Slashdot Mirror
Why You Can't Build Your Own Smartphone: Patents
jfruh writes "In the mid-00s, more and more people started learning about Android, a Linux-based smartphone OS. Open source advocates in particular thought they could be seeing the mobile equivalent of Linux — something you could download, tinker with, and sell. Today, though, the Android market is dominated by Google and the usual suspects in the handset business. The reason nobody's been able to launch an Android empire from the garage is fairly straightforward: the average smartphone is covered by over 250,000 patents."
couldn't give a single f*ck.
Really?
SJW: Someone who has run out of real oppression, and has to fake it.
After all: "We are actually witnessing fewer patent suits per patent issued today than the historical average" http://yro.slashdot.org/story/12/11/09/2156232/patent-system-not-broken-argues-ibms-chief-patent-counsel
The reason nobody's been able to launch an Android empire from the garage is fairly straightforward: the average smartphone is covered by over 250,000 patents."
And it's hard. And it costs a lot of money. And the market is full of very good competitors. Otherwise there's nothing stopping you.
Patents won't touch you if you make 1-10 units.
Other manufacturers won't consider you as worthwhile to legislate against since you most likely won't make any profit from those devices sold.
From US point of view, good luck getting your device FCC approved, that'll be cheap and fun process!
There are no atheists when recovering from tape backup.
How many patents cover assembling a PC from parts, installing your own OS on it, and selling those? Anybody have a list or something? Genuinely curious.
----
On a separate, possibly unrelated note, yet another of my silly dreams: right to access and tinker with the BIOS (or whatever the technical term is).
I understand you might part with that right if you have the item subsidised, but after the contract ends, the root rights must flow back to you.
Oh, and root rights should always be available on an unsubsidised device; the whole "warranty will be void" shtick doesn't fly with me. I mean on my PC, it would be void if I over-clocked it or whatever, not because I installed Linux instead of windows; similarly, warranty should not be voided simply because you change a ROM, only if you use an over-clocking app.
(Do correct me if I am wrong in my rant)
I am an ACCA student. Got a query on Accountancy/Finance? Maybe I can help!
In the mid-00s, more and more people started learning about Android...
Android was announced in 2007 and the first Android phone wasn't sold until late 2008. Even the Neo1970 was from 2007/08, so I don't know what the submitter is referring to.
I can't build a phone because of parents. They won't let me solder in the house.
This is totally ignoring the software and patent problems.
To elaborate on why open-source hardware is hard, and why making a single phone will cost you over $10K.
Why open-source software works is:
Widely available repository of code.
Many people able to review it, or sections of it, and understand it.
Ease of submitting tested patches.
Hardware has problems that don't really fit well with this.
The open schematic is the trivially easy part, and not really a problem.
(though in practice, you need a schematic with copious links to design documents, which isn't well solved by open tools).
The number of people who can review it is rather smaller - as you can't open up a c file, and see a clear error or awkwardness in code that can be edited.
For all but the most basic errors, you are going to have to sit down and read several hundred pages of hardware documentation about how the chips in question work, in addition to having in-depth knowledge about the circuit design, and costings of likely changes.
Now, you've done this, and generated a patch that you think (for example) lowers the supply current by 1%.
Compile - test.
On a PC, this takes a couple of minutes.
For something of a smartphone class, a one-off PCB may cost several hundred dollars. Then the parts will cost another several hundred dollars in small quantities, as well as being difficult to obtain.
Now, you have to solder the parts onto the board, which is a decidedly nontrivial thing - and if you decide you want someone else to do this, it's probably another several hundred dollars.
So, you're at the thick end of a thousand dollars for a 'compile'.
Now, you boot the device, and it exhibits random hangs.
Neglecting the fact that you are going to need several hundred to several thousand dollars of test equipment, you now have to find
the bug.
Is it:
A) The fact that unlabled 0.5*1mm component C38 is in fact 20% over the designed value, as the assembly company put the wrong one in.
