I wonder if Twitter will retroactively apply this to the "pull to refresh" patent they acquired when they purchased the Tweetie and turned it into the official Twitter client:
I don't usually upvote jokes on HN, but when I do, I provide some actual content in the same post.
OT (obviously): This is an example of a facet of humor I find rather difficult to explain. Most comedians pull this trick once or twice per show - the good ones, more often. You take the current topic of conversation, and link it (in a somewhat skewed way) with some previous joke. Somehow, this has an amplifying effect on the old joke, which continues to get stronger as you do it a few more times (to the same target). Anybody know of an explanation / name for this?
This made me laugh because I just realized that the "no semicolons in my Javascript" people are just modern day versions of Kurt Vonnegut: http://www.goodreads.com/quotes/show/17178
It's a very interesting move. I guess the calculation (apart from real benevolence, which I don't mean to undercut) would be that the good will and good press, affection from idealistic engineers and the resulting hiring and retention advantage, and the effect on the rest of the patent landscape (over which they have no direct control) is worth more than the patents themselves.
The thing is, how that plays out depends on whether or not other companies agree to change their behavior. If it's true, as another comment says, that they will allow Apple to use the pull-to-refresh patent, for example, that's some clear leverage lost. If no one else makes similar moves, won't Twitter be at a disadvantage in the context of the "patent wars?" Will the other advantages I listed above still be worth it if the patent landscape doesn't change? Does this effect the value (especially as a potential acquisition target) of Twitter? I don't know enough about their portfolio of patents to know how valuable there patents might be.
In any case, it's at least something new-feeling in the story of software patents. Hopefully it will have some positive effects.
EDIT: Also, hearing all this talk of "defensive" patent use makes me think about Intellectual Ventures; does that sort of thing count as defensive under this plan?
One possibly self-interested effect could be that it may make engineers more willing to go along with applying for patents. A lot of engineers feel very squeamish about software patents, and try to avoid applying for them or letting themselves get coerced into applying for them. A commitment by Twitter that they won't use them offensively w/o the engineer's permission may make recalcitrant engineers more likely to submit/sign the relevant documents.
I said specifically I'm not intending to undercut their motives. Even if that's true, I think it's worth understanding how it impacts then strategically.
There also a question of the legal aspect of letting other companies infringe for a while and then trying to start litigation against them (e.g if they broke condition 2b).
How would courts view this? Or does it not matter?
I don't think Apple could use the pull to refresh patent without fear since they would fall into the 'sued people offensively in the last 10 years' bucket referenced, no?
'The company will not use the patents in offensive litigation without the permission of the inventors.'
Honestly how hard is it to go up to someone and say "Let us troll with this patent and we will cut you in on 20% of the profit". I know quite a few people who would do that deal just to pay for their kids college in the future. I don't see this changing anything except for inventors now getting a cut of the deal.
"If Assignee needs to assert any of the Patent claims against any entity for other than a Defensive Purpose, Assignees must obtain prior written permission from all of the Inventors without additional consideration or threat."
At which point whoever twitter was suing could claim that they had violated their own agreement so they don't have grounds to sue.
The agreement would likely be private, so as long as both parties keep their mouth shut about it (which they're incentivized to do), there is little risk for them.
Personally, I see this as a great potential bonus source of revenue for the inventor down the road, selling the right to use their patent offensively.
Most companies won't do it though, since it restricts the potential market for their patent IP, which decreases the value of their assets on their balance sheet.
The common law principle of privity of contract (see http://en.wikipedia.org/wiki/Privity_of_contract) means that contracts can only be enforced by the parties to that contract (although some jurisdictions have changed this with legislation). A defendant in a lawsuit against Twitter cannot assert an agreement between a third party and Twitter against Twitter.
Well, if this sort of agreement becomes widespread, it would effectively opt them in to being sued by other companies who have patents assigned under a similar contract. Also, it would be in breach of any other IPA contracts they've made with other inventors.
It's a neat concept. The more widespread these contracts become, the more litigation you expose yourself to for using your patents offensively.
