Submitted by Microsoft is generating $2 billion per year in revenue from Android patent royalties, says Nomura analyst Rick Sherlund in a new note on the company.
He estimates that the Android revenue has a 95% margin, so it’s pretty much all profit.
This money, says Sherlund, helps Microsoft hide the fact that its mobile and Xbox groups are burning serious cash.
Microsoft has not written a single line of Android code, yet rakes in the profits through scummy software patents. Crime does pay.
Indeed, and Judas wasn’t even the first.
Kochise
Funny how it IS NOT ILLEGAL but you and Thom label it as “crime” simply because you don’t like the system. Don’t hate the player… Google plays the same game. Just not as well:
http://www.fosspatents.com/2011/10/google-scores-own-goal-google-fu…
Edited 2013-11-07 20:06 UTC
Well, we’re not speaking about Google, currently is how much Microsoft enjoy its strategic move without people gaining anything. I know Microsoft aren’t philanthropic people, but the way they enforce their position is sad.
Kochise
Agreed… calling this kind of thing a crime misses the point – it’s *not* a crime, it’s just the inevitable (and totally legal) outcome of a badly broken patent system.
Citing fosspatents? Really? Oh man, we’re in trouble.
While I am definitely not saying the following is a like-for-like example, it’s useful to note that the genocide of Jews in Nazi Germany was technically legal at the time.
Yet the perpetrators eventually found themselves facing war-crimes charges for their actions.
I use the example only to demonstrate that simply because something is legal, does not mean it isn’t a crime.
I would argue that the stand-over tactics being used by all the companies concerned could be considered criminal, only nobody has challenged it in court yet.
It’s just going to take someone growing a set and digging their heels in. Only problem is that all the companies with the means to do that are contributing to the problem.
Not true. Jews were deported to other countries such as Poland to be executed. No death camps existed in Germany. The systematic execution of Jews and other persecuted minorities occurred in relative secrecy. It was never mentioned directly in official documents and was never codified into law.
And by that time for example Poland was ‘part of’ germany. Relative secrecy? Come on, NSA actions are relative secret and yet legal. Laws are changing all the time.
There were plenty of concentration camps in Germany; Buchenwald, Dachau, Neuengamme, Ravensbrück, and Sachsenhausen were among the worst camps in the Holocaust. All of them located in Germany. Germans knew pretty well what was happening, even though lots of them tried to play coy after the war ended.
I have no idea what the Holocaust has to do with patent extortion though. I guess this thread got Godwined in record time.
Edited 2013-11-10 22:14 UTC
Microsoft spends billions in R&D, they are among the biggest spenders worldwide – all type of activities included.
If their patents and R&D work is valid they should be paid for it, it’s just natural.
I’ve said it before, MS must be receiving $2 to $3 for each android being sold, way below some reports that go to the insanity value of over $10!!
While $2B sounds like a lot of money, $2 or $3 for each device don’t seem to be braking the bank.
So you spend a lot of time on making the perfect burger for your restaurant. You patent it. I live on the other side of the planet and have never heard of your recipe or restaurant. I make a burger that I like for my restaurant and it is very similar to yours. Now I have to pay you 50 cent per burger I sell.
Am I understanding you correctly? If not, please tell me.
In reply to:
“So you spend a lot of time on making the perfect burger for your restaurant. You patent it. I live on the other side of the planet and have never heard of your recipe or restaurant. I make a burger that I like for my restaurant and it is very similar to yours. Now I have to pay you 50 cent per burger I sell.
”
Do you think that would be a valid patent?
For me it seems an obvious invention.
Edited 2013-11-07 12:40 UTC
Yeah, because Google never heard of Microsoft and their patents when implementing Android.
There is for sure a lot to dislike about the current patent system, but it is amazing how much BS we can see coming from techies on forums on this subject.
But then again, Android was already implemented before Google acquired it.
Yes, but not fully (Google did modify it a lot I guess) and not that long before Google acquired it and probably after Microsoft did register their patents.
Is your point that Google should be free to use the patented technologies because it was Android (the company) which implemented the OS before the acquisition? In this case, your point is obviously not valid, because when one company acquire another company, they take the full responsibility for everything the acquired company did in the past.
That is because techies know how trivial those patents are. Non techies think it is really difficult to ‘invent’ bounce back scrolling.
…and/or people with an agenda.
I don’t know how techies who don’t even know what patents we are talking about (in this case as nobody actually know hat patents are involved in the MS deal) could decide whether the patent is valid.
Moreover, in order to be valid, a patent do not have to be difficult to invent. The velcro was not difficult to invent: it was just a very smart idea.
Velcro is a physical invention. It started out as an idea. But anybody can have an idea. The patent was for actually making something.
Imagine that years before the invention of Velcro somebody had the idea of Velcro and patented it. Somebody else really made it but now has to pay Velcro because he patented the idea.
Idea patents are stupid and bad for everybody but big established corporations. And they really don’t need any extra help to make billions.
I agree, but you are talking about patent trolls here and neither Microsoft nor Apple are patent trolls because they do actually use their patents in products.
But Google is presumably using different code (otherwise this is a copyright issue). So the points stands – these are patents on generic idea rather than specific as the code is the specific implementation, and copying the code would be a copyright rather than patent violation.In other words, software patents are dumb.
The issue is stopping innovation because you were the first to get a monopoly. You were not special. You were not genius. You brought nothing in the world. You simply had a bag of money to patent an idea millions of people had before you. Your patent stops progress. Your patent is a deterrent to human progress.
I hope my point comes across this time.
Nope, read back the story of the velcro, the guy ‘invented’ nothing, he just copied nature :
http://en.wikipedia.org/wiki/Velcro#History
People copying nature, patenting the genome, shouldn’t be allowed to patent anything, since there is obvious prior art and anybody could have made it too.
The manufacturing process of making velcro might yet be patented.
Kochise
Way to miss the point.
majipoor,
“I don’t know how techies who don’t even know what patents we are talking about (in this case as nobody actually know hat patents are involved in the MS deal) could decide whether the patent is valid.”