B) C38 has a tiny bridge of solder underneath it that is making intermittent contact.
C) The chipmaker for the main chip hasn't noticed that their chip doesn't quite do what they say it will do, and the datasheet is wrong.
D) You missed a tangential reference on page 384 of the datasheet to proper setup of the RAM chip, and it is pure coincidence that all models up till now have booted.
E) Because you're ordering small quantities, you had to resort to getting the chips from a distributor who diddn't watch their supply chain really carefully, and your main chip has in fact been desoldered from a broken cellphone.
F) Though the design of the circuit is correct, and the board you made matches that design, and all the parts are correct and work properly, the inherent undesired elements introduced by real life physics means it doesn't work.
G) A completely random failure of a part that could occur with even the best design, and best manufacture.
G - may mean that it's worthwhile making two or more of each revision - which of course boosts costs.
Hardware is nasty.
This gets a lot less painful of course for lower end hardware. For very limited circuits, which can be done on simple inexpensive PCBs, and be easily soldered at home - costs of a 'compile' can be in the tens of dollars, or even lower.
Patents won't stop you from building a couple of devices in your garage; but, they'll be as useful as bricks. You have to get the radios FCC certified and then run the gauntlet of certification hoops to convince the cell provider to allow you to connect your garage built device to their network. There are radio modules available that would speed up the process -- basically pre-certified modules that handle the entire cell phone function. You might be able to do it using these... But they're huge, relatively speaking. You won't be building a sexy device like a Galaxy S, iPhone, or Droid with them.
We've done it on equipment we're designing for deployment; but, I have the advantage of being able to call Verizon and say, "I'm Confused, an electrical and software engineer with Big-Company. I am using a cell radio module from A_well_known_manufacturer. I need to activate it on our account for testing..." And, by the way, we won't do that until we're pretty damn sure the thing will work right.
I dare you to name any kind of gadget containing a computer that you can build without infringing on patents.
In the last line of the summary, replace patents with germs and then imagine what a phone might look like with over 250,000 covering it.
As an engineer, I thought I would point out there are two ways we deal with patents:
Method 1: Once you have an idea, do a thorough patent search and verify your idea does not appear to violate any patents. If it does, re-design the widget so it avoids the patent.
Method 2: Ignorance is bliss. Design and build it.
I can tell you, if you use method 1 you will need an enormous staff and risk never getting anything done. Despite it all, you still won't be safe because someone will come along with patent claims anyway, even though you did a most thorough due diligence search. I'm not saying you ignore patents, that would be unethical. Company I work for has a record of the patents related to our products that we have been made aware of. It just doesn't make a lot of sense to go looking for trouble.
GeeksPhone are doing pretty much what TFA claims is impossible. Why haven't they been sued? Too small to be worth the trouble? Jolla (50 employees) aren't exactly a behemoth either. OK, so Jolla haven't released anything yet and thus can't be sued, but the fact that the company was formed implies that they don't consider the 250,000 patents a problem. (Yeah, I know, not Android, but the same principles apply.)
It would seem that patents are company-centric when they really should be consumer-centric. If we change to the consumer view paradigm, a lot of the idiocy of patents would disappear. There really should be just one question for a patent to be viable: How does your exclusive retention of this concept and the implementation of it benefit the consumer? What we have is the exact opposite. If we theoretically have laws and government to aid and support people, then the entire patent system and paradigm is 180 degrees off.
Most sensible countries still don't allow software patents.
I'm pretty sure it's my lack of electrical engineering and general expertise on the topic. Honestly, anything I could cobble together would be some seriously shitty product.
while the overall situation is deplorable, at least it gives someone in their garage extra motivation to ponder what could be the next form factor after cellphones, and get a head start building that.
companies have heavily invested mindshare, factories, research, etc. into their cellular products. the chances of them developing the "next big thing" are slim. and cellphones wont be around forever.
