Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Why You Can't Build a Smartphone (joshondesign.com)
237 points by joshmarinacci on Dec 2, 2013 | hide | past | favorite | 173 comments


I was at UIQ back when the iPhone "happened". It really was as pivotal for the industry as he says. I quipped that the iPhone made us irrelevant and Android made us redundant.

I was surrounded by people with strong ambition and vision, at UIQ, Symbian, at our owners Motorola and Sony Ericsson, who wanted to build the iPhone and knew how to build the iPhone long before the iPhone was announced and before they knew anyone else was doing it.

We knew all about capacitive displays, finger-touch and so on. To be honest my own vision was tainted by Flash-based UIs and the mechanics of that, but the Flash demos I was trying to enable were fancy iPhone-like UIs. The artists had the design ascetic that was the iPhone to a tee.

But the phone manufacturer upper management were essentially very traditional hardware manufacturers with traditional channel relationships and techie vision was never given much attention.

Not taking away from Apple, not saying actually I could imagine it having turned out any different. Its simply that Apple didn't have the management constraints. Apple created new teams to make new products. The entrenched phone makers made existing teams punch out variants and incremental refreshes of the same tired formulae.

I imagine it was much the same in Finland too.

The big question is, without Jobs, will Apple settle into defending what they have - and punching out variants and incremental refreshes of it - or will they have fresh ambitious restarts?


Well they do have that "Special Projects" division.. I'm sure they'll be cooking up something nice for 2014.. at least I hope so


Yours is not an uncommon sentiment, but it’s not really a fair picture: Observers often see the Apple of the past decade as releasing a category-definer every few years (iPod, iPhone, iPad), and awaiting the next one (iWatch! iScreen!).

Really, though, Apple's blockbusters of the 2000s are best seen as one continuous thing: the iPod paved the way for well-designed, vertically-integrated mobile devices (that it was a music player is almost incidental) and the iPhone and iPad are really bookends of a single smart device revolution.

Seen this way, Apple has had two world-changing category definers in its history, the Mac and the iOS device. That’s one every 24 years, if we count to the midpoint between the iPhone and iPad introductions. Will they do it again? Maybe, but I wouldn’t expect it to be just around the corner.


Huh whatnow?

The original iPod (and modern iPods minus the Touch) don't run iOS at all. The iPod gave Apple the time to stall until they could build what Steve wanted to make in 1999, but the world didn't have the capacity to create yet.

These days, Apple is the forefront of both industrial design and fabrication capacity. They aren't in the past—they are in the present and they are moving the world along with them, not waiting for it to catch up to them anymore.


Your parent is grouping iPod/iPhone/iPad. Yes, the original iPod didn't run iOS, but he's just using it as a moniker. We could call them iDevices if that pleases you more.

The iPod wasn't just an infusion of cash to keep the company alive, to me it really looks like the start of the whole mobile line that today runs iOS.


"they are moving the world along with them"

You tried iOS 7, didn't you? I sure hope they are not using "Apple Maps" to move themselves and the world.


Apple Maps has come a long way since the iOS 6 launch. Google Maps is still undoubtedly better, but Apple have been working hard on it.

iOS 7 is not bad. People are still afraid of change (look at what they said about the original iPhone, now look where we are) but ultimately it was in need of a refresh.


> Google Maps is still undoubtedly better, but Apple have been working hard on it.

The big difference is that Apple only allows Apple to make things better, where Google allows the world to do its work for it.

Example: The URL for the Boston subway system (the T) is "mbta.com". Nearly every single T stop showed up as "mTBa.com" in Apple Maps when it was released.

In the intervening time, they've fixed maybe 1/3 of the stops to be the correct URL - and almost all of that 1/3 in the last two months. (I've been watching, because I thought it was interesting.)

First of all, you'd think if they fixed one, they'd fix all of them.

Second, this is the sort of thing that when you report it to Google Maps team gets fixed within DAYS.


Again, Google has had a lot longer to get this right. I still use Google Maps whenever possible but Apple are hiring for multiple positions in my country for Maps staff shows they're at least /trying/ when we don't even have an Apple store.


But they are not "moving the world" in any way except for hardware. iOS 7 is derivative.


iOS7 is more like iOS6 than any it is anything else.

iOS has certainly adopted features from other platforms over time. But what has been adopted in the other direction is far greater.

If you are referring to the removal of skeuomorphic style elements, then it's true that android and windows phone were always less skeuomorphic than iOS, but that's irrelevant to the technology.


True, even the hardware iterations aren't exciting. Rebranding the 5 as the 5C is disappointing.

Where Apple do lead is in software, Android is getting better for the regular user but it's still not even close. Vertical integration works.


for varying definitions of "regular users" and "works"


It depends on whether upper management gives special projects the importance they need or not. Almost every large company has a research division, but in most, it is the playground where they keep the guys that are too smart and competent to fire, but that have ideas too radical for main divisions' heads to accept.


UIQ3 was amazing, and my M600i had the greatest keyboard on any phone ever. Thanks for making the OS for one of my favourite smartphones :)4


I miss that keyboard, I really do prefer physical keyboards to software ones.


  design ascetic
The words are supposed to be "design aesthetic".


Maybe they're going for something extremely minimalistic.


I really thought phone manufacturers at the time where aiming for better Windows Mobile phone, and had nothing like the iPhone in mind or at least nothing more than a napkin corner with a 'moon' tag attached to it because some R&D/Designer broke someone's b*lls enough to get a little attention.


This guy is missing the fact that in a lot of European countries, and elsewhere in the world, carriers mean squat. The carrier is often simply the provider of the SIM card you put in the phone, so carrier support is in no way needed for a phone to succeed. You don't need to be successful in the US to be successful.


Yes, but those are smaller markets.

As much as we're loathe to admit it, the USA represents the largest single market of wealthy (in global terms) customers. That means the USA is the place high-end consumer tech lives or dies. Other markets are more fragmented (so they're more expensive to sell in) or less affluent. There's lots of exciting things happening in Brazil and India, but those are lower-end spin-offs.

I'm a Canadian. Our entire consumer market is based on getting America's sloppy seconds. I have no illusions.


I wouldn't say so. Without carrier restrictions, EU is much easier to sell to. In fact, in EU, the carrier is totally worthless, and the phones they usually bundle with contract plans are in general the lowest-end phones you can have. Even the iPhone deals are usually meaningless, since you can have the phone for less then a 10% difference, with the added choice of getting a better data plan at any time you wish.


Lowest end? The Samsung Galaxy S4 and HTC One aren't exactly low-end, and those are the two phones everybody is offering now. The ones you get for free with low-end contracts are low-end, but you can get some pretty nice discounts on high-end phones too.

At least apparent discounts. You pay for them through your contract, and if the difference with a sim-only contract is small, that's because sim-only contracts are way too expensive.

Carriers are big, but it is possible to get around them. Most people get a phone with contract, but a lot of people really do go sim-only.

Fairphone is a nice example of a small Dutch organization that makes a revolutionary new phone (not available in the US). It's very successful so far, but also because they're not inventing their own platform; it's still Android. But the hardware is indie, so that at least can be done. (And a major carrier ordered a 1000 of them.)


All these articles do. It's annoying.


Every european country has own laws, culture and language. They are individual markets, and each market is too small to break even after all the R&D and hardware costs.


> Every european country has own laws, culture and language.

