PG suggests that this won't stop the trolls but it might deter more traditional companies.
Does anyone have stats on who is doing the most damage to early companies? Given the press, it's easy to think that trolls are the biggest offenders by an order of magnitude. Is there data that suggests otherwise?
The two types of attacks are so different that it's hard to compare them. Patent trolls attack more victims, but all they want is money. Attacks by competitors are much rarer, but more damaging when they do happen.
I'd be surprised if the trolls (except Lodsys) are causing a lot of grief for companies under 25 employees. They seem to be more interested in a smaller number of bigger scores.
Indirectly, this pledge my also help to slow down the trolls. Trolls buy up patents in order to sue or resell. Startups have no money to pay up, so the only option is to resell. But, if a potential acquirer has pledged not to use patents against small companies, buying them off from trolls wouldn't make for a good public image. As a result, trolls won't press on the patent holders, at least until they are acquired by a company that is capable of paying the bill.
... thereby making it cheaper for trolls acquiring patents, like Intellectual Ventures, as there is less market for the patents -- but not slowing them down.
I like the idea of pledging, though there may be some variance (what if a startup violates a startup, margin is high, the technical innovation was real, etc?) That being said, there is a lot to be said in simplicity.
I'm no lawyer - I have to ask the logical question - does publicly stating this pledge bust any opportunity to double back (i.e. it is more legally binding than just a pledge?)
The content of the pledge seems to indicate that there are a lot of (or at least some) cases where large companies are suing very small companies (< 25 people) over patent infringement. Is this the case? I've only heard about the patent litigation between the tech giants, and not anything about small firms getting sued by larger ones. Are there any recent/high profile examples of this that I missed? Or is it just something that goes unreported?
The recent This American Life ep "When Patents Attack" (http://www.thisamericanlife.org/radio-archives/episode/441/w...) started with a segment on a startup called Fototime, which was very small (certainly under 25 people) and had to settle a patent suit at great expense. (As part of the settlement agreement, they aren't allowed to say how much; only that it was just short of what would have caused them to file for bankruptcy.)
Incidentally, for those who are curious, the three patents used against Fototime (as well as more established companies like Flickr) can be found at http://www.google.com/patents?id=uiAPAAAAEBAJ, http://www.google.com/patents?id=0J8DAAAAEBAJ, and http://www.google.com/patents?id=-QEWAAAAEBAJ. The first two actually feel substantial and specific (probably too specific for Fototime to have been infringing); the third one is just a simple database schema and a flowchart, which is madness (though it, too, is so specific that it's unlikely that anyone was actually infringing).
So if a company grows beyond it's sub-25 people are they expected to then license any technology they're infringing on? Could this lead to an even bigger penalty if the company is made aware that they are infringing when they are small and doesn't act on it when they grow, thus willfully infringing?
I think the idea is that if the company is successful enough to have grown to 25 employees, it stands a better chance of having access to competent legal counsel and other resources. The penalty might be bigger in monetary terms, but smaller in terms of making the founders' lives a living hell.
Do non-trolls really bring patent cases against small companies? A company of <25 people probably doesn't have the cash to make a financial settlement worthwhile, and if a small company has a good product it'll have >25 people soon enough.
Companies with less than 25 people are relatively unlikely to have sufficiently deep pockets to attract patent attacks in the first place. And trolls, of course, won't care about the pledge. Nice idea, but doesn't seem that useful to me.
This is ultimately not a helpful avenue to pursue. Asking companies to please not abuse a favorite group of companies (in this case startups) is not a solution to this problem. It's very similar in my opinion to the patent exceptions being carved out in congress right now for the finance industry (their favorite group of companies). We need to be striving to help everyone with patent reform, not just our favorite types of companies.
Yeah, I'm not really afraid of Microsoft -- especially if I don't ship a product based on Linux or Android. And patent trolls aren't going to agree to this pledge.
Patent trolls worry me the most. I'd like to see a pledge around having the tech community pool together against patent trolls. Not sure how you'd do it, but I'd like to see that.
That doesn't scare me much. They were trying to reverse-engineer a MS file format. If I'm doing things like trying to reverse engineer Silverlight DRM or get around PatchGuard, or find a way to circumvent some limitation in a Windows SKU (like installing Media Center on Starter Edition) sure I'd be worried. But if I'm just building product that infringes a patent they have, I'm not worried.
There was a case of a startup shoe sales site with a really elegant interface for selecting the exact type of shoe you wanted. Basically, it allowed you to see many very similar styles on one page to be able to compare similar offerings from different labels. I wish I could remember the name, but a recently VC-funded startup used a patent to go after the first startup, and they ended up closing shop.
It solves the chilling effect of patents on startups. Many startups have avoided entire domains because of patents. Large enough companies can negotiate reasonable license terms or fight patents in court, but small startups cannot afford the risk of a dispute.
I appreciate the good intention of this proposal, but it doesn't really make so much sense. Tech startups are small businesses, but their goal is to grow. With that pledge you could only grow up to 24 employees, and what then? You're ready to be slaughtered?
This problem needs to be fixed at its root, with a different law.
From what I understand, the key points of software patent reform would be (a) significantly raise the bar of "non-obvious" and (b) shorten software patent lifetimes. The related issues of small companies being at a disadvantage (due to economies of scale with litigation and patent portfolios) seems rather orthogonal.
Since this pledge would only address this issue of secondary importance, which seems a lot less salient to the public, I can't imagine it getting off the ground.
I disagree that this will help, because the established companies the pledge would apply to are a secondary problem and mostly seem to fight each other (has Microsoft asserted patents against a startup? has IBM? has AT&T? when they asserted patents it was against multi-million dollar businesses!). The primary problem is patent trolls (see e.g. lodsys / intellectual ventures) for whom this pledge could be considered self-harm.
I will quote myself from [ http://news.ycombinator.com/item?id=2855835 ] here for another solution, one that actually can _easily_ go through government (except for the intense lobbying against it by whoever enjoys the current patent regime); you can read there for some discussion if it is interesting. Quoth myself (with minor editing):
Intellectual "Property Tax". Have everyone declare the value of their intellectual "property" (patents, copyrights, trademarks) - each and every item, for that year, on their tax return, and have them pay 1% of the value as "IP tax", per year.
Clarification: you can set a different value every year. The value may drop to zero because a competitor's patent solves the problem better; or it may go up because it becomes essential to something that becomes commonplace.
