go?

    1. Agreed, I’m getting tired of people’s infatuation with thinner phones/tablets and even laptops, a thinner phone/tablet won’t make it more portable or pocketable, especially when it’s a 5″-10″ flat slab, yet manufacturers insist on making them thinner and thinner. A thinner phone is actually more difficult to hold, and just isn’t comfortable. You could double the thickness of a modern smartphone, thus giving it more battery life, add a physical keyboard (which many of us want/need) and it’ll still be just as pocketable as current slim smartphones… Hell, people already put huge cases on their smartphones that doubles the thickness, such as Otterbox, Tank, Ballistic Hard Core, among others, instead of putting on a case, manufacturers could build them to withstand more abuse as well… And the same can be said about laptops, we’re crippling laptops so that they can fit in a paper envelope, seriously, what’s the benefit of having a super thin laptop such as an ultrabook or a macbook air? They’re going inside a laptop bag anyways, so why do they need to be so thin? We’re sacrificing performance, the ability to upgrade and repair/replace certain components, battery life, the ability to swap out batteries etc… And for what? So that our laptops look better? It’s absolutely ridiculous… People need to stop worrying about how smartphones, tablets and laptops look while they’re carrying/using them, they’re not fashion accessories…

      As for Project Ara, I don’t think it will take off, mainly because I don’t see how the base will be able to support future components past the usual 1 year (at most) cycle, it’s a good idea in theory, but not in practice, the technology just moves way too fast for it to be feasible… Those who would benefit more from this would be makers who use smartphones for robotics projects or other projects that makes use of old smartphones as the “brains”, but for them, there are plenty of old and unwanted smartphones in the classifieds for cheap, not to mention the low cost brand new budget phones available these days… Someone should just make the modules themselves available, as that is what most makers really want, cheap and open 3G/LTE radios, GPS, sensors, camera modules, flash storage, SoCs, memory and especially high quality and high resolution 4″ to 5″ touchscreen displays, all with pins broken out, data sheets available and the whole 9 yards… That won’t happen, but we can dream…

  10. The concept excites me, but I think that the challenge is a society that’s learned to be spoon fed the latest in technology. The idea of running over to att and buying the new camera module as apposed to the new smart phone is foriegn. How often does the average Joe buy a new phone only because he is told its better, without a clear understanding of why?

  11. “want a phone that supports open hardware and lets you plug in “peripherals?”

    I always thought it was odd that Android phones just used USB for charging/debugging.

    Is that lack of thinking on Googles part?
    Or a sign that there isn’t many wired peripherals people want?

    1. Its not a lack of thinking on googles part, its a lack of research on yours.

      You are hard pressed to find an android device from the last year or 2 that *doesnt* support USB host for peripherals. Google also launched the ADK boards a few years ago in collaboration with arduino (the ADK essentially being an arduino mega intended for slaving off of an android device).

      For example, here is a video with my phone model:

      This also works on my brothers kindle fire with my powered hub, my brothers galaxy S2, my old nexus 7 and my dads galaxy ace as well as just about every other android device there is. Even the galaxy tab and galaxy note tablets which have a proprietary connector have USB host adaptors from samsung themselves. Same goes for the transformer tablets, asus sell an external USB adaptor for the tablet itself when it isn’t docked.

      I have used a USB keyboard (on screen keyboard just simply doesnt popup, type on keyboard and it works onscreen), USB mouse (pointer appears, left/right/centre click all do the same but you can move around and click to launch apps and play some angry birds or whatever), xbox 360 controller (retroidnetplay recognises it, dead trigger recognises it, beach buggy blitz recognises it), a few USB thumb drives and even another android device.

    2. There is an international standard that most phone companies are signatory to that defines the USB micro B as the main power and data connector (even Apple was a signatory, but they never implemented the standard).

      1. Apple (and all other cellphone manufacturers) was required by law in the EU to adopt the micro USB connector but they said “FU EU!” and instead of having the connector built into the phone they include a clunky adapter.

  12. I think EMI/RFI issues will be a big hurdle. How does one ensure that the final mix of components meets regulations? There are so many things that can affect RF… all of those interconnects for instance. Even the trace layouts have to be taken into account. Somehow, you need to make sure all of the different modules play well together in the RF.

  13. Why not? A lot of periferial was made to interact with smartphones via SD slot and some of this periferial even fit to SD-card boundings. This is will be just “extended extendable via SD-cards” platform. To be really modular it should be multi-layered, so the one base can be connected to another base via intermediate module.