You're completely disregarding the EEC and European Parliament. As an aside on language, while on the train to Brussels from Köln, I was chatting with some Young Greens who were en route to a pan-European Green conference. One was from Serbia, one from Germany and one from France, and we were all speaking in English, it tends to function as a lingua franca, so language has never been a barrier for me yet. The multilingual capacity of Europeans really puts me to shame, my German barely goes past "I would like a beer" and "I would like a coffee".

But yes, they have different languages, so i18n is a thing you have to pay attention to, and yes, they have different cultures and markets, so how you sell and engage people can differ. But you can recruit a small cadre of local sales experts to do your sales - a sister company of ours develops in the Czech Republic and sells in Poland, Romania and Turkey. The European market is hardly as balkanized as you make out.

If what you say was true, then iPhones would have presumably never impacted in Europe - after all, every european country has own laws, culture and language.


Your "counterexample" is young political activists, probably local party leaders. Those are multicultural, openminded and all. What about the average middle class, representing probably 90-99% of the buying power?

I agree that it's not impossible to sell in multiple countries at once, but it's just times bigger investment to get smaller combined market than US.

For Apple starting with US was a justification to spend probably billions on iPhone. Market was big enough for this gamble to make sense. You can't justify huge investments if your plan to start in Poland and then expand to Romania and Turkey. Or did that company go into all 3 countries simultaneously? And, btw, combined GDP of Czech Republic, Poland, Romania and Turkey is 1.65 trillion. GDP of USA is 16.86 trillion.


Is that not similar to how the individual states in the US work? (Laws, culture to a degree, but obviously not language).


Similar? compare California to Texas, and then compare Spain to Czech Republic. Apple needed to partner with one carrier - AT&T to sell phones in all 50 states. What's the minimum number of carriers you need to partner with in order to have a friendly carrier in each EU country?


They still mean something in those countries, because carriers there still have agendas, marketing budgets, partnerships and retail stores even though they may not dictate cell phone features.


that only ever affects the bottom of the market.

where i came from, you'd actually made fun of people that had the phone from the carrier billboard.


Carriers don't mean squat. They are the king makers.

They control a section of the market that every smartphone mnnufacturer relied on. The Radio frequencies. This gives them power via their marketing subsidies, and it was this power they leveraged before they screwed up and ceded control to the platform. It used to be the case that the carriers controlled your wallet. Every single thing you did on the phone, the money went to the carrier, they held your credit details. AT&T screwed up and gave up consumer power.


The fact is that in a lot of European countries, they don't even have AT&T.


They let the cat out of the bag.

Once AT&T ceded control and the popularity of iPhone as a carrier switcher came along the other carriers had to fall in line. This is not the case with Android, they don't control the updates. (Except in a stealthy way).


And Google put straight back in in a bid to gain carrier favour with Android.


Article is fine but misses a few salient points.

Appstores did exist, Docomo created the first appstore and any phones that were released by the carrier had to comply to Docomo's requirements. Nokia created an appstore with N-Gage, it didn't work and created a carrier backlash.

What Apple did was to stealth in their app store. It was all due to a strategic mistake by AT&T. Sure the Apple was a great device, but remember, the first version was 2G with a crappy camera and an amazing UI/UX and already seen as obsolete in developed 3G markets like Japan (I was working at Nokia in Japan at the time).

AT&T's mistake? Unlike any previous carrier or vendor relationship, AT&T ceded control of software updates to Apple. Apple could update the phone any way they wanted. They already had enough market power in the US that would prevent carrier boycott. At the time, all the carriers were trying to copy Docomo and push out their own appstores, and this decision by AT&T killed it.

In terms of smartphones being a commodity, it already is, anything that sells int he hundreds of millions is effectively commoditized, that doesn't prevent new players from coming in to try and disrupt the market. This is probably the third disruption in mobile phones, from dumbphones -> feature/early smartphones -> current smartphones, however it may be 'too soon' as each cycle has been around for about 10 years each.


I always considered App stores the same as Linux repositories with a payment option.


Curious that the article doesn't mention the word "patents" once. That's what stops you from even starting to build a Smartphone of your own - you'd violate dozens of patents and be sued into oblivion. The only reason manufacturers are able to do it now is because they all have portfolios of their own and are in a state of cold war with everyone else.


Most things needed to make a smartphone is covered by FRAND patents. Software patents are of course a problem for anyone working in the software industry, so that's not limited to smartphones.


"Most things"? Haven't we just learned that if you make a smart phone that looks like a smart phone (rounded rectangle, button below screen) you can (and will) be sued?

I suppose the fact that stuff like that is the area where big legal battles are fought, does imply that a lot of the "hard" patents (radio spectrum/coding, noise cancellation, audio encoding etc) are indeed covered by FRAND patents, though. It still seems to be a bit of a leap to go from there to saying that you'd not run into patents when making a smart phone?


No. The argument re "rounded rectangles" (actually a meme invented by people like yourself) is about trade dress, which in the US is covered by a design "patent", which is not the same thing as a technical patent. This is not difficult to understand. To this day the level of denial around Samsung blatant imitation of the iPhone is staggering.


I'm not sure what (who?) "people like myself" are referring to, but you are indeed right -- while Apple tried to sue on the basis of what I believe is fair to sum up as "something looking like a smart phone (granted the natural technical progression to larger and better touch screens)" -- they only won an injunction for "something that looks confusingly like an iPhone".

I still think that's a rather silly ruling -- and my point was that you could be sued, not that you'd loose (if you had a few million to spare on defending yourself) -- on the basis on a lot of different patent than just the type that are covered by FRAND agreement(s).


By 'people like you' I mean those that continue to perpetuate the meme in an incorrect and dismissive manor.


> To this day the level of denial around Samsung blatant imitation of the iPhone is staggering.

The problem is that I can't think of any other design that a modern smartphone would have other than "piece of glass encased in plastic/metal, with maybe a button underneath". As far as I can see the screen is the phone, and given that constraint there isn't really anywhere else to go.


The thing is that people had been building smartphones since before the iPhone, and none of them looked anything like that. Buttons. Everywhere. The fact that the iPhone form factor is now considered self-evident is a demonstration of Apple's design strength, not of how obvious the form factor is.

Don't forget that to make that form factor work, you need a whole software stack that is able to replace the physical buttons adequately. You need an on-screen keyboard capable of at least equalling a physical keyboard. You need to have virtual buttons that are easy to use. You're going to be doing a lot of scrolling on that tiny screen, so you're going to need some pretty good graphics routines. To make that scrolling anything other than an awful experience you're going to need low-latency right from the capacitive input right through to the content actually moving on the screen. Not easy, especially with the chip used in the original iPhone.

Think about Android. If Android had have been first to market, and they had actually undergone the post-iPhone transformation without the iPhone being present (because don't forget, it was originally going to be a phone with a physical keyboard), they would have released a touch-screen phone with Android 1.0 on it. Android 1.0 was awful. Battery life was rubbish, graphics performance was pathetic, even the keyboard was full of fail. If that had have been the defining capacitive screen smartphone, we may have decided that touchscreen phones were not the right direction.

It took Google, with a large chunk of the smartest engineers on the planet, about 4 years to ge to a point where Android was truly competitive with iOS. It took about 5 years for Microsoft to get Windows Phone to the same place.

Without the software engineering effort, the capacitive screen was not an obvious solution. The fact that Apple invested in that effort made the benefits obvious to everyone, but this was not the case pre-2007.