That amount is what one pays for a compulsory license or if successfully sued, and up to 3 times that for willful infringement, per year -- and no more. (But of course, a patent owner can always negotiate a lower payment, as is done with music recordings that have compulsory license agreements)
All of a sudden, everyone has an incentive to state a reasonable value for their patent. Copyright catalogs that are not being published (old music recordings, old books, old movies) would be assigned 0 value by copyright holder, to avoid tax - which means anyone can freely make a copy. If they believe -- at the end of the year -- that someone is making a profit at their expense, they can set the value as high as they want at the end of that year, pay the tax, and sue the profiteer.
Simple, elegant, and coffer filling.
edit: put missing link
edit: added clarification about setting value each year anew.
Everyone will game the IP tax by valuing it at pennies -- after all, they may not directly be using it may not have a fair market value before they take someone to court.
As a real-life example of people gaming these types of taxes, people buy expensive (>100K USD) cars in europe directly (i.e. taking delivery in stuttgart) to save on the sales tax: driving it around for a bit ensures that the car is technically "used" when brought back here, circumventing new car taxes.
I haven't thought through OP's proposal, but your objection is covered - by valuing a patent at pennies, a company would cause compulsory licenses/willful infringment penalties to also be worth pennies.
No but the taxes would have to be paid only after a successful patent case. The point is that the patent has no real market price until the first successful willful infringement penalty. Until that point, the price is arbitrary.
Until _before_ a patent case. The taxes are what gives you the ability to sue (and a limit to the damage you can be awarded), and there is no guarantee you will win your case of that your patent will not be invalidated.
Again, you have to pay the tax _before_ you can sue, and then you can only sue for the value reflected by the tax for that year, no more (and you are still subject to invalidation, etc).
I think that the idea is that if you declare the taxable value, then later try to sue someone for a much higher value, then you've now committed fraud against the IRS.
Of course, the easy work around to this is the claim that it's value increased suddenly when the patent became more useful; you start paying more tax when you start suing others.
Considering that low maintenance costs don't prevent domain squatting, I'm not sure this would be an effective tactic to prevent lots of silly patents.
You need a carrot for Congress. Right now, a patent reform (of the kind HN would generally like) will:
a) reduce government intake in the form of patent registration and maintenance fees
b) get all the lobbyists employed by current patent holders (who spend billions of dollars on those patents) working against you.
Why on earth would congress do something like that? So that 0.05% of the population that actually has any idea how bad it is would rejoice? (If your answer is "to improve american competitiveness" or something like that -- well, you have no idea how congress works)
However, if you can get a new spending source, you just have the lobbyist problem to take care of.
Improves innovation which will eventually lead to more tax dollars. Although it would be a stretch to get them to take a long term view, would also probably have lawyers lobby against it.
To clarify: they can't adjust the value retroactively, right? If you find out someone is profiting from your patent in 2011 which you valued at zero the beginning of that year, you have to wait until 2012 to re-value (and potentially sue them), correct?
To clarify: the declared value puts a _maximum_ on awarded damages, but does not make it in any way automatic (even if the product is found infringing).
My proposal is that it is set retroactively one year back: when you file your 2011 taxes (on April 2012 at the latest), you would assign a value for it for 2011, and pay it at that point.
If you sue (or even threat to sue) someone in 2011 before paying the tax, that is an admission of tax liability - if you declare less at the end of the year, that's grounds for the IRS to audit you and assume your highest estimate during the year (as shown in the threat to sue) sets your tax, and anything else is automatic assumption of tax evasion.
A future-looking only assessment is also possible; but in that case, you have to be able to set the value (and pay the tax) on a monthly basis -- otherwise, it calls for anyone who is willfully infringing to wait until 1-jan, ask for the IP value, and if it is less than they expected to pay, have a year of "free" use.
My proposal is an attempt to sort-of balance the existing state of affairs -- if at the end of the year, you didn't manage to license it to anyone, you shouldn't be taxed for what you thought it is worth.
I don't get it then. If my company holds a huge portfolio of patents, and we sue for $X damages on some subset of them in a given year, we will simply value those patents at $X and the rest at zero. Then we recover $X in damages, and pay 1% of X in taxes.
In other words, doesn't this just reduce to a 1% tax on patent damages?
(Even if we think we'll only recover on average p% of our claims, we still only pay 1% of X in taxes to get p% of X in damages. Assuming p% is closer to 50% than 1%, not much changes.)
No, it does not reduce to 1%. At worst, it will reduce to (1/p)% (with p<1, you assume p=~0.5), but probably much, much more: All of these processes realistically take months to years (have you noticed how long SCO vs. Novell was going on?) - if you "deactivate" a subset while you're still suing, that reduces your damages. So, you'll have to keep it active, say three years - which with p=0.5 makes it a 6% tax on damages.
However, it does chill the chilling effect - right now, e.g. Microsoft has ~20,000 patents, and they can sue anyone for anything. This way, they will actively have to choose strategically what they want to enforce.
Finally, and I think I haven't made this clear enough, I don't think my proposal is a GOOD patent system in any way. It is bad. But it is not as bad as the existing system, and (I suspect) it has a much better chance of passing than a complete reform.
Furthermore, it stops some of the (currently legal) tax loops that involve passing patents among shell corporations.
The basic idea is to align value with cost. Any system in which the value of the patent is potentially infinite (as in the current system) is materially broken, and it is just a matter of time until someone finds a way to game it (non practicing patent trolls). Any sane system must have cost (of patent, property, copyright, etc) rise at least linearly with value.
Fine, so it's a (n/p) times 1% tax, where p is the average chance of victory and n is the average number of years in court for each suit. This doesn't seem to change the structure of the incentives at all; it's just a tax on patent infringement damages.
I find the non-retroactive possibility much more interesting.
I don't disagree a non-retroactive possibility might be better; I just haven't been able to reach a conclusion about it.
And, it is more than just a tax on patent infringement damages - because you have to pay the tax upfront, which makes a big difference in cashflow.
eolas's $600M verdict against Microsoft, and i4i's $400M verdict would both require posting $6M / $4M per year respectively in never-to-be-recouped taxes, which is a huge sum for the companies in question, and would probably have made them not sue.
Furthermore, it involves the IRS in valuations, which is a good thing (because it is much harder to mess with the IRS) - manipulation of the value of a patent would have to be explained, and not arbitrary to avoid taxes.