You’ve highlighted one prime reason enforcing software patents is absurd: merely identifying patent infringement is an intractable problem. There are hundreds / thousands of new patents being issued on any given day, with a backlog of millions – a large proportion of those have been issued for software. The truth of the matter is that it’s impossible for individuals to keep up, much less identify which patents we’ve inadvertently infringed upon in our software.
Microsoft is merely exploiting the inherent problem of identifying patents. There was a popular example back in the day: IBM’s lawyers paid Sun Microsystems a visit with demands for patent license fees over 7 patents. Sun indicated that they could fight those patents. IBM rebutted “Maybe you don’t infringe these seven patents. But we have 10,000 U.S. patents. Do you really want us to go back to Armonk and find seven patents you do infringe?”
http://www.washingtonpost.com/blogs/the-switch/wp/2013/09/24/large-…
Writing software is in and of itself an inventive process. The problem with respect to patents is that in the course of doing our everyday jobs of solving software problems, we will infringe patents without even knowing it. We worked hard and solved a problem. So when someone sues us over infringement, our own rights to our own work is in jeopardy. Our final product might even be better than the patent holder’s, but none of that matters since the patent system granted them a legal monopoly over the software methods we are using.
“Moreover, in order to be valid, a patent do not have to be difficult to invent. The velcro was not difficult to invent: it was just a very smart idea.”
That’s another problem with software methods, it’s all too easy to solve a problem the same way as someone else already has. In nature “ideas” aren’t a limited resource, the more you have, the better. Under the contrived patent system however, ideas have become a land grab, usually sold to the biggest corps with the intention of charging license fees over everyone else’s use of ideas. The vast majority of software developers feel the way I do, software patents are very bad for our field and have had the effect of rewarding the use of court system to block competitors rather than spurring innovation.
I mostly agree with all your points but I don’t see why they should apply only to software patents and not “hardware” patents.
The Velcro is a good example, because it is a hardware patent which is just a smart idea that mimics nature. On the other side, software patents may be actually very clever and complex algorithms requiring a lot of research (for example compression algorithms or audio/video codecs).
So my point is not that all software patents should be granted, but that some are actually more patentable than many obvious hardware patents.
The “software patents are Evil” thing is thus mostly a dogmatic point-of-view for me from people considering that “software are just maths” which is not more correct than saying hardware is just physics.
I am myself a software engineer, designer and solution architect and I know that engineering a clever piece of software cannot be summarized as a mathematical equation.
majipoor,
“I mostly agree with all your points but I don’t see why they should apply only to software patents and not ‘hardware’ patents.”
I focus on software because that’s my domain. I don’t know if the economics behind patents makes more sense on physical inventions, but I think it would be silly of me as a software developer to tell a chemist (for example) what is best for their industry. I’d be in favor of letting each industry choose what’s best for itself.
Any yet it doesn’t seem to have helped Windows Phone gain significant market share
Doesn’t matter that [you] live on the other side of the planet. [We] both sell burgers where [you] live and where *I* live. So, yes [you] would have to pay 50c per burger. Not [my] fault that [you] didn’t perform due diligence of what patents were in place before selling that burger.
Of course, you could just not pay or not make that particular burger and sell a different one, and then finally patent it to protect yourself so that this doesn’t happen again.
Yes, they spend billions finding trivial ideas and patenting them, after they spent billions in bribe (ahem, lobby) money putting this fucked up system in place. It’s only fair that they earn it back.
O course they do, by definition most of the money spent by a software company goes into R&D. And nobody contests they are one of the biggest in the world.
Old debate over software patents and copyright. Is not that you are totally wrong, bu right now most patents are not used by others because they looked at them and saw: oh, what a nice idea, I would invest X$ if it wasn’t for for this patent that I can buy.
No, usually it is like this:
– a company creates something and runs into a problem, it finds a simple and beautiful solution,
– then another company comes and say: you know, we had a similar problem, found a similar solution, made it general, wrote a patent, and wait to get money for this.
Compare to what ARM gives ad how much money it asks, this is a lot.
It is hard to know the exact sum, but it is less than 5% of the CPU cost according to:
http://www.anandtech.com/show/7112/the-arm-diaries-part-1-how-arms-…
And that is for a complete CPU R&D
“then another company comes” .. “we had” : you are mixing present and past tense which is quite clever but also (willingly?) confusing.
The patent holder must first find the simple and beautiful solution and patent it, otherwise the parent is invalid due to prior art.
That being said, I still don’t understand the fundamental difference between hardware and software patents which would makes hardware patents acceptable and not software patents. Can anybody quickly explain me?
You cannot patent math. 1+1=2 cannot be patented. Software is math. Ergo, software cannot be patented.
Oh god. This again.
There’s math in software, but software is not math.
“A program is the transcription of an algorithm in a programming language, and being every (Turing-complete) programming language equivalent to Church’s lambda calculus by virtue of the Church-Turing thesis, a program is thus the transcription of a mathematical function. Since math is not patentable, neither is software.”
http://en.wikipedia.org/wiki/Software_patent_debate#Arguments_again…
Science > Nelson.
Right. Software is not math, that’s a reduction to absurdity. Quoting one side of an argument has no impact on what I said given that I disagree with its premise.
Its nice that you can have others do your actual thinking for you, but when you get some real world engineering experience perhaps your perspective will change.
I highly doubt it. Let’s ask actual programmers – you know, those who do “real world engineering” – how they feel about software patents, shall we?
http://www.nzherald.co.nz/business/news/article.cfm?c_id=3&objectid…
94% were against software patents. 94%.
I’m sorry Nelson, but I prefer to listen to those doing the actual work and those who have the actual experience working in this field – instead of basing my position on patents on company allegiance, stock ownership, or misinformation.
Edited 2013-11-07 15:08 UTC
The says nothing about software being math, only the first source you could find apparently from NZ.