No, it explains how the purchaser of the patent can get the inventor to go along with their offensive patent lawsuit, which is necessary BECAUSE the inventor maintains veto power. If that weren't the case, the inventor would have no say in the matter.
But what's the point of the "no consideration" bit? What's the point of two parties agreeing that they won't make a new deal later, when they can anyway?
They just have to say if pressed that there was no threat or other consideration, and that they agreed to let the other party use their patent offensively. Then collect their totally unrelated consulting fee.
Edit: Now I think I was wrong in this reply. I did not realize that paragraph 4 of the agreement gives the inventor the right to license the patent to another party to enforce the agreement. However, the broad language of the second paragraph still gives the assignee a lot of wiggle room to argue that they are not breaking any promises made.
---
Original comment:
I mentioned this in another thread, but it is actually simpler than that. As I read it, the twitter agreement gives all rights in the patent to the assignee, just like any other assignment, but it adds on the clause that the assignee agrees to get permission from the inventor if they want to sue offensively with the patent. In the future, if the assignee (whoever it is at that time) decides to sue offensively without the permission of the inventor, the assignee can (and indeed they have the right to, as the holder of all the rights in the patent). The inventor would then have a cause of action for breach of contract against the assignee, but who knows what that would amount to.
Basically, I doubt that a court would read the agreement in a way that prevents the future patent assignee from using the patent offensively, even without the agreement of the inventor.
Employers and investors who have a problem with this are going to have to weigh the cost of these restrictions against the cost of turning away high-value engineers who can afford to be selective about where they work.
During the hiring process, lots of employers are already giving candidates verbal assurances that their patents will only be enforced defensively. Now, software engineers have a practical way to demand that they make these assurances binding. Employers who fail to do so will have a much tougher time explaining themselves.
I'm probably going to start requiring this of any employers that I work for in the future.
But for hardware or biotech startups? Licensing patents is oftentimes a startup's business model. And licensing doesn't work if people can use technology without paying (which would be the case with a purely defensive patent).
I'd say defensive patents are pretty useless for a startup. The idea is to build a patent portfolio that's so large and diverse that the day some random company sues you, you just happen to have a patent that they themselves are infringing on. You also need the financial resources to pull off that counter threat. Doesn't sound like lean startup to me...
A much cheaper defensive measure is to publish your "inventions". There are companies that help you with that, e.g. ip.com.
It's a balance between the benefit of the recruiting/retention edge for making a commitment to your engineering team, vs. the loss in patent value for an acquisition or to stop someone who's really ripping you off later (the latter obviously being a huge gray area).
They can sue, they just need the inventor's permission. In that case, if the inventor is getting their hard work ripped off, they would probably agree to sue.
'An assertion of claims of the Patents shall be considered for a “Defensive Purpose” if the claims are asserted .... (c) otherwise to deter a patent litigation threat against Assignee or Assignee’s users, affiliates, customers, suppliers, or distributors.'
I agree that the language is pretty broad (particularly the idea that ayou can use it against someone who has bought action against anyone, not just to defend yourself) but I don't think you could get away with getting your retaliation in early the way it's written.
Glad to see Twitter doing this, it's definitely a step in the right direction. In my opinion what this really needs is the support from another well-known company. If a company like Facebook announced next week that they were adopting the Innovators Patent Agreement, I think that could get the ball rolling and we would start to see many other companies join the cause.
No, I think it means that the patents will never end up in the hands of people that just plan to use them as weapons to extort other companies vs. their actual useful purpose, which is to protect innovators from having their hard work ripped off.
Twitter can still use the patents offensively if the engineer whose name is on the patent agrees. They're just committing not to do so against the will of the inventor.
> If Assignee needs to assert any of the Patent claims against any entity for other than a Defensive Purpose, Assignees must obtain prior written permission from all of the Inventors without additional consideration or threat.
Doesn't this nullify everything else? I would love to believe we live in a world where inventors would stop their patents being used wrongly if they had the power. But I honestly believe these are a very tiny minority. Won't most inventors just agree with the company they work for anyway? There doesn't need to be any explicit threat, but just working for a company probably means that you'll do what's necessary to keep your bosses happy. Won't companies just get a signed consensus from inventors and keep suing anyway?