Anyway, we could all dream - it is the lobbyists that write the laws, and the patent holders have thousands of them in DC.
> Any system in which the value of the patent is potentially infinite (as in the current system) is materially broken, and it is just a matter of time until someone finds a way to game it (non practicing patent trolls).
The biggest problem with the patent system is it discourages people from using patented ideas - the more patented ideas you use in your product (even if you independently discover them), the more you pay.
A good patent system would have the following properties:
* Investment and risk taking in R&D and sharing the results is rewarded above the costs of that R&D, so there is no disincentive to share with competitors.
* There is no disincentive to using and building on ideas shared by others.
I think the best system is a 'tax'-like system - alongside the existing patent system, businesses can elect to pay 10% of their revenues to a pool for immunity from all patents. Businesses using / reselling patented manufactured products are still liable even if the manufacturer pays the 10%, unless they also pay the 10%. Patent filers tell the pool the actual costs of R&D (and may be audited to prove it). The pool is distributed based to patent filers based on R&D costs (and maybe a per-industry factor for risk) - it might be better to somehow collect some rough metric of value of the each patent to their business from patent users, although there are risks of the system being gamed then.
This system would achieve the desired properties at 10% of all industry-wide research being spent on R&D - it would favour not disclosing R&D worth more than 10%, and if less than 10% was spent, it would be stacked in favour of companies doing R&D.
The problems people note are solvd if you issue a patent free for a short time, like 3 years, and then auction a much smaller number of extensions to that term. This creates a pricing maximum for years 4 to whenever.
I suppose it would be a good start, but the self-interest in this proposal stinks a bit. What's the distribution of employee numbers in companies in which Y Combinator has a stake?
Of course, the bulk of the valuation is in the startups that have grown enough. You still have vested interest to protect startups while they are small, this is when you add most value anyway.
AlexBlox asks in an earlier comment: "does publicly stating this pledge bust any opportunity to double back (i.e. it is more legally binding than just a pledge?)"
A court might well hold a company to such a pledge, on a theory of "equitable estoppel." This type of defense to an infringement charge is always highly fact-specific; here's an example of a case in which the defense succeeded:
A patent owner accused a manufacturer of eyeglass frames---which it had previously sued for infringement---of infringing other patents. After back-and-forth correspondence---in which the manufacturer denied infringement---the patent owner went silent for three years. In the meantime, the eyeglass manufacturer expanded its marketing efforts for the products in question.
The trial court held that the manufacturer was not liable for infringement, on grounds that the patent owner's actions, in view of all the circumstances, had misled the manufacturer into thinking it would not be sued. The appeals court found no error in this holding [1]; it explained that:
"In the context of patent infringement, the three elements of equitable estoppel that must be established are:
(1) the patentee, through misleading conduct, led the alleged infringer to reasonably believe that the patentee did not intend to enforce its patent against the infringer;
(2) the alleged infringer relied on that conduct; and
(3) due to its reliance, the alleged infringer would be materially prejudiced if the patentee were permitted to proceed with its charge of infringement."
In that case, what's stopping large company X from forming a shell company of one employee which performs services for X using patents held by company Y which has made such a (potentially legally-binding) pledge?
I think the whole YC gang is going to promote this aggressively, which means a strong network effect. Remains to been seen what happens outside this network.
Well, Stripe is not YC, but your argument makes sense. I don't know how others are supposed to take this pledge "officially" other than by posting a blog post and emailing pg however. Maybe there's a secret cabal that of startup people use as a channel to converse anyways though ;)
This is a nitpick but I think the pledge should read "No first use of software patents against companies with fewer than 25 people," since people are countable:
It's not immediately clear to me whether this solves any part of the current problem. But on reflection, I believe I understand the motivation here.
Big companies that use patents as a revenue stream (MSFT, IBM, etc.) typically bide their time and bring a patent lawsuit once a new company is established and there is blood to drain. It's the threat of such a lawsuit in the future that can negatively impact investment in a startup, as the right collection of patents could conceivably capture much of the economic surplus of a new venture. Alternatively, a big company might use the threat of a patent lawsuit, now or in the future, to push a young company to agree to an early acquisition.
The pledge doesn't seem to have much impact on these scenarios, even if a big company were to follow it rigorously.
Most of us, I believe, would prefer to see companies make a stronger commitment: "No first use of software patents" [period]. Google hasn't made this pledge, but to the best of my knowledge, they've acted in this way so far. It does seem in line with "don't be evil."
That said, I think I see what PG is going for here. He wants companies to make a pledge that, at a minimum, allows a new product or service to be tested on the market. That way, if it gathers traction, it will attract investment despite the threat of patents, and the new company will be able to mount a reasonable defense.
Perhaps more importantly, though, by allowing the product to succeed first, even in a modest way, it makes the offensive use of patents worse PR for the big company. Killing a successful product with patents is no longer an abstract issue. It takes away from customers and the market something very real.
'Most of us, I believe, would prefer to see companies make a stronger commitment: "No first use of software patents" [period].'
Good idea, but it can't work. If the PTO is willing to grant a patent on a software idea, someone is going to get the patent. If Google stands down, some troll will get it instead, to the detriment of society.
OK, if so, who is going to pay the $10k+ in costs to get each of these patents? Whoever pays to get a patent and not enforce it does society a service, but bears the cost themselves.
This line is repeated everywhere now, and it is hard to swallow. A deterrent is only useful if you employ it from time to time to demonstrate its power.
If you have offensive capacity you will be bound to use it.
Guns don't kill people, People kill people.
Sure, but guns make it much easier, and by design, have the capacity for landing in the wrong hands.
Over 50 million households in the US own 200 million or more guns, collectively. If these guns were usually used this way, the US would be depopulated by gun violence in a decade or two.
I don't understand how one would defend their patented innovation without suing someone else? Or are you suggesting that the patent only be used as a counter-suit after being sued for infringement of a different patent?
I mean, the whole point of a patent is to be able to sue people who are competing with you. How can that ever be defensive?
Unlike trademarks, you can hold a patent without having to defend it. You are not legally required to use it. So the choice to use it is yours. The logic behind defensive patents is similar to creating nuclear weapons as a deterrent and the theory of mutually assured destruction.
I think you're mistaken about superpowers flinging nukes at each other. Only two nuclear weapons have ever been used in war, and that was 65 years ago.
