Is it legal to sell my own implementation of a paper?

If there's a paper on the internet about some new technique, is it legal to write my own implementation of it and sell it?

Example : There's a paper describing a new rendering technique. Source code is not available. I read the paper and implement it myself. Is it legal to sell the code / plugin I wrote?

 

It depends on whether or not there's a patent on it, I'd wager.

 

If they don't state that you cannot in that paper than it's OK, and if not then they'll likely only care if you become wildly successful selling your implementation.

 

You should not take any legal advice on a game development forum. If you really want to know the answer, you'll have to ask a lawyer who specializes in those things. Nothing anyone here can say will protect you from consequences should there be any. If you can't find or afford a lawyer, and you are really concerned about the legality of it, don't do it.

 

If you sign up with LegalZoom.com you get a full free month of legal consultation (at least I remember that being the case for us a few years back) and can schedule appointments with lawyers who practice in a whole bunch of areas, including this one.

You should check it out to see if they still have that offer. Like @drewradley said, you'll want to be getting this information from someone who knows it well.

 

Do you hire a lawyer each time before using any products with all those warnings on them? Or speak. Or do anything really because in a land ruled by laws every breath you take has potential legal repercussions.

If the paper itself doesn't state legal terms of the content of the paper than they are free to implement. They are not verbatim copying the text of the paper and claiming it as their research. Politely though and to save embarrassment the implementer should site the research paper(s) in their implementation.

 

This is completely untrue and I am amazed more people don't call you on your BS. Papers can easily be patented and those patents can and will be enforced.

 

Huh? The OP asked if it were legal to do something. I said to ask a lawyer to be sure. If they had been asking for medical advice I would have told them to see a doctor. If they had asked a tax question, I'd tell them to ask an accountant.

 

Ask the author of the paper.

 

Copyright protection does not apply, copyright only protects the actual written words of the paper, not the concept.

Trade marks don't apply, and would only apply if they coined a specific name for their implementation of the algorithm. In short you can't pretend to be the paper's author.

Trade secrets don't apply, as long as you got the paper from a legitimate published source. If the paper has been leaked from some internal source, you could still get in trouble here.

Patents do apply. You can patent an invention. If their is a patent pending on their invention, they should have disclosed that in the paper. But you can always do a patent search to see if any one has lodged a patent. Chances are they haven't, code is ridiculously difficult to patent and enforce because of the prior art restrictions. But if you want a more thorough legal defence, consult a lawyer who specialises in the area.

 

So very much this.

If you ever have any question about whether some technique in a paper is patented or protected in any other way, contact the authors and ask them directly. If they say it is patented then you need a license (and note that there are patent holders who let people use their methods without a fee - it's uncommon, but it does happen, especially in the OSS community). If they say it is not patented then you are probably free to use it. If they say it's patent pending then you've got a choice: you could license it from them as a preemptive way to protect yourself, or you could just use the technique and hope their patent application is denied, but in that case you are exposing yourself to liability if the patent is granted.

But asking directly is the best way to get an answer.

 

Why should they start mentioning patents? That would only make some research papers and articles even harder to read then they already are. It is the opposite, expected that the work resulted in one or more patents if it is a serious idea.

It is your own responsibility to check if you might intrude on any patents. The patent owner is required to defend the patent.

You don't patent code, you patent ideas that happen to also cover implementations in code. That is why there is sadly so many crazy patents that affect software It should also be noted that patent systems are different between countries and it is therefor important to know your market and if the you have the necessary tools to restrict which countries your implementation is sold in etc.

 

And get the permission in writing, then bring that to the lawyer.

 

Permission of the author doesn't matter, unless there is a patent. The default concept of a paper is that it is funded by the government, as a gift to the world. Sometimes there are patents, in which case you need a license, which is a legal document, not just verbal permission.

 

This also isn't true. Papers produced by universities, for instance, often get patented right out the gate by the university itself.

 

This thread should be stickied as a prime example of why you never seek legal advice in a forum.

 

Perhaps that is common, but that wasn't the case when I was a grad student at Berkeley.
Patents cost thousands of dollars to get, and not every paper is patentable.