    1. Or, for example, you can attach two screens from both sides – from one side large touch-screen, and from another side the screen twice smaller with keyboard or with camera block. Yes, it will be fun :D

    1. Mediocre satellite products? It’s true that there is lots of variety among Android interfaces, but this isn’t necessarily a bad thing. HTC Sense, Samsung TouchWiz, and all of those custom skins give users choice over what interface they want. Similarly, Ara (if it succeeds) will let a given user choose whatever camera-module they want, whether or not it’s made by the same manufacturer as the CPU-module and radio-module.

  14. Ah yes, those experts at Motorola that brought the WEbtop. a great concept of a cheap screena nd keyboard that docked with your android phone. Only it ran a different OS entirely.
    Do we think they will have learnt anything?

    I’d still quite like a larger screen with keyboard dock for my android phone. That’s all it needs to be. A larger LCD (touch or a nipple mouse would do) and a keyboard. No CPU or other bits required. Generic HDMI and USB interfaces to a phone and a 2.1mm 5-20V DC plug for a PSU. Battery optional & generic.

    1. Umm.. Where ya been? Webtop IS just Android now. It has been for a couple years!

      I just got a Bionic and webtop. I would have actually liked to have the original Ubuntu version of webtop. I would like to install gimp, toolchains for various micros and Libreoffice on it. Also, having a windowing system where I could have a browser open in one area of the screen and a text editor in another for web development would be very useful to me. But… that would have meant sticking with Gingerbread and I had things Gingerbread doesn’t support.

      Webtop with anything past Gingerbread is just Android switching to tablet mode. It’s useful but doesn’t really take advantage of the mouse, keyboard and additional screen space like a desktop interface would. Maybe Webtop IceCreamSandwich or Jellybean is exactly what you want?

      Meanwhile I will be trying to get Debian going in chroot for my purposes. And then there is always VNC (not to useful for programing micros though)

  15. This won’t work at all until either component manufacturers give up the ghost on proprietary drivers or Google (or someone else) invents a way to load the same binary drivers into any arbitrary Linux kernel version (through a common framework like WDM).

  16. I think the point is that the hardware for phones, if you push the limits and maybe not choose the most bleeding-edge hw, can shrink enough to get *smaller than what we consider a phone*. That’s the key thing. That means room for the extra large blocks.

    But, phones are expensive and when customers choose them they want max performance (performance can also be something percieved, i.e. not GHz but a nice designed and highly responsive UI). I think this Ara project (which is not like the phoneblocks idea at all, except you swap some smaller *peripherals*) can have some success for tinkerers and business. Being able to strap on a bar code scanner, credit card terminal, OBD-II-plug etc. will be useful. But for the big masses, maybe not so interesting. They’d be better off picking a phone with everything integrated, simply because production volume matters and they will get the cheapest phone with the best performance that way.

  17. For everyone saying that this will never work because of this problem or that problem. Here is a quote from Wilbur Wright.

    No airship will ever fly from New York to Paris. That seems to me to be impossible. What limits the flight is the motor. No known motor can run at the requisite speed for four days without stopping, and you can’t be sure of finding the proper winds for soaring. The airship will always be a special messenger, never a load-carrier. But the history of civilization has usually shown that every new invention has brought in its train new needs it can satisfy, and so what the airship will eventually be used for is probably what we can least predict at the present.

    I am not saying that this will work but, don’t be so quick dismiss an idea because you can’t see how it would ever work.

    1. Hah, peasants! You have yet to learn the hacker way!

      Seriously, though, its’ surprisingly hard to make a piece of hardware secure in that way. Virtually every commonly-used Android phone has been rooted, BL unlocked, and had tens of custom ROMs created. I use an Atrix HD, and I was watching when the first bootloader crack came out; the amount of effort put into it was absolutely astounding.

  18. Spare a thought for the future clerk, who has to fight back customers describing a vague module that their friends just got, while making hand gestures and inventing new terms on the spot to try and describe it, and requesting it in neon blue.

  19. I think this would be better if a phone just docks on a “jacket”-like device which would contain the modules. So you’d still have the option to have a thin normal-spec phone, or a full featured phone that is thicker.

  20. honestly. I dont know how well this will take off. For us techies, may be fantastic, but in all honesty, the general public really doesnt care. Most average users Dont even know what linux is, how can we expect people to change this much stuff.

    Personally, if everyone were like me, it would be fanstastic. I would absolutely DIE to get the newest CPU/CPU I want.

    There needs to be a standard however. So it doesnt become available on one device and not capable on another future modular device. I would never get it if I cant get a new chassis.

  21. When we were talking about the PhoneBloks article, my main objection was that a small player wouldn’t be able to get a decent SoC in large enough quantities. Now that we’re talking about Google/Motorola, that’s no longer an issue.