How about this: you are free to hate software patents as much as you like. You are free to lobby for their abolition as much as you like. And do you know what? I will be on your side (though perhaps not sharing your rhetoric).
However in the US software and process patents are legal, until such time as the Supreme Court rules on the matter again, a constitutional amendment is passed, or a law that will pass the test set by the court. You don’t have to like it, but they are legal.
So software parents might be immoral, but they aren’t unethical, and they are definitely not criminal… in the US.
So I ask you again: Separate news from opinion. You want to rail against patens, more power to you. But do it in a separate piece.
I would absolutely call software patents as they are used today both immoral AND unethical. It’s just not illegal. And yes, this whole software patent crap stuff is just a symptom of bigger issues.
One of my philosophies of government is that policies should not discourage people and entities from doing what is right. In the case of the patent system regarding software and methods that policy currently incentivizes players to do what is immoral and unethical to crush any possible competition.
And yes, I am owner and developer for one of those software companies that uses applied physics, dynamic systems modelling and image processing and to solve problems. And yes, our technology does indirectly compete with some of these big players, including the “big three”. We’re still too small to get any bad attention.
The problem with this kind of thinking is: who is going to decide what is moral or immoral, ethical or unethical? These are concepts that keep changing with time/place and that is precisely the reason we have laws Except for the very basic things, and frequently even with them, people keep disagreeing.
I, for one, would like to keep software only under copyright but with all power and money on disputes over it I, regrettably, don’t see this coming.
Also, algorithms are math but unluckily software is not just it, as someone already rightly pointed, it is akin to engineering what means, it uses math + rational thinking to devise a good enough solution for a need and/or problem.
Of course it is.
If you were to ask software patent proponents like Nelson if they were to support the idea of, say, John Grisham patenting the legal thriller, and demanding a 2% cut of every non-Grisham authored legal thriller, they would call said idea ridiculous.
Yet, they still support software patents. With such hypocrisy, idiocy comes natural.
Thom Holwerda,
“If you were to ask software patent proponents like Nelson if they were to support the idea of, say, John Grisham patenting the legal thriller, and demanding a 2% cut of every non-Grisham authored legal thriller, they would call said idea ridiculous.”
They’re both equally ridiculous, however I think the average person has a better understanding of the literary process than the software process. Software development is voodoo to them. They might assume that the knowledge embodied in the USPTO is actually useful for typical developers who are looking for answers to software problems. Little do they realize how much of a waste of time we consider the patent system to be on technical grounds.
Please tell me what ethical guidelines were violated? Ethics as we generally understand them pertain to the correct behavior within the law or other strictures. Something can be ethical but immoral; which is what I am arguing.
No. No. No. No. No.
Seriously. No.
Ethics has NOTHING to do with the law. Absolutely nothing. Ethics pertains to your innate sense of right and wrong, regardless of laws, regulations, religion, or related matters. You may personally find it ethical to follow the law – but that does not mean you can conflate law with ethics. The same applies to morality, by the way.
Conflating – or worse, equating – ethics with the law is very dangerous, because it leads to people blindly following the law without thinking for themselves whether a law is just or unjust. There are people with agendas – politicians, religious leaders – who would very much like the common folk equating ethics, morality, and law – but you should never get sucked into that.
Not ever.
In general, you must never forget that laws exist to serve people; not the other way around.
Edited 2013-11-07 23:27 UTC
Thom Holwerda,
+1 if I were allowed. We should try to make all laws ethical, however the law should not be substituted for ethics. Sometimes the letter of the law may be contrary to what is right.
We were coming home from a bar very late at night/morning, the car’s owner was inebriated and I was the designated driver. A police officer pulled us over and gave me a ~$120 ticket due to one of the lights being out, even though I wasn’t the owner, I was legally responsible for the car’s maintenance while I was driving it. The police was technically following the letter of the law, but the lesson it taught me in retrospect could not possibly be any worse: In order to avoid a ticket, I would have to give serious consideration to letting the owner drive even if he weren’t completely sober. The police officer either didn’t think or didn’t care about the repercussions of fining myself for having unequivocally done the right thing.
Actually ethics is usually accepted to be related to the group behavior or judging of what is appropriate and moral is more usually applied to individual judgment of what is wrong or right and both may not be related to legality.
They are against software patents, but that’s a separate point from “software is math”…
Typically software algorithms research is just looking at things differently, combining dissimilar concepts that hadn’t been tried before, etc. Typically when discovered they tend to be forehead slappers because they are so obvious in context.
So yes there is some innovation there but in no way would I consider those to be “intellectual property”. More like “trade secrets”.
I have realworld engineering experience and I work with other people who do to. We all agree, software is math.
Patenting software is an expansion to absurdity… I already have copyright to keep you from stealing my code. I already have trade secrets to keep you from stealing my ideas before fruition. I already have my skill to apply to improving my product to maintain competitive advantage. I have all this without patents, and so does everyone else…
What patents do give me is a license to sue other people who independently come up with the similar ideas. I define those people as competitors, not criminals.
Engineering != Science. We are talking about software that has come out of a software engineering process.
Keep in mind that a CPU just basically performs on binary data with just AND and OR in its toolbox. AND and OR are really basic logical operations, yet combined smartly could carry out really complex tasks.
But at the very bottom, there is only AND and OR, and both of them cannot be patented.
Live with it.
Kochise
Yes I understand this. However I do not think it is useful.
I’ve been thinking about it.
I’d say software isn’t math, but software is based on or made possible by math.
It’s bit like saying copyright of books doesn’t exist, because it’s made of letters and you can’t copyright those. Or you can’t patent a warp engine, because it’s based on physics and you can’t patent the laws of nature.
The math making software possible is on a lower level also made possible by physics.
People who are crap at math can still code stuff and being briljant at maths doesn’t make you automatically a programmer, nor does it guaranty you can learn to be a good one.
OK, I see the point even if I indeed do not agree because it is not the software which is patented here but the feature and the way it works.