Just make a device that has a software-defined radio in it capable of transmitting and receiving on what ever band you need, and release it without any software on it. Since it doesn't work as a phone, it's not violating any patents. Then have an unrelated group (i.e. the open source community) spring up and release unofficial firmware that turns the SDR device into a fully functional open source phone.
They can't sue you for making the hardware if it's not actually a phone, and some people beyond your control are hacking your device to turn it into one...
The patent system isn't helping.
But even without the patent system, I kind of doubt you can obtain advanced microchip fabrication technology and use it in your garage.
OpenMoko (http://wiki.openmoko.org/) tried to produce fully free phones, as in free hardware plus free software, and only sold ~10,000 units.
Open Phoenux (http://projects.goldelico.com/p/gta04-main/), one of its successors, definitely should get more attention
Do patents prevent me from building my phone? Or are they intended at preventing from building and selling it?
I believe I can build it if I don't sell it, even with patents.
I'm in doubt if I could sell the blueprints and hardware and software details, provided I don't sell the product itself. Can anyone enlighten me on the subject?
I also think we should be thinking of ways of freeing Android from M$ patents (though not necessarily from all patents); M$ is starting to look like Schwarzenegger's Terminator 1: when you think you killed it, some part of it still aims to get you... (but, yes, I digress).
Oh, come on. The "patent system is screwed up" argument started with software and algorithms, and continued with patents on DNA and organisms. And yeah, in those cases it was pretty clear the patent system wasn't working as intended. I'm with you on that one. But now here at Slashdot we're upset that ingenious physical devices, devices that took years of work to design and whose operation is by no means obvious, should not be patented.
There can be no question that something like a SAW filter is a new, non-obvious, useful device. So at this point, you're arguing that patents should not exist at all. And I won't follow you there.
You really have no idea how much brainpower went into designing the individual components within a cell phone. An iPhone is a miracle the first time you see one, then it's a handy tool, and then your familiarity makes it seem blindingly obvious. But tens of thousands of people spent billions of person-hours figuring out how to build the thousands of unique devices inside. And I think those people should receive some reward for their efforts.
Someone explain again how this is good for innovation???
That Diversity is a necessity for innovation ?? When Steve J tinkered with small computers, there were "much better computers which do any processing job, made by IBM Corporation" around. Silly idea to even THINK about these little, powerless computers, right ?
..if you are an anon software developer. Do we need Fame And Fortune, or do we need Open Source ? Release code via TOR and sign with a pseudonymous GPG key.
Disrespect all these trivial patents, develop code and release pseudonymously via TOR or GNUnet. Run your SVN or git server as a Hidden TOR Service. Fuck the corporate sleazebags who abuse the patent system. They have lost their legitimacy a long time ago,so just ignore crap.
I own one.
Its touchscreen broke after just a couple of months, and when I asked the french reseller if this'd be covered, I was almost insulted (asking this apparently meant I didn't have the FAITH).
But, yes, I have the actual experience of a fully open-source phone (and GPS).
That OpenMoko then got bankrupt means one thing at least: none of you here, OP included, does indeed consider phones should be open-source ;-)
why would anyone want to stifle the public's innovation?
I get it - you came up with it first... congrats. So instead of me doing it just a slightly different way which falls under your patent (even though I had no knowledge of what you were doing), you want to play monopoly.
And we wonder why your average joe isn't pushing out tons of patents - major companies are. And if mister average joe is a Tesla or Franklin, in these days, they would be put away in a secret compound owned by major company.
First, if we try to replicate high-end commercial devices, we are probably bound to fail. As others pointed out, high-end semiconductors and other components are hard to acquire, if you are not a Big Corporate SleazeCorp. But it is indeed possible to buy GPRS chips in small numbers. Read amateur radio maganzines about that. Then build you homemade mobile device, running an OS of your choice. Or take a tablet, run your homemade OS and attach a USB UMTS modem, if you cannot get the internal radio working. Yeah, not sleek, BUT FREE ! Free as in free speech, of course.