It was analogy / metaphor.
All the superpowers == all the big companies (Google, Apple, Samsung, etc)
routinely fling nukes == lots of patent litigation.
If you patent an invention, no one else can. Your patent prevents other people from patenting the thing that you depend on.
If your next question is that the existence of your invention would invalidate their patent due to prior art, that would take costly legal maneuvering to establish. Is there a shortcut around that legal maneuvering? Yes, patent your invention first.
If you want to start a movement, you can't aim too high initially, or you just stall. The patent pledge in its current form is at least fairly easy to swallow. If it sticks, it will have the effect of making patent suits against competitors seem more dubious, which would in turn prepare people for more exacting versions.
How far ultimately would you want to push it? I honestly don't know yet. Patents may have some utility. It's a very complicated question. I've read a lot about the history of technology, and I can't say for sure whether things would have gone better or worse without patents. The topic seems to be one of those where on average the strength of people's opinions is inversely proportional to how much they've studied the problem.
Certainly, biting off too much at once isn't workable. But then again, biting off too little can stall a movement too, as it can fail to inspire as much passionate support as a somewhat larger bite. Starting a movement is a hard problem -- a significantly harder one than making money. I appreciate your effort to address the problem, regardless of which way it goes.
Let me address your last sentence, though. Your statement presupposes a utilitarian moral system. That is, it assumes that a person's position is completely determined by which policy will result in the greatest growth in technology.
I've studied the history of patents as well, and I share your practical uncertainly over which policy inspires more growth. Patents may, in fact, grow certain types of technology faster while inhibiting growth in other areas. Since we need to weigh the harm caused by the lack of things that never existed against the benefits brought by things that do, it's probably impossible to do a quantitative analysis that is fair to both sides.
However, I and many others oppose patents on deontological grounds. That is, even if the existence of a patent monopoly produces some technological benefits as compared to a non-monopolistic system, we would choose freedom from monopoly.
John Carmack may have stated the essence of this position better than anyone since Jefferson [1]:
"The idea that I can be presented with a problem, set out to logically solve it with the tools at hand, and wind up with a program that could not be legally used because someone else followed the same logical steps some years ago and filed for a patent on it is horrifying." -- John Carmack
[1] "If nature has made any one thing less susceptible than all others of exclusive property, it is the action of the thinking power called an idea, which an individual may exclusively possess as long as he keeps it to himself; but the moment it is divulged, it forces itself into the possession of every one, and the receiver cannot dispossess himself of it. Its peculiar character, too, is that no one possesses the less, because every other possesses the whole of it. He who receives an idea from me, receives instruction himself without lessening mine; as he who lights his taper at mine, receives light without darkening me." -- Thomas Jefferson
What is a good example of something that would have turned out worse if there hadn't been software patents? (I know you've argued before that software patents aren't fundamentally different than patents on anything else, but to a lot of people software patents have more obviously negative consequences).
The most famous example is probably Watt's steam engine. Bolton & Watt had high development costs and could not have gotten funding without their patent.
But once established they definitely used patents against smaller competitors. So it is hard to say for sure even in this case whether patents were a net win. They probably were though. The patent probably caused modern separate-condenser steam engines to happen at least a few years earlier.
(I recommend the story of Bolton & Watt to present day founders. It is remarkable how little has changed.)
Oops. I don't know of anything that would have turned out for the worse without software patents. But I only know about a subset of uses of software. There might be cases in pharmaceuticals or manufacturing where patents helped protect individual inventors against big companies.
It seems pretty obvious that if instead of patent system we had a mandatory obligation to publish full specs (schematics and source code) for all marketed products the invention rate will grow exponentially. The development costs will be cut down because of possibility to incrementally improve on existing products.
Strange. That directly contradicts Against Intellectual Monopoly[1], which claims patents did more harm than good in this very case.
One reason is independent discovery: most innovations tend to be discovered independently and nearly simultaneously. Like, when "science is ready", innovations just pop out of several people's mind relatively quickly. (See also Eben Moglen's folk theory of the internet and innovation: "wrap the internet around a planet, spin the planet, and ideas flow out of the network".)
If we accept that, the only thing that could have significantly slowed the development of the modern separate-condenser steam engine down is money. Like, no single one investor is willing to risk making the first step, by fear of being eaten by copycats. If this is so, then we have a tragedy of the commons.
Risk aversion in this case is indeed a problem. I think it is mitigated however by (i) the first mover advantage, and (ii) variations of risk-aversion among individuals (combined with independent discovery, you get a winner-takes-all scenario).
Ah, before I forget: adopting an individualist point of view tend to favour patents: an inventor is entitled to reward (forgetting other's entitlement for freedom of invention). If I didn't have that patent, then I wouldn't have done this much good (forgetting that others may have done it otherwise). I know you didn't make this error, —Cf your §2—, but others do —for instance by taking your §1 out of context.)
I think the key takeaway is that the patent was retroactively extended; so the idea was already created. If the extension was necessary for funding and development, then a case can be made for granting temporary monopolies for anything regardless of the inventor (the idea already exists, so we should find the best implementer) - I think most would be against this idea.
Well, it stinks when you try to do something good and then catch a bunch of crap from people doing nothing, for not doing enough. So I don't want to do that.
But here's the thing - while this is good, the Intellectual Ventures of the world are actually wining and dining congressional aides and maxing donations. The financial industry got that patent carve-out a couple months ago that only applied to their industry. Until tech people get into the game, full-court press in DC and get actual job creators and moneymen in front of congressmen telling them that this stuff destroys jobs, they're not going to be represented.
You're not obligated to go on a personal crusade on this, and it sounds like a big hellish pain in the ass that would be a net loss for you personally and for YC by entangling you in politics. But until someone does, people in favor of patent reform are going to be unilaterally disarmed.
Maybe some sort of industry lobbying group for small tech biz that's not beholden to the big corps would be effective. A small amount kicked in by some of the major angels/VCs, and a limited lobbying mission of "stuff that everyone can agree to" which would certainly include abusive patent use.
That is the precise point of this initiative. Since lobbying for reform is timely, not going to make immediate change, and ultimately becomes a war of finances, the goal is to bring consumer awareness to the issue and create an environment where being on the pledge list is a big benefit to companies. It also puts those companies together on a team for this issue, which may lead to other "rules" that help force the change. Since it's not targeting patent trolls with actual products, you may even end up in a situation where large companies on the list decide that the only people they litigate against are those who are inhibiting innovation from others. For example, Bump won't license its patent to any companies not on the list. Basically, stop blocking innovation, or be blocked.