When patenting pinch-to-zoom, Apple did not patent a formula or a piece of code, but the feature and some details on how to track touch inputs etc.
So hardware patents are the same for me: why would you be able to patent a the technology used to build a chip when it is only a matter of physics and how electrons are moving in copper wires?
because the technique and method of growing the transistor requires special machinery to manufacture the chip. That stuff is patentable. How the circuitry is organized on the chip? That seems to be more a copyright issue than anything.
Some of the non manufacturing related chip patents are also method patents as well and should be tossed in the same bin as software patents.
Sorry Thom, software is not maths. Most of these large projects are Engineering … thus the term software engineering.
Yes some of the theory is based on mathematical principles, so is other forms of Engineering but they aren’t maths, they are the application thereof.
A lot of the effort in producing software has absolutely nothing to do with programming:
* Elicitation (finding out what you actually needing to build)
* Requirements Engineering
* Design (both visual and designing the system).
* Implementation (the actual coding)
* Testing (while TDD is driven by code, the tests are to prove that a component is doing what it is supposed to be doing).
* Release Management
….
I could continue, there is so much that is done on the human side, IMHO on most projects the code has been only a small part of the process.
Edited 2013-11-07 17:50 UTC
lucas_maximus,
Obviously software development processes go far beyond code, however Thom & previous poster were specifically talking about the software itself. Can we all agree that granting a patent to any of these “human” processes would be bad?
Furthermore, can we all agree that both the patent holder and the patent infringer will have put in conceptually similar engineering effort into developing any particular software method? Especially if the infringer was not aware of the patent.
If so, then what is the justification for granting the patent holder exclusive rights on a software method?
I was simply disputing that software was maths, because it isn’t.
I don’t really have an opinion about the patent situation.
lucas_maximus,
“I was simply disputing that software was maths, because it isn’t.”
Then you seem to be misrepresenting the basic argument. You can dispute that more goes into software development process than developing code, and sure enough that’s true. However the real argument is understood to be how the algorithms which make up the resulting software are math.
“I don’t really have an opinion about the patent situation.”
That’s fine, but my point was that the superset of human engineering work going into software which isn’t directly encoded into the end software product isn’t typically relevant to software patent lawsuits. Software patent protection applies to the subset of engineering work that did get encoded into the end product (ie software algorithms/math).
Edit: Of course I realize that none of this debate has any barring on real life. We can effectively be sued over math algorithms in software, regardless of whether they were ever supposed to be patentable to begin with.
Edited 2013-11-07 19:56 UTC
Only in the purest sense. Saying it was maths made sense when software was simpler.
I would say there is theory for say requirements engineering that is without doubt mathematical.
lucas_maximus,
“Only in the purest sense. Saying it was maths made sense when software was simpler.”
Ok, I forgot about the point when software magically escaped the bounds of mathematics…
I am not saying it is not some sort of algorithm. I am saying that I don’t think it is meaningful or helpful to think of it that way, especially considering how modern software is built today.
Well, one could argue that all source code is compiled down to machine language. And if you look at the instructions natively available to a CPU, you’ll note that all of the instructions are math-related (add, multiply, bit-shift, vectors, integers, floats, etc), and (at it’s core) all a CPU does is manipulate memory addresses (numbers) and binary digits (also numbers).
Afterall, what are the major portions of a CPU: the ALU (arithmetic logic unit) and the FPU (floating-point unit). Hrm, those sound like math to me.
😉
And those commands move electrons around an electrical circuit, does it make it electronics.
How far do you go with it?
The hardware is electronics, yes. But the software is, fundamentally, just math.
The point I was making is how far do you go with it?
Most developers won’t care that it is based on maths or can be considered as maths. They see it as a set of components and instructions, each piece making up the application.
Edited 2013-11-07 22:45 UTC
If you’re going to take that route, you might as well include the roads the engineer took to get from his home to his office to code. So making software involves civil engineering as well. Etc, etc.
Software IS Maths, however not all Maths is Science. Also Software Engineering is based on Computer Science (CS), which is ironic because that term, CS, is a bit of a misnomer. I’m just trying to highlight how taking the semantics route is not going to led anywhere.
As usual, there are no “good guys” in corporate antics. This is just a case of an sociopath bully company extorting money from other sociopath chickenshit companies. If Google was in the same situation as MS, i.e. having a noncompetitive mobile OS while sitting on a ton of related patents, they would do exactly the same because their mission is the same: maximize shareholder value by monetizing as much as (legally) possible.
Patents are a flawed and outdated concept needed to make a flawed and outdated socioeconomic system “work.” In the end the guys who actually invented the things in question probably will get orders of magnitude less money from their inventions (most corporations pay a nominal very small fee per patent to the originating employee), than the lawyers and business types doing the suing.
I said I am not particularly interested in the Patent discussion in the other comment tree.
I have to disagree. All the principles are based on mathematical principles, but what normally makes software is how those pieces fit together to form the final artefact.
I would agree that definition made sense when software was much simpler.
I see software development as putting lego pieces together, each block has it own purpose. I wouldn’t call that mathematics, I call that construction.
lucas_maximus,
“I have to disagree. All the principles are based on mathematical principles, but what normally makes software is how those pieces fit together to form the final artefact.”
“I see software development as putting lego pieces together, each block has it own purpose. I wouldn’t call that mathematics, I call that construction.”
This is what math is all about. We give it a different name because it runs on computers, but it’s the same mathematical foundation. Simple concepts get combined to form more complex ones.
Multiplication builds upon properties of addition.
Exponentiation builds upon properties of multiplication.
Discrete mathematics builds upon series expansion of terms containing addition/multiplication/exponentiation.
Trigonometry builds upon discrete mathematics (ie taylor series).
Calculus builds upon all the above.
At some point, the sheer number of computations make it infeasible to do these by hand any longer, so we use software, but the basic premise is still the same. All software, no matter how complex, is ultimately still a logical expression of mathematics.
Data structures build upon set theory.
RSA encryption builds upon properties exponentiation.