I don't think it's fair to say the iphone form factor wasn't out there. There was a rich variety of full-screen touch or pen phones (palm, pocket pc, some one-off designs like the lg prada), and there were models designed with a capacitive screen. A lot of people were trying to corner the smartphone market by foregoing a physical keyboard. I still think palm's graffiti was a better way of inputing complex text than any smartphone on-screen keyboard.

It's not that nobody was trying to figure out how to build a smartphone with a big touchscreen, just that they were approaching it from the enterprise space. Usability wasn't important because it wasn't an enterprise selling point. Pen input was important because signature input was key for enterprise. Apple ignored enterprise at first and designed for the consumer market, which allowed for blasphemies like throwing out pen support. The big change was making smartphones a consumer product instead of an enterprise product.


> It took Google, with a large chunk of the smartest engineers on the planet, about 4 years

Google didn't take people off of other projects, and turn every engineer that they had at the problem of "making Android competitive," so this statement is a bit of an exaggeration.


>""piece of glass encased in plastic/metal, with maybe a button underneath"" That isn't what the design patent says. It's far more nuanced.


> "Most things"? Haven't we just learned that if you make a smart phone that looks like a smart phone (rounded rectangle, button below screen) you can (and will) be sued?

Only if you sell in the markets where such patents matter. Note how Jolla is conspicuously absent from the US.


"Most" is not enough. If you have a single not-FRAND patent for Samsung, one for Apple and one for Nokia that you are allegedly violating (even if you're not), then consider your product dead before arrival.


Yes, there are hundred of thousands of FRAND patents that will cost "only" a small percentage (1%-5%) of the product's selling price to license each.

Anyway, no, the patents don't stop you from making a phone for yourself. Nobody will sue you if you do that.


I work in the software industry and have never once worried or even thought about patents in the things I work on. It really depends on what field you work in.


This is only a valid strategy until you have enough money/exposure. Then you get targeted by asshead extortionists and the law is on their side. After the first lawsuit, you will never "never once worry" again.


Well that and whether you are making millions from software sales.


I was about to say the same thing -- glad to see I'm not the only one who recognizes the real reason why there will be no more competition in this space.


There's a very strong argument here but the claim appears to rely on the author's expansive definition of "smartphone", in which he encompasses app stores, cloud servers and networks. Perhaps he has a point about the difficulty of investment in taking on these well established players in the field, but I don't feel it's entirely accurate to claim that smartphones themselves cannot be successful.

A new phone with a bare-bones linux-like OS designed to piggy-back on existing networks may have a small place in the market, and others like Jolla that seek to grow toward something big may pick up fractions of a percent as well. Even surviving a decade with a low, but consistent, share is a success (e.g., Apple in the 80s).

Finally, there's the novelty and growth of industry to consider. The current big players in the smartphone market are already scrambling to address the emerging wearable revolution. The "internet of things" is right on the horizon as well and there's no telling where the chips may land. To be fair, the author did mention innovation as a key to potential success. We should consider the grander implications of a constantly evolving business model and offering in that sentiment, though. It's not just a matter of making a neat new feature, but of the entire game changing every 5 years.

While the author makes plenty of good and valid points, I don't think the outlook need be so bleak.


Exactly rather than focusing why a smartphone can't succeed, I'd like to see how and why a phone can succeed.

If anyone thinks that the mess in the Play Store and the development environment of Android doesn't leave a bad taste in your mouth they should really re-consider thinking that Android is going to be around forever. I would like to ditch this platform the first moment I get. There are a lot niches that are not addressed by market players today, and the overlap between them can wedge a dent in the market. Google has serious weakness in hardware, and it looks like they will never get over that hump. Can you honestly expect them to optimize performance and battery life, if they can't even make their own devices.

Niches like:

Open-Source/Flexible development environment

Sustainability

Emerging Markets

Actual Innovation

Disruptive technologies and emerging markets shouldn't be counted out either. Nor should finding ways to monetize open-source. People, developers especially are increasingly looking for open platforms/devices, that market is NEVER going to get smaller only bigger.

Imagine actually being able to use Javascript or Python or Clojure for your apps, or scripts. Will you actually discount a platform where you can program without managing a thousand configuration files or worry about the gatekeepers(Google/Apple). See these problems aren't necessarily problems in Linux(not to say there aren't other drawbacks).

The great(maybe not?) thing about innovation is that if your to busy looking at the technology today your going to completely miss out what is coming next. - At least it's great if your competing.

I can't help thinking about what Jobs says, "Experts" are clueless.

>Experts—journalists, analysts, consultants, bankers, and gurus can’t “do” so they “advise.” They can tell you what is wrong with your product, but they cannot make a great one. [1]

[1]http://news.cnet.com/8301-13579_3-20117575-37/what-i-learned...

People seriously underestimate the power of a great product. Products that don't rely on massive feature lists that mean nothing a day after buying your device. I regret falling into this trap and wasting precious time instead of thinking about creating a better product experience. But again this is easier said than done.


I'd like to see how and why a phone can succeed.

There is an enormous market for a good dumb phone. Or, well, idiot savant phone: makes calls, makes calls well, and does nothing but makes calls. Between those who just don't want a smartphone, and those who carry enough computing power to not want yet another Cray 2 in a pocket, there's a large niche for a phone that is just a phone, syncs contacts effortlessly, eliminates every unnecessary/redundant call step, is just the right size & shape, uses every available network aggressively & seamlessly, leverages every service to save money, and has as clear & pristine a sound as makes audiophiles drool. Instead, we have fat folders with idiotic interfaces designed to make you spring for a smartphone just to get something thinner & easier.

I saw Smith Corona collapse. A market for typewriters remains to this day, decades later, untapped. Dumbphones are following the same path: a viable market wrecked by an industry dazzled by the sometimes undesirable glitz of a competitor.

Whither the iPhone nano? Many want it.


I don't really believe this market exists. And even if it does, I expect it's tiny - otherwise, manufacturers would be targeting it.


Last night I had dinner with my parents. They lamented that they could not find a phone for my elderly grandmother. They just need something that just makes and receives calls, is dead simple, has (very) large buttons, and is loud. Where is this phone?

We have an ageing population.


While this might not be the exact answer you were looking for, but people can buy a cheap android phone and put this launcher.

https://play.google.com/store/apps/details?id=name.kunes.and...


Something like http://www.telstra.com.au/mobile-phones/mobile-phones/telstr...?

(There was one with larger buttons, but I can't find it)


They lamented that they could not find a phone for my elderly grandmother

Check out Doro (www.doro.com/). Quoting from the top of their website:

"Doro is a Swedish company developing telecom products specially adapted to the growing worldwide population of seniors."


iNO Mobile here in Singapore has a whole series of elderly-targeted phones, here's one:

http://www.inomobile.com.sg/2012/07/ino-cp10/

actual retail price now is SGD 79, which is about USD 65. I'm not sure if I can get these at our 7-Elevens, but I think they have something similar with large keys.


There is an enormous market for a good dumb phone

I just ordered an Nokia 515 off ebay. Hopefully it will be what I'm looking for.


another idea:

a credit card sized e-paper device that can only do sms. It could be efficient enough to be powered by a peltier chip, so no need to charge it ever. Seems perfect for the text generation


Not enough power for the RF transmission; even if you go off-network while not in use (so you can't recieve SMS), then you still end up needing to do quite a few 2W burst transmissions to get back on and send the SMS.