Yeah, since posting I realized that a list of pledgees is exactly the kind of list you want if you're going to form an industry association based around this issue. So good on pg in general.
i cant see a place to take the pledge so i added one on my own site http://littlebiggy.org/2005087. its still in alpha but it oughta work since you need to use your real name.
Part of the problem is that the negative impact of patent trolling doesn't yet register with the general public as an issue comparable to the frivolous med mal suits that necessitated tort reform. We need Michael Moore to produce a documentary on the harms caused by patent trolling (and bad patents), so that the full extent of the problem is presented to a wider audience.
> The topic seems to be one of those where on average the strength of people's opinions is inversely proportional to how much they've studied the problem.
For me, at least, the experience has been the reverse: I used to think the patent system was basically sound with only a few problems, but the more I've read on the history of patents, the more I've realised how broken it is, and has been for a long time.
For example, in the 18th century James Watt used his patent to stymie development of high-pressure steam engines.
In the 19th century, there was costly patent litigation over barbed wire, which benefitted the economy not one jot. Nor did the sewing machine patent thicket and ensuing litigation benefit the consumer.
A patent is supposed to teach people how to make an invention. But in the 20th century, Frank Whittle wrote his patent application for the turbojet several years before he had actually made one. His patent couldn't teach people how to make a turbojet, because at the time of writing he didn't know himself! -- it took Whittle several years of hard engineering before he built a working jet engine.
I think it's implied that the companies making this pledge will tend to be the huge corporations with massive brand recognition and existing user bases that ought to be able to dominate the market unless the 25 person shop has found a way of significantly improving the product offering. In which case most of us would prefer to see acquisitions over litigation.
> Big companies that use patents as a revenue stream (MSFT, IBM, etc.) typically bide their time and bring a patent lawsuit once a new company is established and there is blood to drain.
Doesn't the legal doctrine of laches prevent this?
Patent trolling big companies is just as unethical as trolling small ones.
PG: Red Hat, a multibillion dollar business, already has a working patent pledge - they won't use patents except defensively against people who attack them first. Copy that and use it.
I disagree. Patent trolls often just aim for a comparatively meager payout for licensing from their victims, who are usually too small to be able to afford the much more expensive option of taking it to court (and perhaps winning). Therefore a loophole in justice.
Not necessarily a meager payout. They don't want to kill the host (as per PG's essay), but non-practicing entities are much more likely to swing for the fences - asking for, litigating toward, and receiving large damage awards. They have less to lose.
I understand that some victims are weaker than others. However that doesn't change the ethics of the matter: stealing from a poor old lady is as ethically bad as stealing from a rich kid. The impact may be greater, but the ethics are just as dubious.
Consider ID discovering a mathematical transform. If Creative signed up to PG's meaningless patent promise, they'd still be asking ID Software for protection money as they have more than 25 employees.
My personal pledge is that as a programmer I refuse to work for any company that goes on the attack with software patents, this obviously includes Apple and Microsoft. I also refuse to participate if asked by my company to help create a patent, I am willing to be fired over this.
Since good programmers are a scarce resource if enough of us took this pledge it could really start having an effect.
When I was writing software patents for a living, programmers hated talking to me. They considered it a total waste of time at best. That's part of the reason I decided to switch sides and do a startup instead ;)
The problem with that worldview, is that Patents play a number of very important defensive and value creating roles in a small company, that is not related, whatsoever, to their use in an offensive (in both senses of the word) manner.
"We do advise the companies we fund to apply for patents, but not so they can sue competitors. Successful startups either get bought or grow into big companies. If a startup wants to grow into a big company, they should apply for patents to build up the patent portfolio they'll need to maintain an armed truce with other big companies. If they want to get bought, they should apply for patents because patents are part of the mating dance with acquirers."
That's the type of pragmatic individual reasoning that collectively just perpetuates and reinforces the current system. If we can't rely on government then the only way to change it is by being unreasonable.
The thing about being a leader, or a bellwether, is that you need to walk outside the mainstream, be just a little bit crazier than your colleagues, but not so crazy our out of the mainstream that you leave everyone behind.
PG's position is clearly not in the mainstream of the business world (though, one might argue that, if anything, he's more conservative than his hacker audience) - but he's close enough to their interests that he may encourage followers. Or not. That's always the risk of being a leader - you may step out, and people may not follow.
He does speak to their interests which is, "If you want to hire great developers, you should align your corporate ethos with the best and brightest that you want to attract" - and, in general, large companies virtually never, ever, sue _small_ companies for _software_ patents - so they have little to risk.
Software inventions, no matter how original, should never be patented. Where would we be if someone had patented web-crawling or unit testing or hyperlinks or MVC? Plenty of people are getting rich, tons of innovation is going on. What are software patents supposed to be good for?
> Software inventions, no matter how original, should never be patented
Why not?
The whole idea of patents is to prevent trivial duplication of results that were achieved through considerable research expense. I don't really see much difference between spending a lot of time on getting a mechanical design right or developing and testing a drug or researching a sophisticated algorithm.
Paul, there is a major oversight here. The site http://thepatentpledge.org/ doesn't even have a contact form. Also, you may want to make the links nofollow.
I think this is a great idea and that it will prevent pledging companies from engaging in patent abuse, but I wonder how many large companies will bother to sign up for it. I imagine that if Apple/Google/MS all just take a pass that they won't catch much flak for doing so.
I wonder though if we could make the whole thing more effective by also adding an underlying threat to the pledge:
That any company, patent pledging or not, who violates the <25 rule will have their talent actively recruited away by those companies that have pledged.
One line from this essay has me a little worried...
"A clumsy parasite may occasionally kill the host, but that's not its goal"
This came up in a previous discussion on HN where I made essentially the same point. As someone pointed out in response, a parasite can get away with killing off the host as long as there's somewhere else to go next. In fact, a parasite could wipe out an entire species as long as it can make the jump to something more resilient.
it was just a short aside, but here's a link the the thread...
203 comments
[ 3.0 ms ] story [ 280 ms ] threadDoes anyone have stats on who is doing the most damage to early companies? Given the press, it's easy to think that trolls are the biggest offenders by an order of magnitude. Is there data that suggests otherwise?