Compression builds upon properties of mathematical probability theory.
Graphics engines, AI, UI… and so on and so forth.
I’ll grant you that most software developers are working with extremely high level abstractions of the underlying mathematics, which enables them to specialize even further than they could if they had to constantly be thinking about the low level underpinnings. Even if we treat these high level software components as ‘elemental’, like legos, the way in which we combine them is inescapably governed by mathematics.
It is a sufficiently high level of abstraction that saying it is maths isn’t helpful or meaningful when creating software.
That is why on a separate branch of this thread I have said it is only correct in its purest form.
It depends how much of a pedant you wish to be.
I was interjecting my own opinion, independent of yours. This is an open discussion board.
Confusing the abstraction(s) of a concept, with the concept itself is fairly common.
I am not confused which is which. I just don’t think it is helpful or practical to think of it that way.
It made sense to think of it this way when programs were:
1) Small (relative to today)
2) Were used for performing calculations.
This isn’t the case anymore.
When I am building a web application, I think in terms of components and how they fit together. The tooling and libraries reflect this.
The mathematical nature of programming is not dependent on the size or complexity of the programs.
That you can’t see the mathematical notions is a testament to the effectiveness of the abstractions you are being presented as tools.
This is why software patents (or patents in general) is such a sticky field.
Edited 2013-11-07 22:23 UTC
I am saying it is not helpful to think of it like that as the programs become larger, not that it doesn’t exist.
I do see them, but they the mathematical part is irrelevant at the level I work at. It simply isn’t useful or important for me to think at that level.
As far as I am concerned and people on this thread keep on ignoring is that whilst the concept of a program being mathematics is fine and correct. It simply isn’t useful concept when creating an application.
It is a nice intellectual curiosity when talking down the bar with my colleagues but nothing more.
Edited 2013-11-07 22:41 UTC
As a web designer you’re probably not that involved with or knowledgeable of the intrinsics of the technology you use.
In my line of work, for example, the mathematical nature of programming is more exposed and far from an “intellectual curiosity to talk at the bar.”
I am not a web designer. I was until a few years ago a front end developer. Now I spend most of my time re-engineering large .NET solutions that happen to have a web front end. I would now call myself a Software Engineer these days as I barely touch CSS or Markup.
I also work on bespoke web applications using PHP or Python.
Yes. Software Engineering as a discipline is not Math, I agree. But the end result of software engineering is software, and software is math, it is pure math.
A software program is a series of encoded instructions which when given specific inputs produces specific outputs. Thats it – the rest is fluff. Its math, and as such it is nothing more than a particular way of expressing universal truths, i.e. software = math = 1+1 = 2.
No argument there, but those things have nothing to do with whether software is math or not. There is a lot of similar effort that goes into developing a mathematical proof, and yet the result of all that work is still just math.
Of course this whole argument is moot since the modern courts now use the “too complicated to do with a pencil and paper” test – math is no longer excluded from patentability as long as it is too hard for a 5th grader to do, more or less.
The problem is if the federal courts were following precedent of the Supreme court established in 1972 we wouldn’t have software patents today. Instead in in 1994 in In Re Alappat they decided to simply ignore the Supreme Court and now here we are…
But saying it is maths isn’t a helpful or meaningful definition for software developers. Essentially any program running on a CPU is pushing electrons the right way around a circuit … is it electronics then?
It is more meaningful to think of software program as a series of components (you also create better software this way).
I liken them to virtual lego pieces. Most modern tools and methods of building software are geared for this approach.
e.g.
When building a web application you may have:
* A set of presentation components (bootstrap)
* Client side MVC frameworks (Angular, Ember, Knockout)
* Routing logic
* Server side MVC frameworks
* Server side View Engine
* Data Mapper layer / ORM framework
They are better thought of as components IMHO.
Edited 2013-11-07 21:03 UTC
lucas_maximus,
“I liken them to virtual lego pieces. Most modern tools and methods of building software are geared for this approach. ”
I think you missed the point I was getting at before. I don’t disagree software can be thought of as a bunch of components. But the same is said for mathematics as well.
When we use sin/cos/tan, we rarely think about how they’re mathematically implemented. That’s can be abstracted away so we think in higher level terms. People can use sin/cos/tan without knowing their implementation.
When we use 3d vector * rotation matrix transformations, we don’t think in terms of the individual multiplies or trigonometry, that can be abstracted away. Someone can conceptually understand a vector transformations without strictly knowing how they’re implemented ( making a sin*cos matrix and doing vector math, etc ).
When we want to render complicated 3d objects, we don’t have to deal with individual 3d vectors / polygons / shading / etc, we can use an abstraction without knowing how any of it is implemented.
Each level of development creates more powerful abstractions that hide the implementation. You are arguing these abstractions aren’t math, but I’m pointing out that it’s the exact same kind of convenience abstraction that we’ve been applying in mathematics all along.
This comes back to in it purest form it is Maths, but with modern software this is so abstracted away it can’t be thought of like that.
I would say for things that map directly to a mathematical technique then it should be called maths. If not it isn’t helpful.
e.g.
A recursive function that works out the value of pi to n number of decimal places, I would agree it is maths.
A implementation of a video encoding algorithm, I would say is maths.
Showing a dialog box saying “Press yes to confirm”, I wouldn’t say would be considered mathematical.
lucas_maxiums,
“Showing a dialog box saying ‘Press yes to confirm’, I wouldn’t say would be considered mathematical.”
Ok, I understand what you mean now. That’s a good example. The UI concept is purely non-mathematical, even if the mechanics under the hood are mathematical. The same would be true of most UI elements, even the good old dos prompt.
I don’t think UI elements should be patentable (trademark/copyright perhaps, but not a patent). Never the less, if someone patents a UI concept without patenting the underlying algorithm, what does that really mean? Well, in an indirect way, it would mean that all algorithms that render this UI concept would *effectively* be infringing regardless of implementation. If the patent office were to allow such UI patents to be granted, that would be a subtle yet radical shift in scope from patenting an implementation of an idea to patenting all implementations of that idea.