Also, peltier requires a temperature difference; keeping it warm in your pocket is not enough.


What you describe seems exactly like the target of Firefox OS.


I actually hate the developing for the web, so FirefoxOS has no real interest for me. I'm grateful for the work those guys are doing, but I will never create HTML5 apps. I'm more of a systems, and robotics programmer and I love it but somehow these interests can never overlap with the way FirefoxOS functions.

So there isn't much of a value proposition for me.


Sorry, but you don't seem to be offering anything new here. We know there are large capital costs for hardware businesses. We know network effects exist for phone applications. We know UI performance is a key issue in smartphone adoption (currently).

Maybe "It's Hard To Build A Smartphone In 2013" would be a better title. But to suggest people will be using iPhones and Androids in the year 2050...Yikes. I pray that future doesn't come to pass.


The way I read this article, It is less about building new smartphone but more about the next technology shift that is going to change the direction. Basically, as oft repeated Ford statement goes, 'if you asked people what they wanted, they would have said faster horse'. So, at this point, smartphone business is settling into a faster horse business.

So, I would really expect the next big thing to be not a be an iPhone but something utterly different. Neural implants, anyone?


The biggest problem the smartphone solved was boredom. That's the reason for it's popularity.

At it's core ,boredom might be manipulated neurologically, so maybe you won't feel bored and just feel content with whatever. If we're going to aggressive means as neural implants ,this seems a more interesting option.

Another options(without opening skuls) is something that offers far better experiences.

Maybe more immersive (up to a point we can't differentiate from real life).maybe something that enriches our daily interactions without interfering with them. Maybe something that help us manage emotions and brain states optimally. Maybe something that augments fluidly our cognitive capabilities without interfering with what we do.


It solved boredom by enabling to play a game a megadrive would run.


Yes but a megadrive did not fit in your pocket, and the game gear, while backpackable, had awful battery life. The smartphone is successful running game gear level games because it will do so all day, and has a setup time of four seconds from impulse to game.


Plus the game gear was more a master system than a megadrive even if both would run columns (I was wondering if ppl from the US understand what a megadrive is since I think they call it genesis). Talking about battery life, I had a lynx which was even worse I think.


I think you've got the tone right. Also... I hope you're right about Neural implants! I've wanted one ever since I read that MT Anderson book in JR high school (minus the dystopian homogeneity and fields of filet mignon).

Maybe Amazon will debut them on 60 Minutes prior to cyber monday 2014!


I'm sorry, but fields of filet mignon sounds great. Make it fields of bacon-wrapped filets and I would call it my own personal utopia (for as long as my heart held out)


> So, I would really expect the next big thing to be not a be an iPhone but something utterly different. Neural implants, anyone?

Google Glass seems like a good candidate.


Yep, augmented + wearable tech seems like the future


They won't be. Smartphones will be disrupted by something else. People aren't using PDAs anymore either; they use smartphones. Eventually, something will come along that will make people stop using smartphones.

His point is that you can't be successful building something that is essentially the same thing as a smartphone. You can build something that disrupts the current smartphone market, but it would likely need to be something more than a smartphone (a la Google Glass.)

The network effects you mention are exactly the point: network effects are especially powerful in things like mobile app stores because they create a tipping point market. That means if an early entrant or set of entrants into the market can capture enough market share, they create extremely high barriers to entry. The more market share they have, the more powerful the network effects are and the less likely a new entrant is to have success.

If you're a smartphone developer, chances are you develop for iOS and Android. It would take tens of millions of users adopting a new platform for you to start developing for this new platform. But you're not likely to get tens of millions of users without a large app library.


Particular areas of software development rapidly change them plateau. I have an issue of Byte magazine from 1993. It was all about how RISC processors and next generation microkernels were about to take over the market. And yet here we are. Every new phone runs either an OS built in the 1980's (iOS), or a 1990's clone of a 1970's OS (Android). A RISC had a nice little run (ARM), but ancient x86 seems poised to wipe it out just like it did with all its previous competitors.


Modern "x86-based" CPUs are actually mostly RISC architectures under the Hood. Instruction sets are a mere front-end nowadays and don't tell much about the inner workings. Think of bytecode.


>Maybe "It's Hard To Build A Smartphone In 2013" would be a better title. But to suggest people will be using iPhones and Androids in the year 2050...Yikes.

Don't take any title literally. He is obviously not talking about 2050 or 2525 (if people are still alive, still alive).

He's talking about the market as is -- and an obvious timeline of around 10 years comes naturally from what we know of the tech industry and the pace of change.


> Challenge: you are now competing with the iPod Touch.

It disappoints me that nobody does this. Every attempted "competitor" I've seen that stands against the iPod Touch either costs more than the touch or is reviewed as unusable garbage.


> Every attempted "competitor" I've seen that stands against the iPod Touch either costs more than the touch or is reviewed as unusable garbage.

Maybe this speaks to Tim Cook's operations wizardry? If no one can get components at the same quality or price as Apple then they're stuck either selling a crappier device for the same price or a good device at a higher price.

If you can't beat Apple in hardware price or hardware quality then you'd have to win on software cost or software quality - sounds like a perfect place for Android. It's likely that the companies with the resources to do this have decided that the market is too small or they don't want to cannibalize their phone sales. Otherwise you'd think HTC or Samsung would be releasing Android-based media players.


Samsung made a line called Galaxy Player. The cheapest one was $150 and sported a 3.6in 480x340 pixel screen, at a time when you could get an iPod Touch for $200.

There expensive one, the Galaxy Player 5, cost more than Apple's offering.

And they all ran Android 2.3

Similarly, there was Archos. They made a whole family of Android player devices at the 2.3 era, and where infamously buggy.

Sony made the Walkman Z as a phoneless counterpart to the Xperia Z. It was $250. Supposedly a fantastic device, but that's not really undercutting Apple very well when Apple was selling iPod 4 devices for less.

I mean, Black Friday doorcrashers this year were iPod 4 at $130. I can't think of any android counterpart I'd want at that price. Maybe Motorola can figure out a phoneless Moto G with the right nudge from Google.

A big problem is also marketing - these companies are only now figuring out how to market their handheld devices in the carrier-controlled countries since up until then they just handed the devices off to carriers and maybe put up a few ads.

Meanwhile, Apples had their own stores.

Samsung could stop shoving money into ads and could start putting together coherent store-displays of their non-carrier product line. Sony should revive their store-brand too - they were Apple back when Apple made beige boxes. Sony could be Apple again - they've stopped the race to the bottom and seem to be investing in quality. I'd love to see Sony put together a coherent adversary for Apple using the Android/Win8 ecosystems. They even have the kind of market breadth and to put together a counter to the iTunes/App Store with their gaming and media holdings.


I used to think well of Sony but my PS3 has lowered my opinion of them as much as the issues with Sony laptops that family members own. I'd love to see them invest in quality and turn things around but I don't expect them to. It's like they've lost the ability to think through the whole experience so even their nice products have extreme low points.


Sony hit a quality nadir 5ish years ago, from what I've heard they've decided recently they can't compete with Samsung and LG at making cheap crap with good components and they're trying to be a quality company again. The Xperia Z hawt secks for example. I haven't touched a PS4 yet or a VitaStation, so I can't speak to their other flagship products.

I'd love to see Sony take on Apple in making quality well-designed equipment and metal laptops. Even if Sony is just as bad as (if not worse than) Apple on IP law.