It only takes one obscure archival services contractor to sue Google out of existence in 1999.
But operating a single-founder startup myself, I am alot more worried about the patent trolls than big companies right now.
I'm no lawyer - I have to ask the logical question - does publicly stating this pledge bust any opportunity to double back (i.e. it is more legally binding than just a pledge?)
Incidentally, for those who are curious, the three patents used against Fototime (as well as more established companies like Flickr) can be found at http://www.google.com/patents?id=uiAPAAAAEBAJ, http://www.google.com/patents?id=0J8DAAAAEBAJ, and http://www.google.com/patents?id=-QEWAAAAEBAJ. The first two actually feel substantial and specific (probably too specific for Fototime to have been infringing); the third one is just a simple database schema and a flowchart, which is madness (though it, too, is so specific that it's unlikely that anyone was actually infringing).
I see 2 problems currently.
1. Microsoft suing Android makers, and other similar examples, where large companies burn billions of dollars of our economy over something pointless.
2. Patent trolls like Intellectual Ventures and their shell companies suing startups.
How does this solve either of these problems? Who really needs this?
Patent trolls worry me the most. I'd like to see a pledge around having the tech community pool together against patent trolls. Not sure how you'd do it, but I'd like to see that.
How?
This problem needs to be fixed at its root, with a different law.
Since this pledge would only address this issue of secondary importance, which seems a lot less salient to the public, I can't imagine it getting off the ground.
I will quote myself from [ http://news.ycombinator.com/item?id=2855835 ] here for another solution, one that actually can _easily_ go through government (except for the intense lobbying against it by whoever enjoys the current patent regime); you can read there for some discussion if it is interesting. Quoth myself (with minor editing):
Intellectual "Property Tax". Have everyone declare the value of their intellectual "property" (patents, copyrights, trademarks) - each and every item, for that year, on their tax return, and have them pay 1% of the value as "IP tax", per year.
Clarification: you can set a different value every year. The value may drop to zero because a competitor's patent solves the problem better; or it may go up because it becomes essential to something that becomes commonplace.
That amount is what one pays for a compulsory license or if successfully sued, and up to 3 times that for willful infringement, per year -- and no more. (But of course, a patent owner can always negotiate a lower payment, as is done with music recordings that have compulsory license agreements)
All of a sudden, everyone has an incentive to state a reasonable value for their patent. Copyright catalogs that are not being published (old music recordings, old books, old movies) would be assigned 0 value by copyright holder, to avoid tax - which means anyone can freely make a copy. If they believe -- at the end of the year -- that someone is making a profit at their expense, they can set the value as high as they want at the end of that year, pay the tax, and sue the profiteer.
Simple, elegant, and coffer filling.
edit: put missing link
edit: added clarification about setting value each year anew.
As a real-life example of people gaming these types of taxes, people buy expensive (>100K USD) cars in europe directly (i.e. taking delivery in stuttgart) to save on the sales tax: driving it around for a bit ensures that the car is technically "used" when brought back here, circumventing new car taxes.
Again, you have to pay the tax _before_ you can sue, and then you can only sue for the value reflected by the tax for that year, no more (and you are still subject to invalidation, etc).
Of course, the easy work around to this is the claim that it's value increased suddenly when the patent became more useful; you start paying more tax when you start suing others.
Considering that low maintenance costs don't prevent domain squatting, I'm not sure this would be an effective tactic to prevent lots of silly patents.
a) reduce government intake in the form of patent registration and maintenance fees b) get all the lobbyists employed by current patent holders (who spend billions of dollars on those patents) working against you.
Why on earth would congress do something like that? So that 0.05% of the population that actually has any idea how bad it is would rejoice? (If your answer is "to improve american competitiveness" or something like that -- well, you have no idea how congress works)
However, if you can get a new spending source, you just have the lobbyist problem to take care of.
My proposal is that it is set retroactively one year back: when you file your 2011 taxes (on April 2012 at the latest), you would assign a value for it for 2011, and pay it at that point.
If you sue (or even threat to sue) someone in 2011 before paying the tax, that is an admission of tax liability - if you declare less at the end of the year, that's grounds for the IRS to audit you and assume your highest estimate during the year (as shown in the threat to sue) sets your tax, and anything else is automatic assumption of tax evasion.
A future-looking only assessment is also possible; but in that case, you have to be able to set the value (and pay the tax) on a monthly basis -- otherwise, it calls for anyone who is willfully infringing to wait until 1-jan, ask for the IP value, and if it is less than they expected to pay, have a year of "free" use.
My proposal is an attempt to sort-of balance the existing state of affairs -- if at the end of the year, you didn't manage to license it to anyone, you shouldn't be taxed for what you thought it is worth.
In other words, doesn't this just reduce to a 1% tax on patent damages?
(Even if we think we'll only recover on average p% of our claims, we still only pay 1% of X in taxes to get p% of X in damages. Assuming p% is closer to 50% than 1%, not much changes.)
However, it does chill the chilling effect - right now, e.g. Microsoft has ~20,000 patents, and they can sue anyone for anything. This way, they will actively have to choose strategically what they want to enforce.
Finally, and I think I haven't made this clear enough, I don't think my proposal is a GOOD patent system in any way. It is bad. But it is not as bad as the existing system, and (I suspect) it has a much better chance of passing than a complete reform.
Furthermore, it stops some of the (currently legal) tax loops that involve passing patents among shell corporations.
The basic idea is to align value with cost. Any system in which the value of the patent is potentially infinite (as in the current system) is materially broken, and it is just a matter of time until someone finds a way to game it (non practicing patent trolls). Any sane system must have cost (of patent, property, copyright, etc) rise at least linearly with value.
I find the non-retroactive possibility much more interesting.
And, it is more than just a tax on patent infringement damages - because you have to pay the tax upfront, which makes a big difference in cashflow.
eolas's $600M verdict against Microsoft, and i4i's $400M verdict would both require posting $6M / $4M per year respectively in never-to-be-recouped taxes, which is a huge sum for the companies in question, and would probably have made them not sue.
Furthermore, it involves the IRS in valuations, which is a good thing (because it is much harder to mess with the IRS) - manipulation of the value of a patent would have to be explained, and not arbitrary to avoid taxes.