I am not interested in the patent discussion. I was only ever interested in the conversation about “What is a program”.
The code that triggers the display of the dialog box, the code that positions and sizes the dialog box, the code that coverts the resource representation of the dialog box into a bitmap, the code that loads the bitmap into the frame buffer, the code that draws the dialog box on screen, the code that translates mouse movement and button presses into coordinate events, the code that manages window elements and distributes events to relevant programs, the code that actually handles responding to such events? All math – lots of math actually (and that is an extremely lacking breakdown of what is actually happening – it is FAR more complex in reality)…
Just because you leverage someone else’s code doesn’t mean you are no longer in the realm of mathematics… You just don’t think of it as math because someone else has abstracted all the work away for you (the OS)…
Edited 2013-11-07 22:55 UTC
Lot of maths I don’t have to worry about when I call dialogbox().
At my level I see it as a component, nothing else. It isn’t useful to think of it any differently.
I think you want to force us to think of software at a too high level. Yes, components, yes, modules, yes, applications, yes, they can be combined to perform an action. Yet, all of them are software, code, algorithms, and yes, that is math.
Risking being called an idiot, I’ll still say that all this is not that complicated, it’s just continuously being tried to turned hazy and fuzzy by two parties: those who don’t know enough and misunderstand most related concepts, and those, who understand all too well and purposefully work to confuse as many as they can to keep the current system working. Meanwhile, those who understand and see all the wrongs, try to argue against all the nonsense, trying to explain basic knowledge. Still, informing people is the only way.
Yes it is maths if you are being a reductionist, but I wouldn’t call it that at the level at most developers work at.
I am not forcing my view (how can I?), I just have a different view. I think both are right depending on what level you want to consider what software is.
I only care about what is useful for me to get the job done. It isn’t useful for me the majority of the time to think of it as a mathematical algorithm. What does work for me at my level that I think of each system and subsystem as a series of components.
Consider this:
http://pattern-lab.info/
http://demo.pattern-lab.info/
It has broken down common use cases for creating display components. The guy has created the CSS version of lego.
I don’t care if 10 layers down the browser is working out with some algorithm how to render this, it is irrelevant to me. I only care that it does.
When creating software at this level it is construction. I would say it would be like building a house.
The irony isthat by using the steps described in BABoK you are showing the main issue at hand – business process patents. That process could have been patented as business process, and therefore unavailable to anyone to use. That is how software gets patented not through a category known as software patents, but process patents.
And everything else is just physics. Ergo no hardware patents are valid.
Physical objects can be described by mathematical equations; that’s how physics and engineering work.
Are you arguing that physical objects cannot be patented?
I think that what Microsoft is doing is very immoral, but shouldn’t we blame Google too, for not offering protection to Android OEMs, aka for not standing by their product?
They know how to pocket the 30% cut from the Play Store, but when an OEM gets patent-trolled, they are all like “we just offer source code, you are on your own”.
Anyway, this probably won’t continue for ever. Motorola hasn’t paid MS a dime in extortion fees, and probably won’t, so, once the other OEMs realize that making an Android rom free of MS patents is possible (just do what Moto did), they will stop paying the extortion fee, just like companies stopped paying the linux extortion fee to SCO (but if Google had stepped up to protect Android like IBM did with Linux, the Android extortion fee would have already been history, damn!)
Edited 2013-11-07 11:19 UTC
They tried to somehow. I am not saying that they cannot do more, but they did try to form a patent pool to help against trolls. This is why they bought Motorola. They should be more aggressive on this.
They should have offered real protection for OEMs, just like Red Hat did. Take the whole responsibility.
1) Different times
2) Red Hat offered it for customers who payed, not for the free fedora
3) It was important for Red Hat to do this, because they had all kinds of clients, while mobile OEMs are usually huge companies, and also the money asked are not so much, compared to SCO case.
4) I’ll give you this: I love google compared to other tech giants, but Red Hat is a true open source company… mentality, business, involvement… Over all the greatest open source company
[QUOTE] Anyway, this probably won’t continue for ever. Motorola hasn’t paid MS a dime in extortion fees, and probably won’t, so, once the other OEMs realize that making an Android rom free of MS patents is possible (just do what Moto did), they will stop paying the extortion fee, [/QUOTE]
I would love this story to be true.
But unfortunately, those “other” Android players must take into consideration that they are “foreign asiatic” companies, being brought into a US-based jurisdiction against a US-based company.
They are not stupid, they know they will lose, for “valid reasons” which will be essentially “invented on the ground”; they won’t get a fair fight, as Motorola would get against Microsoft.
And Microsoft knows that too.
Edited 2013-11-07 11:33 UTC
When did this happen?
Motorola got around ITC import bans by assembling local. That is, you can’t ban imports of non-offending components — the final assembled product (with Android installed) is what is infringing.
Monetizing your platform with money from a competitor is pretty much a master stroke, and Google has to hate every moment of it.
Android’s troubles seem to be about to get worse now that Nokia is weaponizing their patent arsenal. Pretty soon one has to wonder how much longer Android remains cost effective to OEMs.
With Google’s patents from Moto pretty much neutered, and Rockstar knocking at their door, it looks like Google is powerless to stop the impending shit storm.
The fact that you are willing to sacrifice the progress of technology just to stop Android speaks volumes about you and your ilk – and all just because of the stocks you own.
Sickening.
I’m not doing anything, except investing after the fact. And that’s only because I like putting my money where my mouth is.
I just live in the real world and assign a tangible value to a company’s patent portfolio. IP makes up a considerable portion of my own valuation of Nokia. And FWIW, I’m not a MSFT shareholder.
Nelson,
That just proves that you have common sense when it comes to investing.
This will not end well for the OEMs. Google can pay and will. Many of them run thin margins. Rockstar, Nokia, and other patent holders will attack them and knock out many of the smaller players like Pantech.