 

I don't get this cottening to lawyers to interpret words that one themselves can interpret. Well..I do get it..but I am fully capable of parsing legalese and looking up case law. You do not need a lawyer, or attorney or barrister for a case like this. You simply need to read case law. We are talking about folks who can understand very intricate subjects in depth around here who can dive into a SIGGRAPH paper and implement it but need to definitely talk to a shyster of the BAR.. I don't buy it. Starve the lawyers. If we dumped all the lawyers in the ocean it would stop global warming.

 

And many, many more times they don't. For instance I saw a stat that only 250 patents were issued to Universities in the US in 2008. Furthermore these are some other facts found in a survey.

- Patenting high-tech inventions made on university campuses may not be a profitable undertaking, even at those universities best-positioned to profit from tech transfer. Based on the patenting and licensing activities of survey respondents, I estimate that university patent programs collectively earn a negative rate of return — an overall loss of more than three percent — on funds invested in high-tech patenting.

- The prospect of obtaining patent rights to the fruits of their research does not appear to motivate university researchers in high-tech fields to conduct more or better research. Eighty-five percent of professors report that patent rights are not among the top four factors motivating their research activities. Moreover, fifty-seven percent of professors report that they do not know how, or if at all, their university shares licensing revenue with inventors.

- University patent programs may, instead, actually reduce the quantity and quality of university research in high-tech fields by harming professors’ ability to obtain research funding, to collaborate with faculty from other institutions, and to disseminate their work to their colleagues.

- University patent programs seem to be, at best, a modest benefit to professors seeking to commercialize high-tech academic research. Entrepreneurial professors report that these programs hinder their ability to work as consultants with companies that show interest in their research, and fewer than half of university spin-off founders report that the ability to patent their research affirmatively helped their commercialization efforts.

 

Which can be looked up on the US patent Office website. If it ain't there it can be ran with. What he said is not completely untrue. That is a falsehood.

 

Didndt NMS get in trouble because they stole some universe algorithm

 

A patented one, yes.

 

For the same reason I "cotton" to Unity to build me a game engine so I don't have to waste my time on that part of it. Plus, you know, they're fricking better at it.

Have you ever actually done LexisNexis requests and seen exactly how much case-law gibberish is out there? Have you ever talked to an attorney about anything of a truly serious nature and followed the byzantine wanderings of legal theory? I have. This isn't a shader, this isn't a mathetmatical theorem, the law has conflicting rules and standards, there are points of procedure that must be followed, and overall it's a whole lot more serious than a SIGGRAPH paper.

 

(Pauses. Looks left and right.)

I'll be right back...

(Runs to the patent office.)

 

For what it's worth I once consulted with a lawyer about stuff for my first (now abandoned) game and regarding my question of what having a EULA written would cost me, he told me to just read the EULAs of some well established games and write my own by averaging what they say. He said that would literally be what he would do and it wouldd make no sense for me to pay him for ~4 hours of his time (at a 200,- € hourly rate) for that. Talking to him for 45 minutes cost me 150,- €. Fair enough, at least he didn't try to bullshit me into thinking I'd need more help than I actually needed.

 

That's why I've been replying to these legal threads about finding a good lawyer. The good ones don't gouge. One of my lawyers regularly tells me if he's going to have to write it up, or if he has something ready to go on file, or a combination of the two. If it's something boilerplate sometimes he doesn't even bother charging me. And once you develop a relationship with a good one, you can generally get a lot of free advice and you can probably expect repeat-customer pricing. They're a business like any other.

About the only time prices skyrocket is if you're unfortunate enough to be facing a courtroom (no matter which side of the argument you're on).

 

If they're having to ask this question they either (a) didn't read the legal document, (b) didn't know where to find the legal document for the paper (do they normally come with one or a link to one?), or (c) they didn't understand it.

 

I worked in a law office trading commodities and know UCC as well as how to write a contract that was accepted by top 10 world banks.. Yes. I know how to do the requisite and zero in on case law..not that hard. i don't know why everybody deems this esoterica. Black's Law Dictionary is your friend. It is very similar to a mathematical procedure. Many SIGGRAPH papers are much harder to comprehend than legalese and UCC under which this type of situation as to papers and patents is related.. So..instead of scaring the bejesus out of folks maybe back down. He wants to implement something from a paper. Simple. Go to the patent office and look up whether the implementation is patented. that is what an attorney would do. if not it is a go. If he is not to take legal advice from a forum then why are you giving him such in telling him he MUST seek an attorney or be frozen in place when the chances are he will not have any issues.??