Secondly, Go Clandestine. Government and corporations are running over so many long-established and cherished rights so that we, the people now have the moral right to GIVE A FUCK ABOUT PATENTS and ALL THEIR SURVEILLANCE SHIT. Of course, the fatboys at FBI don't want to move out of their fat chairs to perform surveillance; they want to write National Security Letters to sniff into your personal papers and affairs at Google Docs and Facebook, etc. If you are politically active and piss off Mr O'Bomba, your can be sure they will find a pretext to sniff into your Commercialware IT products or "cloudy" services. No criminal action required whatsoever.
Clandestine operation means we de-compile code, release it via TOR and USENET. We write programs irrespective of corporate sleazebags having patented the single-linked list and the rounded corner. Of course. that means staying pseudonymous, signing with GPG and releasing via TOR, GNUNet, etc.
Use TOR as A Matter Of Principle. Rich and powerful corporations have the power to intercept your DSL line and they will do that if they think you "violate" their "precious" patents. Fuck the law; it does not apply to Cisco or Apple as it applies to you. Don't give away any hints by googleing or traceable chat messages. TOR ALWAYS.
Expect the sleazebags to send their reconnaissance assets into hacker meetings. Expect your home to be raided at the behest of Apple, Google, MS or any other of these slimecorps. The government is in collusion with them to illegally sniff into your data, so the government will bend over to "help them protect their patents and IP". So, encrypt your full disk using TrueCrypt and just pull the power plug when the SWAT team comes. And believe me, they have the power send the SWAT team for "patent infringement".
To protect your code, put it on Flash memory chips, drop that into an empty plastic bottle (Coke plastic bottles as sold in Europe work excellently, firmly close the screw, put it into an additional sealed plastic bag and then bury in the woods. So when they have taken away your hard disks on some pretext, you can recover your last backup from the woods. After all the "you are a software terrorist" bullshit has been eliminated by your lawyer, of course.
Yeah, all cloak-and-dagger stuff, but I have absolutely no moral problems with that, as the government routinely breaks their own laws, acts based on flimsy pretexts and of course unfairly aids the big fat corporations in their quest to eradicate the little guy who "violates" their trivial crap or their attempts to corner markets.
The idea that patents are what stops small competitors from producing smart phones is just not true. To make a commercial smart phone, you have to be able to do it at a cost and quality level that is similar to what big electronics manufacturers can do. There are many reasons why a small shop can't do that.
You would have to hire several electronics engineers and mechanical engineers who are as skilled at packing a lot of functions into a small package as those employed by Apple, Motorola, Samsung and HTC.
You would have to employ them for at minimum 1-2 years before you sold your first phone.
You would need your own production facility or invest enough to have a CM make a run of your phones.
You would need a sales and marketing staff capable of getting consumers' attention and trust.
You only need to start thinking about patents when you are confident you can solve all the above problems. Only companies with a lot of money to invest ever get that far. A number of companies have broken into the Android phone market. What they all had in common was a lot of money to invest in starting up a smart phone business.
Of those, I think the last might be the hardest. A smart phone is going to be an expensive item. Such items are sold on trust. The consumer has to trust that they're going to work and that they won't suffer buyer's remorse when your phone turns out to be a piece of junk and they could have had a brand-name phone that works.
according to patents, it is against the law to sell anything that someone has a patent on... however, you could if you spent the time, create a do-it-yourself instruction set on how to build a smart phone. this, according to patent law, is frowned upon, but is unpunishable. you can build whatever you want for yourself (or for giving away for free), even if it is patented
The notion that hardware is "more difficult than software" is just ridiculous. If you have no experience in hw design and debugging, only then it is difficult. But that also applies to software. Just talk to a guy ordered to "make that single-threaded package multithreaded". He will tell you why hardware is so much more easy than sw development.