In my experience, the inventors often are working somewhere else by time a patent gets litigated. In particular, patent trolls usually acquire patents without also acquiring the labs, inventors, etc. that created them.
They wouldn't be violating under the terms of this agreement - they're granted a perpetual license.
But unless I'm misunderstanding you I believe you're missing the point. This is pretty much written with the assumption that engineers wouldn't grant permission to a first strike action whoever it were against and that that's fine.
It's turning patents into tools weapons that can only be used for retaliation.
> 2(b) against an Entity that has filed, maintained, or voluntarily participated in a patent infringement lawsuit against another in the past ten years, so long as the Entity has not instituted the patent infringement lawsuit defensively in response to a patent litigation threat against the Entity
So it counts as a defensive purpose if they're suing anyone who's filed a patent infringement case in the last decade? That's pretty friggin' broad.
That's hugely broad. Basically they won't sue startups unless the startups sue them first.
I love this approach, but it should also be pointed out that this is a very easy approach for a patent-poor company to take. If you don't have many patents, you aren't going to sue other companies who do have patents and you'd sure like to look like the good guy if you infringe on their patents.
Actually, it appears to mean that they won't sue startups unless the startups sued _someone_ about a patent in the last 10 years, unless that suit also qualified as a Defensive suit, which means that the suit is against _another_ company that hasn't sued anyone about a patent in the last 10 years, unless _that_ suit was a Defensive action.
frankly i think the recursion is both deliberate and good--if it's written properly (i haven't reasoned through the fine points completely), it partitions the world into two groups: those who've never started a patent fight (against anyone), and those who have. members of the first group can't sue each other without being reclassified into the second group, at which point everyone left in the first group can sue them without being reclassified.
People got to get off their utopia world-view thinking companies should be giving away their core-business' secret sauce - the essence that keeps them in business.
There is a clear distinction between core-business-specific platform and the technology their platform is built on which facilitates it (i.e. OSS projects/libraries).
Twitter is built on open-source, they enhance & contribute to existing OSS projects & open source key technologies their platform is built on:
I know. All that is nice, certainly better than many companies. But that doesn't make them role models. I'd like to think we can shoot a little higher in finding role models than companies that utterly lock down their platform but throw us a few scraps for free.
They are perfect Role models since their a successful company who employee great talent who use and contribute back to open source. If something doesn't exist they'll build it and open source key parts of their technology platform and are more than open about what they use and how they're able to scale their incredible load:
Giving away your core-business is corporate suicide, they would be a shadow of themselves if 100 clones surfaced overnight offering the exact same service - and by extension be unable to contribute back to OSS as much as they have.
Your other reply is "dead" for some reason, so I'll reply to it here.
You can build services that interoperate with Gmail over the same protocol. I can run my own email server, even (can you run your own Twitter server?). That interoperability is how Gmail came to exist, disrupting the entrenched hosted email industry and providing a massively better service to its users. If Hotmail had 99% market share and only allowed email to be sent to other Hotmail addresses, then Gmail never would have happened. Yet Gmail survives by providing a great service.
Now compare with Twitter. An open equivalent standard is StatusNet, which powers Identica and other services. But none of them can work with Twitter, because Twitter is a walled garden with 99% of the users. Thus, if Twitter starts to provide a crappy service, it will be much harder for competitors to instigate a Gmail-like disruption because they won't be able to interoperate with existing Twitter users.
Twitter has an API, so you can syndicate your Foo feed through your Twitter account, and download your incoming Tweets to publish elsewhere. How is that different from SMTP, from the accessibility angle?
Yahoo Mail, Hotmail, and my own personal mail server can all interoperate with Gmail as first class citizens. I can easily send an email from Gmail to someone using Hotmail. That's how Gmail was able to easily get a foothold in the market, because Gmail users could email users of other email providers.
Identica, other StatusNet installations, and other services cannot fully interoperate with Twitter as first class citizens. How do I send a Tweet from Twitter to someone using Identica, for instance?
Gmail isn't open source - Open APIs/Open Data is not Open Source. People use the Gmail service they allow external integration to broaden Gmails reach/ecosystem - this makes the Gmail service more attractive.