Anyway, we could all dream - it is the lobbyists that write the laws, and the patent holders have thousands of them in DC.
Incidently, patent trolls are not a new phenomenon: Elias Howe was on in the 19th century. http://volokh.com/archives/archive_2009_04_26-2009_05_02.sht...
A good patent system would have the following properties: * Investment and risk taking in R&D and sharing the results is rewarded above the costs of that R&D, so there is no disincentive to share with competitors. * There is no disincentive to using and building on ideas shared by others.
I think the best system is a 'tax'-like system - alongside the existing patent system, businesses can elect to pay 10% of their revenues to a pool for immunity from all patents. Businesses using / reselling patented manufactured products are still liable even if the manufacturer pays the 10%, unless they also pay the 10%. Patent filers tell the pool the actual costs of R&D (and may be audited to prove it). The pool is distributed based to patent filers based on R&D costs (and maybe a per-industry factor for risk) - it might be better to somehow collect some rough metric of value of the each patent to their business from patent users, although there are risks of the system being gamed then.
This system would achieve the desired properties at 10% of all industry-wide research being spent on R&D - it would favour not disclosing R&D worth more than 10%, and if less than 10% was spent, it would be stacked in favour of companies doing R&D.
I get 'narrower' but what does 'but open source' mean here?
A court might well hold a company to such a pledge, on a theory of "equitable estoppel." This type of defense to an infringement charge is always highly fact-specific; here's an example of a case in which the defense succeeded:
A patent owner accused a manufacturer of eyeglass frames---which it had previously sued for infringement---of infringing other patents. After back-and-forth correspondence---in which the manufacturer denied infringement---the patent owner went silent for three years. In the meantime, the eyeglass manufacturer expanded its marketing efforts for the products in question.
The trial court held that the manufacturer was not liable for infringement, on grounds that the patent owner's actions, in view of all the circumstances, had misled the manufacturer into thinking it would not be sued. The appeals court found no error in this holding [1]; it explained that:
"In the context of patent infringement, the three elements of equitable estoppel that must be established are:
(1) the patentee, through misleading conduct, led the alleged infringer to reasonably believe that the patentee did not intend to enforce its patent against the infringer;
(2) the alleged infringer relied on that conduct; and
(3) due to its reliance, the alleged infringer would be materially prejudiced if the patentee were permitted to proceed with its charge of infringement."
[1] Aspex Eyewear, Inc. v. Clariti Eyewear, Inc., 605 F. 3d 1305 (Fed. Cir. 2010) (affirming summary judgment in favor of accused infringer), http://www.cafc.uscourts.gov/images/stories/opinions-orders/...
---
These companies have agreed to be the first to publicly renounce aggressive use of software patents on small companies. Please join them!
A Thinking Ape, Airbnb, Bump, CarWoo, DailyBooth, Disqus, DotCloud, Greplin, Hipmunk, Justin.tv, Loopt, Songkick, Stripe, Weebly, Wepay
----
I think the whole YC gang is going to promote this aggressively, which means a strong network effect. Remains to been seen what happens outside this network.
http://englishplus.com/grammar/00000214.htm
Big companies that use patents as a revenue stream (MSFT, IBM, etc.) typically bide their time and bring a patent lawsuit once a new company is established and there is blood to drain. It's the threat of such a lawsuit in the future that can negatively impact investment in a startup, as the right collection of patents could conceivably capture much of the economic surplus of a new venture. Alternatively, a big company might use the threat of a patent lawsuit, now or in the future, to push a young company to agree to an early acquisition.
The pledge doesn't seem to have much impact on these scenarios, even if a big company were to follow it rigorously.
Most of us, I believe, would prefer to see companies make a stronger commitment: "No first use of software patents" [period]. Google hasn't made this pledge, but to the best of my knowledge, they've acted in this way so far. It does seem in line with "don't be evil."
That said, I think I see what PG is going for here. He wants companies to make a pledge that, at a minimum, allows a new product or service to be tested on the market. That way, if it gathers traction, it will attract investment despite the threat of patents, and the new company will be able to mount a reasonable defense.
Perhaps more importantly, though, by allowing the product to succeed first, even in a modest way, it makes the offensive use of patents worse PR for the big company. Killing a successful product with patents is no longer an abstract issue. It takes away from customers and the market something very real.
I don't think PG agrees with this. He seems to be much less anti-patent than the general HN (commenting) community.
Good idea, but it can't work. If the PTO is willing to grant a patent on a software idea, someone is going to get the patent. If Google stands down, some troll will get it instead, to the detriment of society.
If you have offensive capacity you will be bound to use it.
Guns don't kill people, People kill people. Sure, but guns make it much easier, and by design, have the capacity for landing in the wrong hands.
Are you suggesting people who own guns should kill someone from time to time so nobody will fuck with them?
Over 50 million households in the US own 200 million or more guns, collectively. If these guns were usually used this way, the US would be depopulated by gun violence in a decade or two.
I mean, the whole point of a patent is to be able to sue people who are competing with you. How can that ever be defensive?
If IBM is not deterred by Microsoft's patents, why would they be deterred by mine?
If your next question is that the existence of your invention would invalidate their patent due to prior art, that would take costly legal maneuvering to establish. Is there a shortcut around that legal maneuvering? Yes, patent your invention first.
How far ultimately would you want to push it? I honestly don't know yet. Patents may have some utility. It's a very complicated question. I've read a lot about the history of technology, and I can't say for sure whether things would have gone better or worse without patents. The topic seems to be one of those where on average the strength of people's opinions is inversely proportional to how much they've studied the problem.
Let me address your last sentence, though. Your statement presupposes a utilitarian moral system. That is, it assumes that a person's position is completely determined by which policy will result in the greatest growth in technology.
I've studied the history of patents as well, and I share your practical uncertainly over which policy inspires more growth. Patents may, in fact, grow certain types of technology faster while inhibiting growth in other areas. Since we need to weigh the harm caused by the lack of things that never existed against the benefits brought by things that do, it's probably impossible to do a quantitative analysis that is fair to both sides.
However, I and many others oppose patents on deontological grounds. That is, even if the existence of a patent monopoly produces some technological benefits as compared to a non-monopolistic system, we would choose freedom from monopoly.