This will have the net effect of causing cross-licensing with the big players like LG and Samsung to alleviate the costs, and the removal of smaller OEMs.
This is market consolidation.
Hey, how do you get the cool orange bar? 😉
What about Fraunhofer and the MP3? are they criminals?
I can’t see any evidence in what you have written showing that progress in technology has been impeded by patents.
Patents and licensing schemes are the only means we have of protecting past invention in technology which motivates new R&D, both large and small players benefit financially from patent rights and licensing.
LOL, there’s one born every minute, isn’t there?
Thom isn’t it just a sickening to purchase the products of companies like Microsoft Nokia and Apple. Do you own any products from these companies. Will you continue to purchase products from these companies?
A ‘master stroke’ would be creating product value with which they could actually compete against Google in the open market place, but they can’t so they instead turn to whatever dirty tactic they can come up with (all in the grand old Microsoft tradition), which in this case means software patent aggression and smearing campaigns.
I find them a shitty company through and through, they are even so far gone that they are now backing Oracle’s insane attempt to have API’s be copyrightable.
How any developer can stand behind such a company is beyond me.
In a world where certainly no tech company is a saint, Microsoft truly belongs at the bottom of the barrel.
Microsoft is a huge company and as so it is an amalgamation of distinct, sometimes unrelated and on other competing interests. I am sure there are people inside MS that don’t like situations like those with Android royalties and current form of software patents. Unluckily, the top management and the lawyers are not between them.
Who is going to prevail in the end? My bet would be on Google. SCO failed against Linux with all the help Microsoft could provide, Oracle failed against Google, again with help from MS and Apple, and Nokia/MS may very well fail also.
They are monetizing a competitors product and using it to finance their own product. That’s strategic genius. Your rant aside.
I don’t think “genius” is the proper word on this case but the strategy is lawful. And so was also slavery in some of humanity history. :-p
That you consider the act of patent extortion to be ‘genious’ says a lot about you, it’s a simple form of a shakedown which has been around for ages.
In reality it’s nothing genious it at all, it’s just a matter of being willing to stoop so low as to use the insanity that is todays software patents to try and attack your competition.
Of course Microsoft is on the forefront of such behaviour, as it’s seemingly part of their DNA.
And while Google can put up a defence, that is not something the vast majority of tech companies can, which means that the huge tech companies who amassed these ludicrous ‘patents’ (and obviously had a huge hand in having software patents made legal to begin with) are now using it to try and destroy competitors that they would otherwise have to compete with in an open market.
Even going so far as to purchasing a bunch of third-party patents all in an effort to attack a competitor with which they can’t compete by the value of their own products, -‘if we can’t create products consumers want, we must instead try to remove the other products!’.
The end result is a very bleak world for developers and entrepreneurs, but even more so for consumers, as when there is no competition to speak of, you are effectively locked to a platform because there are no viable alternatives, with technology stagnation and high prices as the result.
This is what software patents ultimately brings us, a future where the entire software market is controlled by few large tech entities who owns and continously file new broad patents to encompass any possible functionality with which they can easily crush any smaller company through threat of long costly legal battles.
And even if a small company would end up winning, ‘there’s always another broad software patent to litigate with’.
Again as a developer I can’t understand how people who claim they are developers can stand behind this behaviour.
I guess it depends on whether you want a healthy competitive market or a few extremely rich companies that own the market.
In the latter case it could probably be considered “genius”.
Can someone please post a link to an article/ analysis that says how much money MS receives for ‘Android patents’? I’d also linke to know what these patents are about.
I’m asking this because as far as I know Microsoft only gets money from companies that sell Android-stuff AND Microsoft-stuff or are in “cross licensing deals”. Those companies pay one big invoice, and Microsoft says that it is for Android – for whatever reason, possibly to spread FUD in the market.
One of the few companies that only sold Android but no Windows, was Barnes & Nobles. When the received an “Android patents claim” from Microsoft they immediately started a lawsuit. Soon after the (very weak) patents were disclosed, Microsoft ended this embarassement by simply buying Barnes & Noble….. Problem solved.
The only “patent” that might have some value is the xFATx patent. Still doubtful, because there is prior art.
Anyways, could you guys name me ONE company that does not sell Microsoft stuff and is not in “cross licensing deals”, but that pays Android patent money to Microsoft?
I’m doubtful because 1)if the patents were strong, why not disclose them? 2)they have never been tested in court 3)MS didn’t contribute to Android 4) how likely is it that a company that ‘s been struggling for years with it’s own mobile OS, would hold patents for a different (but highly successful) OS?
Suppose General Motors was struggling to develop a good electric engine & battery management – how likely would it be that they had essential patents in these fields, and forced Toyota, Nissan and Tesla to pay for it? Very unlikely, I’d say. Hey MS, if you’re so good at mobile Operating Systems, how come your own OS sucks so hard?
You just don’t like Microsoft?
Patents are public. Details of the license agreements, lawsuits, and settlements are not always.
Sure they have.
http://www.geekwire.com/2012/microsoft-invests-300m-barnes-nobles-n…
Sure they did.
http://www.linuxfoundation.org/news-media/announcements/2012/04/lin…
Very likely when we’re talking about software patents.
Microsoft’s contributions to Linux don’t exactly count as contributing to Android… most of that work involves support for virtual machines and related server-oriented stuff… nothing of interest to Android or other mobile devices…
I said “Microsoft’s alleged Android patents have never been tested in court”. You said “Sure they have” and included a link to an article about Barnes & Noble versus Microsoft SETTLING the dispute. Hey, I knew that all along, that’s why I said that the claims have never been tested in court. Microsoft seems to avoids that (and why would that be? heh heh heh….)
Then you say that MS didn contribute to Android, and you link to an article about MS contributions to Linux, which are all about enabling Linux to run on Microsoft’s hypervisor…… in other words, not a contribution to Android.
Compared to Microsoft, Oracle, or Apple, Google is young company and they basically just started buying it’s own patents. There’s no way that they can match with older corporations in the next 10 years and everybody knows that. After some time they’ll have something to cross license.