 

"Get an attorney" is not legal advice.

 

Nor necessarily good advice in regards to draining his financial resources for something that can be determined by simply scanning the patent office records online. You are basically paying an attorney/lawyer/barrister to do a search for you at 150+ USD an hour. Felony charges are a whole other ballgame. Chances are that because this is code that there is little if anything that the papers writers can do. My bet is they go "cool dude..that is why we released this research..so that folks can use the knowledge". See my above post on surveys as to why researchers rarely patent research papers.

 

That is why I interjected. So that he would know the procedure instead of hiring it out at 150+USD per hour.

 

Unless you are the original author of the paper, you can't patent it. One of the conditions of patenting is that the invention cannot be obvious based on prior art. And if there is already a paper on it, that's definite proof of prior art.

One of the reasons code patents are so difficult is because of prior art. Programming seldom produces something that is obviously new and different from what was before.

 

@ippdev has it right. Patent searches are not hard to do.

The only reason I would consider a lawyer on this one is if you did find a patent and you couldn't figure out if it covered you. Even then you would want someone pretty specialized who understood both code and law.

 

Searching the patent database for something like a paper is relatively easy to do since you have some tangible words to base it on. My only wish is that it was this easy to search for general items. Take Apple's patent on overscroll bounce. Let's also say you were unaware of this before you implemented something similar. If you search for scroll or something similar you are likely to get a bazillion results (ok 56829 as of right now). If you are unaware of the marketing term that was used (I would have guessed spring scrolling) you likely are going to find it difficult to narrow it down. Couple this with the fact that if you spend any time in the database you realize that almost all basic ideas are patented and sometimes multiple times. In the above scrolling example you can find patents for bouncing, tilting, predictive scrolling, etc. At this point what is actually left? You can find this in many common aspects to user interfaces, algorithms, etc. You realize that things are utterly broken and are left utterly frustrated.

You might say that there is prior art and of course you'd win a case for the majority of these cases and of course you are right. However, that is not how patent trolls operate. They operate on the premise that the court case is more expensive than their license fee, so it will encourage you to settle. They also selectively pick their targets, not too small where they'll fold, but not too large where they already employ lawyers and would get their patent invalidated. They want someone where they'll easily extort there fee from but have little worry that the target can afford a long drawn out case.

I realize for OP they can easily search the database and if it's in a research paper there is a high chance that it's a new topic and not patented by anyone else. I just wanted to highlight though that for anything else, and likely things that are actually showing up in games or assets even by indies, that it's almost impossible to avoid stepping on patents even if you do your due diligence. After all you likely have scrolling, a certain algorithm, a network architectue, etc. in your product and many times the choice is either proceed anyways or leave something out. It's a very broken state of things right now.

 

Yeah. It's worth noting that my experience with the patent database is all chemical engineering based. It's not difficult to find if a phenol group with a chlorine in the 2 and 4 position and a acetic acid acid attatched to the -OH group is patented*.

Finding if scroll bounce is patented would be much harder.

* In case anyone was wondering, it's not. The specific molecule has been around for a long time and most of the patents have expired. There are a couple of recently discovered derivatives still under patent.

 

The lawyer isn't being suggested because it's a hard question, it's because advice from the Internet is notoriously unreliable (see the conflicting responses in this very thread - at least one side of each is wrong!), where a professional should be the opposite.

If I was having a medical issue I wouldn't trust my health to some random on the Internet, I'd go see my doctor. Sure, it might be an easy diagnosis for someone who knows what they're doing... but how do I know if any of the Internet randoms giving me (conflicting) answers know what they're doing?

Same deal with legal advice. If the question needs asking then I should as it of someone who's answer I can trust.