John Carmack may have stated the essence of this position better than anyone since Jefferson [1]:
"The idea that I can be presented with a problem, set out to logically solve it with the tools at hand, and wind up with a program that could not be legally used because someone else followed the same logical steps some years ago and filed for a patent on it is horrifying." -- John Carmack
[1] "If nature has made any one thing less susceptible than all others of exclusive property, it is the action of the thinking power called an idea, which an individual may exclusively possess as long as he keeps it to himself; but the moment it is divulged, it forces itself into the possession of every one, and the receiver cannot dispossess himself of it. Its peculiar character, too, is that no one possesses the less, because every other possesses the whole of it. He who receives an idea from me, receives instruction himself without lessening mine; as he who lights his taper at mine, receives light without darkening me." -- Thomas Jefferson
But once established they definitely used patents against smaller competitors. So it is hard to say for sure even in this case whether patents were a net win. They probably were though. The patent probably caused modern separate-condenser steam engines to happen at least a few years earlier.
(I recommend the story of Bolton & Watt to present day founders. It is remarkable how little has changed.)
One reason is independent discovery: most innovations tend to be discovered independently and nearly simultaneously. Like, when "science is ready", innovations just pop out of several people's mind relatively quickly. (See also Eben Moglen's folk theory of the internet and innovation: "wrap the internet around a planet, spin the planet, and ideas flow out of the network".)
If we accept that, the only thing that could have significantly slowed the development of the modern separate-condenser steam engine down is money. Like, no single one investor is willing to risk making the first step, by fear of being eaten by copycats. If this is so, then we have a tragedy of the commons.
Risk aversion in this case is indeed a problem. I think it is mitigated however by (i) the first mover advantage, and (ii) variations of risk-aversion among individuals (combined with independent discovery, you get a winner-takes-all scenario).
Ah, before I forget: adopting an individualist point of view tend to favour patents: an inventor is entitled to reward (forgetting other's entitlement for freedom of invention). If I didn't have that patent, then I wouldn't have done this much good (forgetting that others may have done it otherwise). I know you didn't make this error, —Cf your §2—, but others do —for instance by taking your §1 out of context.)
[1]: http://www.dklevine.com/general/intellectual/against.htm
I think the key takeaway is that the patent was retroactively extended; so the idea was already created. If the extension was necessary for funding and development, then a case can be made for granting temporary monopolies for anything regardless of the inventor (the idea already exists, so we should find the best implementer) - I think most would be against this idea.
But here's the thing - while this is good, the Intellectual Ventures of the world are actually wining and dining congressional aides and maxing donations. The financial industry got that patent carve-out a couple months ago that only applied to their industry. Until tech people get into the game, full-court press in DC and get actual job creators and moneymen in front of congressmen telling them that this stuff destroys jobs, they're not going to be represented.
You're not obligated to go on a personal crusade on this, and it sounds like a big hellish pain in the ass that would be a net loss for you personally and for YC by entangling you in politics. But until someone does, people in favor of patent reform are going to be unilaterally disarmed.
Maybe some sort of industry lobbying group for small tech biz that's not beholden to the big corps would be effective. A small amount kicked in by some of the major angels/VCs, and a limited lobbying mission of "stuff that everyone can agree to" which would certainly include abusive patent use.
For me, at least, the experience has been the reverse: I used to think the patent system was basically sound with only a few problems, but the more I've read on the history of patents, the more I've realised how broken it is, and has been for a long time.
For example, in the 18th century James Watt used his patent to stymie development of high-pressure steam engines.
In the 19th century, there was costly patent litigation over barbed wire, which benefitted the economy not one jot. Nor did the sewing machine patent thicket and ensuing litigation benefit the consumer.
A patent is supposed to teach people how to make an invention. But in the 20th century, Frank Whittle wrote his patent application for the turbojet several years before he had actually made one. His patent couldn't teach people how to make a turbojet, because at the time of writing he didn't know himself! -- it took Whittle several years of hard engineering before he built a working jet engine.
http://thinkprogress.org/yglesias/2011/08/31/309483/patents-...
What's poorly thought out about this pledge is that a shop of far less than 25 people could easily clones of web and mobile apps at scale.
Doesn't the legal doctrine of laches prevent this?
PG: Red Hat, a multibillion dollar business, already has a working patent pledge - they won't use patents except defensively against people who attack them first. Copy that and use it.
Consider ID discovering a mathematical transform. If Creative signed up to PG's meaningless patent promise, they'd still be asking ID Software for protection money as they have more than 25 employees.
Since good programmers are a scarce resource if enough of us took this pledge it could really start having an effect.
See: http://paulgraham.com/softwarepatents.html
In particular:
"We do advise the companies we fund to apply for patents, but not so they can sue competitors. Successful startups either get bought or grow into big companies. If a startup wants to grow into a big company, they should apply for patents to build up the patent portfolio they'll need to maintain an armed truce with other big companies. If they want to get bought, they should apply for patents because patents are part of the mating dance with acquirers."
PG's position is clearly not in the mainstream of the business world (though, one might argue that, if anything, he's more conservative than his hacker audience) - but he's close enough to their interests that he may encourage followers. Or not. That's always the risk of being a leader - you may step out, and people may not follow.
He does speak to their interests which is, "If you want to hire great developers, you should align your corporate ethos with the best and brightest that you want to attract" - and, in general, large companies virtually never, ever, sue _small_ companies for _software_ patents - so they have little to risk.
What if it covers an actual invention?
What if it is meant to be used defensively, i.e. not for the litigation, but rather to discourage other companies from suing based on their patents?
Why not?
The whole idea of patents is to prevent trivial duplication of results that were achieved through considerable research expense. I don't really see much difference between spending a lot of time on getting a mechanical design right or developing and testing a drug or researching a sophisticated algorithm.
Paul, there is a major oversight here. The site http://thepatentpledge.org/ doesn't even have a contact form. Also, you may want to make the links nofollow.
I wonder though if we could make the whole thing more effective by also adding an underlying threat to the pledge:
That any company, patent pledging or not, who violates the <25 rule will have their talent actively recruited away by those companies that have pledged.
"A clumsy parasite may occasionally kill the host, but that's not its goal"
This came up in a previous discussion on HN where I made essentially the same point. As someone pointed out in response, a parasite can get away with killing off the host as long as there's somewhere else to go next. In fact, a parasite could wipe out an entire species as long as it can make the jump to something more resilient.
it was just a short aside, but here's a link the the thread...
http://news.ycombinator.com/item?id=2760148