But yeah, you also need the right tools and knowledge. Oscilloscope, multimeter, often some sort of logic analyzer. Most of all, you need serious EE knowledge to do EE. As much as you need serious CS knowledge to develop complex software. And don't come with that "self-trained developers are better" myth. Simply not true. If you never had any electrical engineering training, you better go to the next uni to get some basic courses on components, complex math, fourier transform, filters, transmission lines, transistors and all that. Have an experienced guy showing you how to perform some non-trivial oscilloscope measurements. Discover the capacitor value by means of a self-built osciallator and the scope. Build a Z80 based computer. Then FPGAs. Then an illegal AM radio. Then some day, build a phone prototype. It will be as big as a PC, but that''s also how the commercial guys do it. Only when the big prototype works, they will miniaturize it. So will you, the amateur, do it. Use FPGAs liberally, as this is the only way you can miniaturize large logic circuits. These days big FPGAs can host non-trivial CPUs.
Stop thinking about how Samsung would do it. They can build ASICs, you have to use standard components, including FPGAs.
I once thought about creating a matlab implementation for a specific domain and use case. After looking at the patents which Mathworks holds, i stopped that. Many of these are for sure trivial, but if you would place somthing which they consider competition, you will not survive the lawsuit.
Is there a patent for 2 tin cans & some string already?
[Sorry, I'm too challenged to do my own patent search!]
I'm not saying you ignore patents, that would be unethical
Illegal, yes, but unethical? Excuse me for being blunt, but that's asinine. Patent law is founded on coercion (quite unlike laws intended to protect against natural human rights, which preceded government and in fact provided government with its first justification). In the abcense of coercive authority, patent law would not exist. It simply would not be possible in a truly free market scenario.
Now tell me who's being unethical.
Is that trying to source the components, and then solder them together, is just not possible for the vast majority of open source and free software enthusiasts.
Have you ever opened up a smartphone? A handful of relatively specialist chips, all with connectors to the motherboard that would require a soldering iron held by fingers the size of human hairs.
It's just too hard.
You are not alone. This is not normal. None of this is normal.
This raises an important question: How does one pronounce "mid-00s"?
Over 250,000 patents? That's probably more security than a medieval fortress with a thick, heavy steel fence and barbed wire, a surrounding moat and drawbridge, snipers, and dozens of guards operating turret guns and cannons. I'm sorry, but that's just uncalled for for something as simple as a god damn phone. If that's not stifling the fuck out of innovation, then I don't know what is. How the hell has this come to be without the government seeming to even give a fuck, even with the heavy lobbying of companies with their own interests? Corruption at its finest.
Uncle Sam could build it as long as it is for "fair use".
literally.
What would stop a DIY'er making and selling a device that:
Ditched capacitative screens. Used keyboard and joysticks and resistive touch screen.
Used casing designs that are friendly for home-brew fabrication devices such as makerbot.
Built around an open platform, like Rasberry PI.
Use standard rechargeable batteries.
I bet they will send some very nasty DMCA lawsuit into your inbox. 50k lawyer costs later you know why your should have used TOR an have been anonymous.
http://www.reichelt.de/LCD-Module-Touch-Grafik/TFT-DIS-7-0-MT/3//index.html?ACTION=3&GROUPID=3011&ARTICLE=101753&SHOW=1&START=0&OFFSET=500&&SID=12UKAx0H8AAAIAAC92Reoe51de364d8941e1e6a12685f4f976c93&LANGUAGE=EN
http://www.reichelt.de/LCD-Modules-touch-graphic/TFT-DIS-3-5/3/index.html?;ACTION=3;LA=2;ARTICLE=101746;GROUPID=3011;artnr=TFT-DIS+3%2C5;SID=12UKAx0H8AAAIAAC92Reoe51de364d8941e1e6a12685f4f976c93
As you wrote, standard AA or AAA batteries already have excellent capacity (I know because I operate the toys of my daughter) and you could easily have more reserve batteries in you pocket for contingencies.
http://www.kh-gps.de/tb2.htm