    There are basically two kinds of peripherals that will attach to the endoskeleton: those that use buses, and those that use dedicated connections. Keyboards, SD cards, etc. are examples of the former – they don’t have particularly great bandwidth requirements and can be connected over something resembling USB. Similarly, extra batteries could just be connected to a power bus of sorts, with maybe some monitoring electronics hooked up via I2C. The stuff requiring dedicated connections is harder – the screen, audio/ADC, radio, and possibly camera will probably need to be connected directly to specific pins or GPIOs on the SoC. That means that the position of each of these peripherals is constrained, which may or may not be problematic. e.g. the screen is always going to be in the same place, as is the camera.

    The other issue is drivers – the Linux kernel being what it is, changing the peripherals will probably require the user to modify the device tree, unless everything uses I2C to make itself auto-discoverable somehow (given that ARM in general is yet to do anything like this, this would be quite impressive in its own right).

    In short, it’s now feasible, it’s just a question of how well they execute (and market) it, and how much demand there is for it.

  22. Not that I know much about it I’d think it would be doing good to see modularity with a particular manufacturer’s products much less an industry wide modularity. As for the size of devices with the demand for lithium could we see larger sizes combined with less power capacity down the road? I still use some older devices with Ni-cad or Ni-MH batteries & I’m reminded as to how much a PIA they as compared to my cell phone battery. a dumb phone that I use so little I only charge it twice a week.

  23. I’ll admit to being skeptical over the whole thing, but hopeful at the same time. it is kind of suprising to see this coming out but the venerable laptop is still a monolithic slab though I’m sure at some point, given enough traction, a tablet/laptop, endoskeleton will be developed. what I fear more for the [i]blocks[/i] concept is that it will cater to much to makers and electronics enthusiasts rather than every day consumers.

    something on the scale of a PC or a laptop is tenable because the hardware is not awkwardly in hand as loosely locked modules. I look at the mic or screen modules on Ara and wonder how many people are going to fidget with their front panels until they have to hold them down with shims or tape (it’ll happen I assure you) worse still buyers remorse from people who buy the “bear bones” versions and rant about the phone being cheap without fully understanding they bought a skeleton with the lowest common denominator parts.

    at least in western economies its become common practice, for general consumerism, to just replace devices that are out of date rather than upgrade them. for some, especially if this phone gets revisions on the endoskeleton to frequently, it would make more sense to just get a slab phone and deal with it for a few years.

    I do see the potential for swapping parts though, its one reason I resisted the urge to buy a phone for so long (in the past six months I bought my first smart phone and I’m a bit underwhelmed) I want my device to be a computer that makes phone calls, something I could dock and use as a full desktop or at least a thin client with a decent VNC to my workstation at home. While the state of the Android ecosystem doesn’t quite allow that, opening up the closed hardware ecosystem for phones- short of fully embracing ARM in PC/Server builds, encourages more robust software that is not limited to the form factor of the device threw an app store. in essence creating a general computing app category that requires baseline hardware performance that can be easily meet threw sand-boxing ones device and investing in it as the need arises.

    I also hope Motorola, and Google by association, rolls this out rite because there is a lot of room for this citadel to collapse on launch. with Samsung breathing down their neck on control of Android thanks to their line of phones with heavily modded roms the software risks fracturing beyond Google’s control but at least with a modular system it opens of the possibility of compatibility with proprietary drivers. threw hardware deals Samsung has developed their growing niche of the device market by leveraging Android and if Ara, and similar devices, took hold it would level the playing field- not even restricted to Android as the operating system.

  24. I’m never happy with a computer I didn’t put together myself so why should I be forced to accept a pre-fab phone?

    Some of the things I would be thrilled to swap out?
    CDMA and GSM so I could switch between carriers.
    3G, 4G, LTE, wimax, ect.. They are always changing.
    It would be great to order a Screen on Amazon so I could replace the cracked one next day or maybe I would just keep a spare like I keep a spare battery.
    Cameras are improving pretty fast. Let’s swap them out.
    How about memory? I stuck a 32gb micro sd in my phone but wouldn’t it be nice to make that 32gb of the better internal memory?
    I miss the Infra red modems in all of my old phones. I want one of those so I can control the old IR stuff in my home.

    Great now I upgraded everything else on my phone over the last 5 years but the main processor is outdated. Putting that on a module might not make a lot of sense. Hey wait a minute? The modules are all interchangeable with other bases. I can upgrade that and stick all my modules in the new one.

    If they can make it work then I love the idea!

  25. This article really got me thinking. What kind of hacks could we do with all those modules they would be creating for phones? It would be a whole new line of cheap easily available parts for homebrew projects.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.