They still make the best reasonably-priced studio headphones in the business, the MDR-7506. It's a fairly old design at this point, though.


So do you think they're not trying? Or it's just really difficult because of Apple's operations efficiencies? It seems like they could make up for Apple's hardware lead (cornered market on flash, vertical integration) by starting with Android and winning on software development time but maybe iOS is mature enough now that the cost to continue building it out is the same as the cost to repackage Android.


>>they're stuck either selling a crappier device for the same price or a good device at a higher price.

Or they just need to accept a lower gross margin than the ~50% Apple gets.


If you have the money/competence/manpower, there's more upside in selling phones in the current market. And if you're making phones, the carriers would really rather you not dip your toes in data-first devices that might accelerate their dumb-pipe destiny.

Which leaves the devices we see as sadly inevitable.

Minus someone with the power/resources of, say, Amazon or Microsoft, doing it carriers-be-damned.

Though I'm still not sure why Apple is so slow to this party. A $300 "iPod Touch With SIM slot" sounds like a great product for an awful lot of overseas markets. (Where a considerable amount of their sales and growth are.) Though maybe that's where the 5C experiment is going.


Moto G 2/3 of the price? Nexus 5 for $50 more? It doesn't have to be a "music player" of sorts, when the iPod touch is an iPhone without the phone part. So as long as there are decent Android phones out there with the same specs, that's the alternative. You could buy a Moto G and never use it as a phone if you don't want to.


You are a bit mistaken on the price differences. The 16GB iPod Touch is $230. The 16GB Moto G is $200 and the 16GB Nexus 5 is $350. The $300 iPod Touch is 32GB, which is $100 less than the equivalent storage size N5 (and there stops being a Moto G at this point). So if you have an extra $100 you could get an N5 instead of an iPod Touch - or you could spend that extra $100 to get a 64GB iPod touch.


How's the headphone jack on those devices? I've had terrible experiences with Android devices in that respect, including the Galaxy Nexus. Unusable in a matter of weeks, whereas my iPods have always had rock solid connections even after years of abuse.


I've a Samsung-based phone where the headset-jack (it has a mic, it's a headset-jack not headphone-jack) has taken years of abuse and still going strong. Jogging with an armband and using it in bed and getting snagged and clipping it to the jogging stroller and subsequently forgetting I was tethered when I step away.... none of that did damage.

My phone is a Samsung Focus, which came out a year before the G-Nex. I'm surprised the G-Nex gave you so much trouble.


Yes, but the Moto G isn't being marketed for that - consumers don't even know that usage is possible. Unadvertised features don't exist.


Tons of games are sold on Touches, still a big part of handheld gaming. Noone has even challenged it, yes tables/pads now are there and minis to add more options but for a long time noone even tried after Apple. The game appstore market was initially won with iPod Touches.


I began reading his article hoping to have him talk a little bit more about smartphones in emerging markets, but he barely touched on it. Africa has arguably the fastest growing middle class in the world right now. They're also poised to make a cultural leap far greater than that of China's in the 90s. I read that 60% of Africans will have smartphones in 2019. The continent is already using smart technology in innovative/cost saving ways (i.e. sending funds via sms).

It seems to me that if I am looking to join the smartphone industry, I make myself indispensable to Africa. I think Africans are less likely to care for features like Siri or luxury components. My thought is that developers for devices in that part of the world will have the freedom to think outside the box--what will a smart phone look like on a continent with minimal financial resources, but avid interest in quick data and communication? What do people want from their phone when they don't have necessarily have a computer or strong infrastructure?


This guy is so ex-Nokia, it's sad. He rates the latest Maemo/whatever as a higher chance of market presence than Firefox OS. OK, Nordsman... we understand you are grieving for your lost mobile device business... but you can do it in private.

Linux, via Android, commodified smartphone platforms by giving away the OS and thus commodifying the position of mobile device hardware manufacturers. This was coming anyway, but was a brilliantly timed commercial play by Google to head off Apple's strength in devices. Google's allies were the carriers, who were unhappy with Apple owning (via signup-time credit card) who they perceived as their customers.

FirefoxOS and that new Jolla thing the Nords are working on basically use exactly the same hardware and OS platform (ie. Linux). There's no huge investment there, the investment is in higher layers of the software stack, and in FirefoxOS' case in marketing to different segments (the developing world, which still has 'growth', and necessitates less competitive component acquisition for manufacturing batches).

To say a small team can't do something similar is plain wrong. I'd say the effort required is similar to launching a new Linux distribution. As long as there is a niche you can cater to, there is no reason it won't survive and thrive. CyanogenMod is only the beginning: cross-pollination is a good thing.

Start with preferably https://github.com/mozilla-b2g/ (or otherwise http://source.android.com/source/downloading.html) and begin thy mission to conquer portable computing!

(Edit: Added content after the initial paragraph after it was downvoted. Fair enough, it was a bit snarky :)


First paragraph snark aside, I agree.

To say that any tech market is completely closed to new entrants seems slightly ridiculous. Sure, the barriers to entry have increased - if you're chasing mass-market appeal, a strong app platform is as important as excellent hardware. However there will always be some niche to serve or an opportunity to tackle bloated or less open, established competitors.

I think Josh's option is very biased and colored by operating in a certain type of environment. Not that it's any less valuable or valid than any other, but I disagree with him.


The approach by Jolla and FirefoxOS guys is actually quite brilliant. How to get into markets where it's relatively cheap to make a new android phone but anything else requires very expensive contracts with the manufacturers?

Solution: Just use the android drivers directly. In my Jolla phone there is the good 'ol android /system/lib/. As an example the libGLESv2.so (the OpenGL ES 2.0 library) in /usr/lib/ simply opens the counterpart in /system/lib with android_dlsym and calls the equivalent functions (and in case of floating point arguments moves them to int registers). And what's the overhead of this? 9 instructions and 2 loads (for the function pointer). So it's essentially completely free.

Same approach applies to rest of the system. They just take the core of the android and run on top of that with just the bionic -> glibc layer on top. Brilliant.

The actual lib can be found at: https://github.com/libhybris/libhybris

That means they can basically just walk to some SOC vendor (like they did to Qualcomm) and say "I want to buy a mid range Android SOC" and they have everything ready.


* A few new Android manufacturers may join the game*

For anyone interested, Fairphone are doing so pretty successfully right now.

http://fairphone.com/


Fairphone's website says they've sold 25,000, and have another 8200 on the way.

Apple sold over 9 million iPhone 5s and 5c models in the first weekend they were for sale.

While it's impressive that an independent group is putting out a phone, their sales are orders of magnitude away from what Apple puts out in a single weekend. Sure, they're "joining the game", but they're not in the same league.


This comparison is just silly. Look (guess) at Apple's development budget for the iPhone. Look at Fairphone's (or again, just guess). Look at the size of Apple. Look at the size of the Fairphone team at the Waag Society. I don't think I need to say anything else. This is majorly impressive.


Of course, you're both right - it's very little compared to the market leaders, but for a new player it's definitely impressive. Especially considering they were all sold before they were even made.


The power of crowdfunding! That is really the major game changer that hasn't been taken into account yet.


I think its impressive, nonetheless. I do wonder what prices they were able to get components when they sold a "mere" 25,000 though.


I once read a lovely article online (which I've never been able to find again, I think it was by a founder of tripod?), recounting the early days of search engines. He was thinking of starting one, but the market was already filled... then came another search engine, and now there really was no room for another one... this iterated over several engines til he finally got to google. And added: now it's really filled...