The fact is they'll never Open source the Gmail client which is their secret-sauce/USP that others are looking to replicate. Like twitters application platform, that is not something anyone should ever expect to be given away - its bad business.
Twitter isn't open source either. In that way they are similar to Gmail. Twitter is also headquartered in the US, same as Google. There are innumerable other ways that they are similar.
There are also some major differences, such as the one I described about interoperability. Another difference: unlike Gmail, Twitter's "secret sauce" isn't their website. It's their walled garden social network. Twitter could open source their whole website's code, and it wouldn't much help anyone build a Twitter competitor. StatusNet is already pretty nice.
Also, since we're veering way off topic here, note that I am not arguing that Google is "OSS's best role model", like you said about Twitter. I'm just using Gmail as an example to illustrate how Twitter's walled garden is different than how email works.
The word is federated. Email is federated. Google Wave was federated. Twitter is not federated. You can run your own email server and be a first class citizen in the network. You can't run your own Twitter server and be anything but a layer on top of the Twitter API.
More importantly, federation is inherently decentralized. If GMail goes down, your email server still runs and still inter-operates with the rest of the world.
Suicide? Gmail seems to survive fine. I'm glad people didn't think like you back when email was invented, or we'd still be stuck on our old 5 MB Hotmail accounts (or worse!). People who designed and built systems with their users in mind... those are much better role models than companies like Twitter who build proprietary walled gardens. And going as far as to say Twitter is "OSS's best role model" is just absurd.
I'm sure that's what he was talking about, if he wasn't being sarcastic. But there's a difference between releasing a bit of code on the periphery of your walled garden and being a "role model".
I would never ask a for-profit company to open source their core software if it doesn't meet their business model (sell support, etc.). Here's the thing though: I don't care one bit about how they process tweets internally.
However, the "bit of code on the periphery" that Twitter has released is far more interesting to me. Bootstrap lets me prototype web UIs much faster, for example, and who knows? Maybe some day I'll end up using their contributions to MySQL, Cassandra or Mahout.
So, if only for purely selfish reasons, I say kudos to them.
Kudos for the good things they've done, sure. And criticism for the bad things they've done. But "OSS's best role model"? Laughable. That was the post I replied to.
This is good. Personally I would rather in a world where every so called "invention" is immediately documented publicly so that from that point on there is prior art and nobody can claim a patent on it.
Tipping the balance of offensive to defensive patents will definitely also help.
This move strikes me as a truce agreement proposed by someone with no sword.
How many technology patents does Twitter even HAVE, not counting Loren Brichter's "pull-to-refresh" UI innovation?
Could a public company's leadership even adopt such a measure without getting successfully sued by its shareholders? They would be putting liabilities on the company's IP assets, decreasing their market value, which is not in the interest of shareholders.
Would it just fly under the radar? I guess the company would have to argue that this allows them to pursue and retain better talent, which might be an easy case...
To protect your idea/product using patents is OK but using them to stop innovations is a dark side. How to define the balance between using patents as weapons and using them to secure investments is the big question. If your innovative project uses something well-known, but adds the essence on top of that, do you have to pay a license for all the bricks you use?
Wow, I didn't know of Docracy [1] yet. It looks interesting, but doesn't seem to provide version control outside the web interface. From that perspective, GitHub might be a better place for this.
A nice idea, but many developers' employment contracts already state that they assign all inventions created during the course of their duties to the company, free and clear. How would this help such a situation?
So so I'm clear on this, when they say patents can only be used for defensive purposes, are they saying this will effectively nullify patent law in regard to this particular patent?
No company in their right mind should sign this. This throws the baby out with the bathwater. There are actually legitimate uses for patents and if you're working on something non-trivial your company might want to (and might even be obligated to in order to do right by shareholders) use the patent offensively.
Whatever patents in general have any legitimate uses is at best arguable (see http://levine.sscnet.ucla.edu/general/intellectual/against.h... ), but that software patents are a complete and utter disaster that doesn't benefit anyone other than trolls and lawyers has become quite clear by now.
http://appft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sec...