 

This is not a medical or life threatening or incarceration issue. It is an issue of outsourcing that which can be done by an individual without lining the pockets of someone who will be doing that which you could do anyways. Call me a penny pinching old fart if you will but if I can search a database for a few hours and come up with a conclusion versus someone else who will do the exact same thing at 150 bucks an hour or more then I will most definitely save myself those several hundred dollars and learn something (a valuable skillset) in the process. I did this exact thing with an invention i patented, including studying the format of patent verbiage so i could write my own patent and saved myself a minimum of 6K USD. In this case given the ratio of research paper gone to patent is extremely low the chances are the OP would be free to implement his code version of the idea..which was probably released in the research paper so that the greater interested community could make practical use of the research.

 

On the other hand, you could just be setting yourself up for a lawsuit which will cost you thousands of dollars. There is no way delving into a few databases and spending a few hours looking up legal stuff can make up for years of experience in that exact thing. If you care enough to spend that much time, you should get a lawyer to be sure. If you don't care, just do it and don't ask if it's legal or not. Like you said, you won't go to jail for this so who cares? If you don't care about getting sued, I don't care either.

 

Unfortunately that POV has already in hindsight been proven 100% wrong. Your concern is heart touching. But I guess I am not much into being paralyzed by professional use of industry proprietary language and verbiage. Just as the Unity SDK and docs help you understand the proprietary language of the game engine, Black's Law Dictionary will help you understand the SDK..so to speak..of UCC Admiralty courts, copyright and patents. Chances are with patents that if you type in the exact verbiae for your patent search and it does not come up you are good to go in this kind of a case. As well there are Utility patents..which are per this example. a running shoe has been patented..refrigeration has been patented.. put them together to make an air conditioned running shoe and that has not been patented in that form so you can file a Utility patent..the idea is that you have come up with a new confuguration of the lever, gear, pendulum and spring so to speak.

 

Well, to paraphrase Abe Lincoln: He who takes legal advice from themselves is taking advice from a fool. Granted he was a money grubbing lawyer so perhaps you're on to something. All I know, if I really, really wanted to know, I'd consult a lawyer. I can make more than enough to pay for a lawyer in the amount of time I would spend researching it anyway so I guess I'm luckier than most.

 

According to someone on the Internet who's legal background I've no reason to place faith in. It's not that you don't know what you're talking about, it's that I've no idea how to tell. That's why we have professionals and certifications and such. I don't know about the law, but this university and that law firm both do and they give someone the thumbs up so I know I can trust their advice.

Your cart's ahead of your horse, there. You can do an asset database search and save the cash because you've got prior experience and knowledge that allows you to do so. If the OP has to ask the question then they don't have that knowledge and experience. Your answer presupposes that we know what databases to search, that we know how to read legalese, and so on.

Also, consider that the "few hours" it may take you is probably a fraction of the time it'll take someone who doesn't have your background in doing exactly this. And that's not considering the increased risk that a small mistake might expose you to. Again, I'm not saying that it will happen in this case, but I don't have the knowledge or experience to make that judgement for myself.

So once again, if the question is really worth asking then it really ought to be asked of someone who's answer I can trust.

 

In general the only thing to say is the same thing that always has been said "attempt at your own risk". This holds true to the above, if you try to redo the plumbing in your house on your own, change the ball joint in your car by yourself, etc. There isn't anything inherently different and certainly if you do your due diligence you can save money. If you don't do your due diligence you can drown your house in water, get injured or die if your car falls on you or you wheel fails at high speeds, or you can be financially responsible if you tread on a valid patent. Plenty of people successfully do it themselves and plenty of people don't. One only needs to watch renovation realities to know that people don't always do their due diligence (some is scripted but some is not).

Trusting advice on the internet (especially forums) especially if it's your only source of information on any of the above topics is foolish, but it's not like you can't verify from reputable sources whether a real lawyer, nolo, a mechanic, etc. It's not impossible to do I've done plenty of DIY projects but it will be very time consuming vs. a professional, and it's important to go slow and steady because a mistake can be fatal. It's basically a cost vs. time proposition. It's also foolish to think every lawyer, contractor, mechanic, is good as well, so it's a good habit to always verify things anyways.

 

What's the worst that can happen? They can sue you for a percent of your profits.

If you make a million dollars and they sue you for 25% you still pocket $750,000. And they pocket $250,000. So everyone's happy.