Of course, today google still rules, yet they act incredibly threatened by facebook; and niche engines like duckduckgo are making inroads.

Is this true for smartphones? All we can really say is that he (and I) can't think of a way to enter... but that doesn't mean there is no way.

  Just as I wouldn’t suggest anyone build a new line of PCs or cars
Like electric cars...


Seems to me it's easier than ever before to build a "smartphone": Android has everything you need and is open to everyone. What's hard is to build a smart phone operating system. And in some ways (not all!) that's a good thing - you need a pretty big justification to reinvent a giant wheel. If you have one, great. If not, do your innovation on top of Android.


Agreed.

I am still waiting for the Facebook Phone. They already have a fully customized Android Home Screen. The biggest problem is probably that Facebook has no experience in selling physical products.

Anybody could slap together an Android customization. On the other hand, what is the point? If you do not profit from selling the hardware (like Samsung) or the software (like Google), then why sell a smartphone? Maybe Salesforce could sell enterprise smartphones bundled with their software.


Mr. Ondesign (can I call you Josh?), your website sure does use a lot of CPU when it's in the foreground. What is the deal there exactly?


Animated spinning ... things... rectangles? Around the planet logo. Rendered into a canvas without any throttling.


Hmm. I could have sworn I was using requestAnimationFrame.


Maybe. If you are, then you're doing so at 60fps at least, when this scene seems like it'd be fine slower than that.


From the article:

> Small to medium businesses will use apps on standard devices like iPads

I used to get lunch at a Specialty's sandwich place in SF. You could place your order at the counter (to a human, who would mess it up) or use their iPad app at the self-service counter (this got the order right every time). The iPad app worked well, but it was a source of bemusement to me that the self-service counter was just a counter with a row of iPads mounted on it. I once hit the home button on one, to see what would happen, and was taken to the home screen. I feel like for business use you might want more of an ability to lock the device down.


You can lock down iPads via the Ease of Access feature. Apple uses it to lockdown iPads in their stores, for example. The restaurant didn't have them properly configured, although it is an oddly termed feature in the settings.


A sushi place* I went to had an iPad on each table with an app that let you order stuff to the table, very handy I must say.

*Sakae sushi in Beraya Times Square, Kuala Lumpur.


cough Ubuntu cough

In all seriousness, if he means smartphone from a hardware perspective, I might actually agree. The startup costs there are crazy high.

But, if he means software, no. Ubuntu is doing it and doing it very, very well. I've said it before, and I'll keep saying it. All Android using hardware manufacturers are crazy to continue with Android now that Google owns Motorola. Look at the Moto X. Nothing can compete with it and no one will be able to in the future.

Samsung and the rest should very actively be looking to remove Android from their handsets as quickly as possible. Ubuntu is by far the best choice.


Ubuntu is doing it well? How many phones have they sold? What is their installed base? What % of marketshare?


It's this type of comment that really disappoints me about HN these days.

Ubuntu just released V1 of their phone in October. Anyone who knows phone manufacturers and carriers knows how long they soak something in their labs.

Ubuntu has all the pieces in place to be an awesome option. Plus, they have their carrier advisory group with some great names showing their is some carrier interest.

Ubuntu has a company behind it and an owner who seems to want to compete in this market and has very deep pockets. Ubuntu itself has pedigree and history as well as working with hardware manufacturers and chip companies.

I mean, if your only metric of what is possible based on what is current numbers, Android would never had made it out of the gate, or should have folded after the Nexus One or even their next phone (can't remember the name), both of which were TERRIBLE and sold poorly.


Canonical released a phone in October? I think you're mistaken.


Canonical released v1 of their phone OS for developers in October.


Will they magically conjure up an app store full of apps? Unless the plan is to leech off existing Android apps somehow.


Somewhere around 0.0M units with an approximate 0.0% marketshare.


Another reason you can't make new smartphones is lawsuits. I imagine Ubuntu OS will be sued out of existence by patent trolls should Samsung adopt it. They don't have an army of lawyers like Android/Google.


Rather than vocally adopt it, wouldn't the better plan for Samsung to silently allow their phones to play well with something like Ubuntu? The community will try to build up Ubuntu OS to a usable level without Samsung expending significant resources. They have their Plan B without opening themselves to any liability in the short term.


that is my main fear.

Today even the most open android phone have the bootloader and radio firmwares closed because of patents.

no matter how good ubuntu will be, it will end up only being offered ancient radios by qualcomm and the likes. and they won't be able to work around since the standard is patented by dozen different people. You'd probably will only have single band phones in the begging. or maybe i just worry too much...


This should be called "Why You Can't Build a Smartphone Ecosystem"

Building the phone itself is easier, license the software from google, develop some cool hardware, build and sell it.

Admittedly, selling it is probably the hardest part, hard to innovate on features, you can innovate in packaging, price, assembly place, any number of other things.

If you have a big enough bucket of money, you could build a new ecosystem, it wont be easy though. Even without the bucket of money you could build a new ecosystem - but you might only capture 2-5% of the market.


Another far-out option is building a solution not dependent on the carriers, which have already captured the IP space. There is no more internet, it's all just "telecom" now. That's the first mistake people make when starting an "internet" business. Business is reliant on communication and your communication is reliant on telecom. It's pretty much the same landscape as before the internet existed.


The article is kinda paradoxal...

> the iphone changed the game

> no device can enter the market without carrier help.

The iphone was successful exactly because it sold to rich people before, with no carrier help. in fact, in the beginning the carriers would throw anything at you to move you from an iphone. In fact, the reason you have to pay a "smarphone fee" until recently was because AT&T greedily charged a mandatory $30 extra for "iphone data" if you ever wanted a subsided iphone.


iphone started without carrier support, maybe. (There was some sort of an at&t deal, i dunno, doesn't matter) But the point is that was then, we now live in a different world. Apple gave invented the smartphone market as we know it. Nowadays market is mature - it's nearly impossible to disrupt it.


apple is for a long time ridding the frivolous upgrade market. I bet if it wasn't so easy to sell your phone on the grey market (for it to be consumed on the grey market overseas) nobody would be justifying upgrading to the latest number or "s" every year.

That is a fragile market as the market nokia had before. which everyone assumes to be "nearly impossible to disrupt" as you put it.


I think many of the author's points are valid, but I do still believe that companies like Blackberry and Nokia can build Android or Windows Mobile phones successfully, as they won't have to deal with a) user inertia from changing to a new OS or b) developing entirely new UI and operating systems.

As far as I can tell, there won't really be any new carriers or any new operating systems, but that doesn't preclude competition in the hardware space.


I hope Blackberry and Nokia can produce some decent phones in the future. Android has come a long way, but Windows Phone is still pretty awful (more importantly, Nokia are never going to produce an Android with Microsoft as an owner). I used a HTC G1 and Motorola Milestone (1.6 and 2.something) as well as a Galaxy S3 (on 4.something) - back at 1.6 it was nothing special, the S3 on 4.x was hugely improved and a joy to use. If I wasn't so heavily vested in iOS it would be a worthy contender. Windows Phone on the other hand I can't say so much for. Most of the people I've spoken to that are non tech-literate don't have favourable experiences. A family member (electrician by trade) had a high-end Nokia Lumia for a year and didn't have a whole lot to say about it, until he took it in for repair and they gave him an iPhone 4S. He couldn't stop talking about how much better he found it to use, and thus as soon as they demanded it back he switched to an iPhone. Until these manufactures delight the user in the same way that the iPhone does, or differentiate themselves enough, there's little hope.