Frankly, I don’t like how the patent system works (or, usually doesn’t for that matter). Software patents in particular are terrible and the trolls who really abuse them SHOULD be considered criminals (that would be those who just own patents and do nothing with them).
In this case, however, Microsoft uses its patents and is actually trying to actively compete in this market (and were before Google got involved with phones at all). So I won’t hold it against them.
Microsoft is a business after all and must do what it perceives to be in its best interest as such. Business is a war and wars aren’t pretty.
Patents are not contracts between companies. Patents are contracts between holders and a government, representing it’s constituents.
Those constituents loose in the long term when patently abuse happens.
After establishing of patent system more complex inventions became more achievable, i.e. bar of obvious invention should have been brought up. Instead companies learned how to game the system and pushed the factual bar down making trivial ideas patentable.
Here is why is that pattern breaks the patent contract: http://blog.didenko.com/2013/03/patentless-it.html
*If* you want “software patent protection” over and above copyright, trade secrets etc..
(A) your code needs to be scrutinised harder than currently and held to a much higher of ingenuity and inventiveness by a PANEL of known experts in various software types
PLUS (B) if your patent is granted, all your code has to be be then published on a central repository of patented code. So basically if you want and or need this added protection of the patent, your code has to be effectively open sourced – not ostensibly for any political reasons, purely for visibility and openness, people NEED to be able to check their not infringing existing code when they’re developing new code after all.
All current patents should be assessed against the new standard and either (1) tossed out, maybe with a refund of the application cost but no more.
Or (2) offer the chance of becoming a new upgraded patent but with the proviso that the code needs to be published on aforementioned public repository.
*or, the alternative* will be to not patent your code, have existing copyright protections and trade secret protection etc, and get to keep your code secret longer term too – obviously open to reverse engineering but you pays your money and you takes your chances..
Is that a silly idea…… ??
Please note that I actually agree with your point of view on this – the problem is you aren’t understanding the nuts and bolts of the current system…
There is currently no such thing as a “software patent” – what people are really talking about are ostensibly method/utility/design patents, but applied to software. The problem with code review for patents is there is no code to review – no one submits code because they are not patenting software, they are patenting what the software does (specifically when combined with the hardware it is run on). Source code is immaterial to how current patents are reviewed and granted.
I’m not disagreeing with your intent, but ultimately before you can actually call for code review you would have to actually legitimize the fact that these things are really software patents – something those with skin in the game do not want to do because it removes the cover that the federal courts have created for them with this silly “process patent” concept.
In short, you cant have code review because having code review means you have legitimized software patents, and the minute you decide to call a spade a spade the federal courts will be forced to deal with the fact that they have been ignoring supreme court precedent for the last 40 years…
Again, that makes no sense under current law (and has no hope at all of ever being accepted as new law – an open source requirement is simply not tenable).
Copyright != patents…
For example:
1. If I verbatim copy a copyrighted complex function (say over 2000 lines of code involving complex logic) and use it in my code I am violating copyright law.
2. If I do the same thing from a “patented” product, where that function is central to the patent, but I remove a bit that isn’t relevant to my use case and use it for a different purpose… I am in fact NOT violating the patent, assuming that the patent does not cover my actual use case.
3. If instead, I right a completely original function (with no knowledge of the patent) that technically works in a very different way, but subjectively performs the same function as the patented product, following the same high level steps, and is used in the same manner for the same purpose, I AM violating the patent.
Publishing the code of “software patents” won’t have any effect on making it easier or harder for developers to know whether or not they violate one, because the patents have nothing to do with the code used to implement them.
No, its not silly at all. Its just not applicable to the patent system, because patents have nothing to do with source code. I don’t know of a good solution other than to just outlaw the application of patents to software – I have seen nothing else that makes sense to me.
galvanash,
You hit the nail on the head. Still, we should thank the OP for trying to come up with actual solutions instead of just complaining about the current system.
Alas, I don’t think anyone has the political power to stand up to the corporate machine in government to change the status-quo. Anyone challenging the incumbent corporate positions is virtually unelectable because of how important corporate support is for politicians. For it’s part, congress never authorized the “software patents” that are being issued by USPTO. If they were to take action today it’s more likely that they’d be pressured into officially recognizing them.
Disregarding what the patent system is today, it would make much more sense for patent protection to be exclusive to copyright protection. After all, one of the justifications of patents is that the public gets to duplicate the invention. Politics aside, logic would dictate that we really *should* be entitled to the source code.
mistersoft,
“All current patents should be assessed against the new standard and either (1) tossed out, maybe with a refund of the application cost but no more.”
Good lucking getting the funding for that
Many patent holders will probably return their patents on the basis that they weren’t making money anyways.
Edit:
Someone else on osnews (I forget who) suggested that perhaps instead of a patent system, the government should just buy inventions outright to make them public. They’d get paid for the invention, and the company would no longer have claims to the invention. This idea sounded quirky and wasn’t well received here, but I think it was nevertheless an insightful way to motivate genuine innovators, truly maximize public utility, and completely eliminate this very costly epidemic of patent lawsuits. It’s controversial and would never pass, but it’s good food for thought on the subject of alternatives.
Edited 2013-11-09 08:48 UTC
I feel that copyright protection should also have the requirement of “source code” (particularly for music or films, to remix them easily after the copyright lapses), stored in escrow…
Only for books (etc.) the work is at the same time its “source code” – but text is what copyright protection was created for.
Edited 2013-11-12 19:16 UTC
Software patents hold down innovation. Every idea you come up with, you have to lookup because it could be infringing on some rediculous patent that says “a method for moving from point a to point b”.
The only thing that should be patented is specific code, as to avoid plagiarism. Code style is unique and if someone copies an implementation from code, it shows and is proved quite easily.
If books had the same system of patents one wouldn’t be able to write a story about… space travel because Paramount owns the rights to Star Trek.
I am very glad we don’t have software patents in Europe….