I believe that smartphones are actually a slice of the embedded systems development market, the same as desktop computers. They are probably the most profitable slice, but that is exactly why the biggest players in the market pay top budget to compete about the shares. Because of that understanding of mine, I think you don't really get into the "smartphone market" as a start-up, but you get into the embedded systems market. And that is quite possible. You just need to find a slice of the market that is not so highly competed in and you can enter there, e.g., producing small Linux computers for public transportation companies with different, very specific tasks, or building remote controls for model airplanes/quadrocopters. There you get all the know-how of building your own computers, working on your own operating system, finding suppliers and when you are really, really good, you might even come up with a platform of some sort that you can get other developers on to make Apps for, just in the business-to-business area instead of the tough endcustomer market.


With a bit of tongue-in-cheek...

1. Data networks

It's an open question of using WiFi as the universal carrier - we're not at that point yet; but it's not a question of using existing GSM networks. It's going to be hard for carriers to deny, after some point, minutes to paying customers.

2. The cost of entry - OS, cloud store, apps

This happened time and again. Remember when Linux has appeared on the stage? Seen porting projects, like Wine, PetrOS, Mono? Realize that Android was built not from scratch, at least concerning interfaces? Second time is almost always cheaper - and in our times, by a wide margin.

3. Access to hardware components - need good hardware...

No. I don't care much if the phone is 5 mm or 2 mm thick, 100 g or 250 g heavy - just like a word processor user doesn't care if letter appears on the screen in 10 ms or 90 ms after he hit the keyboard. At some point it becomes irrelevant, and competition shifts elsewhere. See Firefox phone as another example.

4. Retreat to the low end

Yes, you see it. The fact that the product is a commodity doesn't mean it doesn't exist.

5. A disruptive UI

It was surely a pain to switch from sleeky Nokia UI to strange future of IPhone, no? How about Sugar, from One Laptop Per Child project, which was done anew?

6. Be realistic?

I certainly see a point in what you're saying; but don't you see the power of fashion - ones all important things are by default? As soon as changes become unimportant, returns dwindling and defaults good enough, attention goes to other things - a similar thing happened in automotive world, where a lot of customers buy not the functionality, but trendiness.


femtocells (wifi size cells) are virtually never cost effective, not enough density to support the infrastructure.

a combination Microcells and picocells, however might be doable, you use mesh networking to tie the picocells to to each other, and back to the microcell, ideally its a slow speed low power network, around 900 mhz, providing a access method for text messaging, and push notifications (sub 100kbps, think of it as a faster control channel), Data could be mesh even between handsets, passing (some) data thru to the nearest base. A higher speed network used for more intensive data use situations that goes directly to the nearest micro or picocell, For voice calls or in rural areas, you could default to the regular BTS.


wifi is not designed or has the bandwidth to support the density of devices that the mobile phone network needs


I don't want a "smartphone". I want a handheld computer, the size of a "smartphone", with a cool form factor like a "smartphone". But, it needs to have Ethernet ports, USB ports, SD card slots, CF card slot, etc. An RS232 serial port would be nice too. I want versatility: inputs and outputs. I do not need to connect to any cellular networks. I do not want a closed source baseband operating system, SIM cards, etc. I want an open source bootloader, like U-boot. And I don't want to be forced to run Linux. It has to run BSD, too. So, I don't need to build a "smartphone". I only need to build a handheld computer.


I want a handheld terminal for a belt-mounted computer. A belt can easily support quite a bit more battery, for hard core computation, longer-range transmitting, and GPS.


A belt with integrated batteries throughout and a micro USB cable would be so useful, and the batteries could easily be integrated fashionably. I don't know why anyone doesn't make that. You could charge your phone in your pocket


Ethernet port on its own is thicker than most smartphones. Having said that, I had almost all of those things you mention on my Toshiba Libretto back in 1998. Wouldn't fit in a trouser pocket but would fit in my coat's inside pocket.


Rasberry Pi sounds pretty like what you are describing. Just doesn't come with a display as far as I can tell.


Perhaps we eventually see generic Retina displays with open standards ports.


"I owned a string of PalmOS devices during this period. Their ‘app store’ was literally boxes of software in a store which you had to install from your computer. No different than 1980s PCs."

That's not really true. Mobile devices around that time were designed as peripherals of the PC, so you could download apps to PC and install them by cable-syncing. Palm apps were just .PRC files.

It really wasn't too complicated and was a far sight from buying shrink-wrapped software in a physical store. It's just a shame that Palm, MS, and other platforms left this stuff to third-party websites instead of making it a more seamless experience for users and a more profitable one for them.


I'd say that downloading software on my original iPhone was probably the 2nd most OMG moment of owning the device (the first being the capacitive IPS display that made the icons seem painted-on). Seeing an app download over a 2G cell connection and be usable in 2 minutes - that was amazing. No clickwrap installer, no license keys, no entering payment details. It foretold the doom of non-AppStore software sales in a wonderful zen moment.


There still were shrink-wrapped PDA apps available in stores, though, which is probably how a lot of people acquired them. I never owned a PDA back then, but I'm assuming they were just the same files with a shiny installer.


There were some early (third-party) online app stores too - sites that would handle payment processing and host your .prc files for you (e.g. https://web.archive.org/web/20020124152252/http://www.palmge...).


I understand where the author is coming from but I disagree completely.

Imagine trying to build a smartphone 15 years ago, that would be really hard. You need a decent screen you need a good cpu etc.

All of that is off the shelf now. Anyone can buy it and its cheap. (esp in volume)

In a market in China (I dont remember the name) there are vendors upon vendors who sell all sorts of electrical parts, gadgets etc. You can literally walk down the isles and pick the parts you want in your new smartphone.

So hardware wise, now is a good time. Sure you wont get as polished as the big players right away, but prototypes or a limited set sure.

Now software. There is quite a bit of software that will run on the arm board. With all the hacking on the android platform (sadly a lot more than on the Windows or Apples phones), you can (if you want) at least create a very customized different Android based OS if you want. (would make updates a bitch).

But you can also get some version of linux working and I am sure some other strange os.

But there has never been a time when it was easier to get some dev boards, and start writing the next mobile OS.

Its something you can get started on at home now for less than $1000.

As far as the cloud services and integration that has never been nearly as easy. We have all sorts of providers with stable and "open" APIs.

Providers like MS azure provides a full stack. Database, Data storage, device sync, messaging, computation etc. and they are not the only ones.

I cant think of how much building and setting up all that cloud stuff would have cost in itself. Its out there now, and if you are tesing and writing your app, or have it deployed with low values its very cheap. Sure you wont have your own Music store right away (but licensing is the biggest issue there) but you can piggy back Amazon mp3.

So we have good access to cheap hardware. We have development tools, boards and compilers. We have excellent possibilities cloud sync, integration, etc. We have open source operating systems we can hack at, learn from and understand how things work.

As others have pointed out the future is probably pure data phones. That is a great market to attack right now. Forget the 1984 "phone" part and the nobody really uses anymore SMS.


But that's the thing though, companies like Apple pretty much ignore the whole "phone" and SMS thing, Apple has even shot the operators in the back with iMessage.

You're right that it's easy to make a phone nowadays with off the shelf components, like the article mentions that's exactly what the chinese/asian OEMS are doing now. The question is whether you will be able to make something that can compete with Apple / Google / Samsung. Even HTC is having a hard time doing that with some pretty decent phones.


That is very true.

I do feel that the best phones out there today as far as sales maybe the iPhone and the Samsung S4? Will in 5 (?) 10 (?), be like the first iPod, big clunky and obsolete.

And the huge phone makers of yesteryear: Nokia, Blackberry etc are now small. I can easily see in a 10 years period a a new player entering the market with success. (if they can keep from being bought by the big 3). Facebooks phone partnerships so far have sucked. Maybe they will find something better. Maybe it will be someone else. Or maybe no one will :(

I am still hoping for a new PC os :o


What's wrong with the existing PC OSes? =).

In the free/open space, we already have fifty billion variants - Linux, the various BSDs (FreeBSD, NetBSD, OpenBSD), GNU Hurd, ReactOS, OpenSolaris, Haiku etc.)...

Do you want more? I mean, I get the whole diversity thing, but after a while, I suspect you get diminishing returns...


I didn't actually know that GNU Herd was stable yet so that is actually cool I will install that later on (VM)

We do have a lot of great Unix inspired operating systems and they are doing an amazing job. I count OS/X in that category too, though its debatable.

Then we have Windows, the big giant. I would really like to see them do a Windows NT Squired, NT NT, a new generation that can break compatibility.

Haiku I am a little excited about. I remember how cool the BoBoxes were many years ago when I saw a demo. How far ahead of Windows it was back in those days.

Dont forget about LoseTheOS/Temple. I dont care what people say about the genius who made it. Its a crowning achievement. It is exciting because it isnt, and was never intended to be a be all do all OS for everyone. It has a purpose.

I would like to see more in Plan9.


Following the same logic we'd have something better than windows8 dominating the market by now. It's so easy to bring a new desktop pc or a laptop to market. Where are new disruptive desktop OSes? Where were they when the market was still growing?


I'm the original author. You can probably guess this from the blog, but smartphones don't interest me anymore. I loved my time at Palm, but I'm working on new things now. Stuff that needs good innovation right now:

* Internet of Things, esp. on the usability side.

* 3D printing. making it more useful and accessible.

* Changing how we teach math and science. I can't stress this enough.

* Improving programming. We still do it like it's the 1970s.


Did they not do it on punch cards in the 70's? I think things have improved in the ten years or so I have been in the industry. Java was the hot language in those days, because you could write it once and run it anywhere.


> Carriers request features, change specs, and pick winners.

I'm not sure this is true outside of America, where we have less of a carrier mono/duopoly. Certainly here in the UK - carriers are required to unlock phones if you ask nicely, >6 months after purchase, and most smartphones can be bought unlocked (I've bought my last 3 smartphones this way.)


I think the point of carrier relevance is importance. Somehow Apple still seems to capture much of the volume.

I have to believe there is still room for disruption at the bottom of the market. Someone could devise a very cheap way to just deliver texts. Or a safe simple network for kids, that they would somehow stick with.

Just when firms seem most invincible, they are also most at risk.


Now that it is officially not possible I am sure it will happen :-)

I agree that it would be exceptionally hard to build a smartphone to the same design formula of the iPhone or Android, as an exemplar we have Microsoft who as spent billions on it and has yet to succeed.

However, you might have said the same thing before Apple entered into the game and you would have been right. Had Apple tried to build a Blackberry or Nokia equivalent slab phone they would have had a hard slog. They went back to first principles, imagined a new kind of "phone" and rolled it out.

So the next smartphone isn't going to look like an iPhone or a Galaxy S4, but it might look like Google glass, or like a pebble watch an earphone and a wallet sized thing you carry in your backpack. Or a communicator badge you clip on your chest. But agree it won't be a clone of the iPhone.


I disagree. Apple is in a serious lull innovation wise. Google has won the eyeballs of the majority, and their ad business is not an innovation driver. MS is stuck in a deadly dance trying to isolate developers from themselves and/or direct hardware access with their closed source Apis and CLR.

I think firefoxos has a serious chance of pulling a ... Er... Firefox circa 2002 if the threat does not force the hand of their financial supporters.(google)

Thanks to google, there is a ref platform now, it's mostly a software problem; necessity is the mother of invention. With apple blocking anything remotely menacing off their app store(coinbase? Seriously? Who does that?) and android being a boring spying endpoint, cool hardware like the nexus 5 could get an new skin with a cool cloudphone thingamajigger.


I disagree with the article. I am still waiting for the device which can be a complete replacement for my laptop. That means

* Ability to "dock" to get a keyboard+mouse+monitor Interface.

* True multi-tasking (The music should not stop just because I switched apps)

* A full fledged command line terminal (xterm).

* support for ssh, Perl, python, ruby, C/C++ etc.

This device may not start out as a phone, May be it will be a raspberry type hobby device with a computer on chip running linux which will "grow" to acquire a pluggable phone module, a touch screen and a docking station.

Still waiting for it.


The latter three are covered by Android, and the first was covered by the Motorola Atrix (which bombed) and the upcoming Ubuntu phone.


check out https://wiki.ubuntu.com/Touch/Install

The docking functionality is still in the works, will land in a future release of Ubuntu. Eventually the desktop OS and the touch OS will become the same Ubuntu.


I agree with the basic premise that a typical hardware company is not ever going to have the management style necessary to beat the iPhone or Android. But I don't agree that the current horses in the race are invincible.

Who is to say that cellphone networks will even be relevant in 10 years? What if city-wide, state-wide, country-wide WiFi were to be developed? Such a public works project would blow the market completely open. And in emerging markets, mesh-WiFi might be the only way to get data to everyone.


My bet is a device like google glasses (or contact lens) with a sleek headset that has mind controllable sensors. I don't want to type, or talk to control it. I just think of calling my friend and the phones dials, and I see her image. I am willing to carry the computing unit in my pocket so the glasses can remain light.

I bet that it is 5 years away. If it is like the cellphone, it won't be google that builds it because there is just too much technology that is not yet ready/affordable.


This is an interesting perspective but it also fails to imagine a shift in paradigm. The way I see it and the way the article acknowledges it to some extent, you can't build a new smartphone with the existing paradigm where carriers have control. What happens if the communication infrastructure that they control gets disrupted is anybody's guess. But maybe by then what we call a smartphone isn't called a smartphone anymore...


The one possible exception to the rule outlined in this piece is probably Xiaomi. They have successfully cornered a really meaningful market, specifically middle class Chinese that is responsible for a big push in revenue. I believe their phones currently run on Android, but with a very strong revenue stream and probably government support where they need it, in time I wouldn't be surprised if they round out their own platform.


I've been thinking this for a while, but I have a feeling there is a decently large market (at least enough to sustain a small company and team) for really nice "dumb" phones.

No data; just text, and talk (and in my vision a really nice camera) with the ability to sync/backup contacts etc.


I agree you can't just build a smartphone. Heck, even building a dumb phone from scratch is a monumental task, let alone being successful with it. What was Apple thinking... Oh wait, they redefined the rules of the game. So that's what you do to shut up the naysayers.


Please, especially if you are going to write a blog about design, do NOT use an animating logo that sticks with the page as you scroll. It makes it impossible to read the content without being constantly distracted.


Let's just build a Wine-like linux variant that can run ios apps


Microsoft loves this kind of guy!!




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: