Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

> “The ruling is a watershed moment in the history of copyleft licensing. This ruling shows that the GPL agreements function both as copyright licenses and as contractual agreements,” says Karen M. Sandler, executive director of Software Freedom Conservancy. Sandler noted that many in the Free and Open Source Software (FOSS) legal community argue incorrectly that the GPL and other copyleft licenses only function as copyright licenses.

If the GPL is a contractual agreement, does that mean that a court can require that an entity to release the closed source portion of infringing software that uses the GPL?

If that is the case, the legal jeopardy involved with using GPL software just got a lot higher.



>> The funding bodies want us to detail all expenses and justify all the changes in the plan, because they don't want people to say that the public funds are mismanaged. This is just an example. We also write tons of useless papers, to justify that we are productive with this money.

That has always been the case. The alternative would be to disclaim the GPL and face straight copyright infringement, which in the US can be as high as $150K per copy. No company making wide use of GPL and infringing it would want to go the copyright route.

The GPL is not really dangerous except to someone with no legal clue who is deliberately ripping off free software.


> The alternative would be to disclaim the GPL and face straight copyright infringement, which in the US can be as high as $150K per copy.

Per work, not per copy.


Oh my bad. The wording is per infringement, which does seem to mean per work rather than per copy. That makes it perfect for going after individuals but not corporations - because it's peanuts to them.


It might be peanuts if it was a one-time thing, but if they kept infringing I would expect they could be sued again for the additional infringements.

Statutory damages in the US start out in the $750-30000 range, and can be raised up to $150000 in the case of willful infringement. The burden of proving willfulness is on the plaintiff, but that shouldn't be too hard if the defendant already lost once and is continuing to infringe.

I'd expect courts to start going higher and higher up the range for damages the more times they see the same defendant losing for continued infringement of the same work.

I'd expect that this would get the defendants to start complying with the license and release the source.


Companies should be more worried about injections than infringement. Sure they might pay court fines, but courts can also order them to stop infringing immediately. Continuing to infringe after a court order is a very different issue, and isn't about copyright law it anymore.


You pay for all the software you use, unless you are in a country with public domain and the software is clearly cast into it.

Some software you pay for in dollars. Some in postcards. Some by exchanging patent rights, or promising not to sue. Some by accepting contracts to contribute back changes you make, under certain circumstances.

There is no legal jeopardy involved in using GPL software that is different from buying a math library: you read the contract, you pay the price, you use the software in accordance with the contract.

Violate the contract, penalties attach, starting with enforcement of the contract.


> the legal jeopardy involved with using GPL software just got a lot higher

Not at all. If you are a developer, this only protects you from patent trolls or any confused company that might sue the wrong person.

Same if you are a user.

The only "legal jeopardy" is for those who repeatedly violate developers and users right. And that's a good thing.


That was always the case. This simply alters who can enforce it.


The requirements were there, but yes, now you go from a relatively friendly group of developers eager to see code shared around to the SFC, which has very strong views on things like unlocks on hardware - I think it is potentially a pretty major change here.


I believe both of these statements are correct.

One of the interesting effect of finally litigating the GPL is that the 'theoretical' implications become a matter of fact. And with that comes a bunch of clarity around what the risks are of using GPL software in your products.

It will be interesting if GPL "wins" and a bunch of companies immediately tells their employees to strip every bit of GPL code out of the system and rewrite from first principles any function that is required for the product to operate. There will be a market for programmers who do systems programming (which is currently not a big market, but it will get bigger.)


> what the risks are of using GPL software in your products

"Risk" is a weird way to phrase it. It's a cost for taking advantage of someone else's work, just the same as if you paid for a library. If you're not willing to pay the cost, don't use the software, just like companies would tell their employees not to use a pirated version of Photoshop.

> It will be interesting if GPL "wins" and a bunch of companies immediately tells their employees to strip every bit of GPL code out of the system and rewrite from first principles any function that is required for the product to operate.

This seems unlikely. Is it really that common for companies to violate the GPL today?


I'm interested in what makes you think the term 'risk' is weird. In my experience most, if not the majority, of business decisions a company made were composed at the management level as a trade-off between risk and reward. The decision process included a list of things that could go wrong, or take longer than expected, Etc and those things were in the risk category, and things like time to market, operational efficiency, better margins, Etc. were in the reward category.

From that perspective the choice to use open source software in their product seemed to often include two risks, one was how would you respond if it broke and the other was what would you have to do to replace it if it became unavailable for some reason. To use a pretty well known example, the Android system from Google, which you could compile and run on your phone for "free" but the risk was that Microsoft would (and did) come after you for patent infringement if you used it in a phone [1].

The risk here was to margins, where a manager might assume the cost of shipping this software was $0 (no license cost) and have it end up costing $Y because of patent fees.

When Blekko was acquired by IBM we had to "blue wash" our code, which was to go through and identify every copyright, every comment that might imply ownership, and figure out if it was "okay", "not okay", or "needs more research." That entire exercise was described as "Minimizing the risk that the IBM company will be sued because they are using the code for your product."

So that is where I'm coming from here.

[1] https://finance.yahoo.com/news/microsoft-may-relinquishing-b...


> Is it really that common for companies to violate the GPL today?

Every company I've worked at is careful about GPL. Some are more careful than others, but they all put at least some effort into ensuring nobody brings in GPL libraries.

As a lawyer explained to me, they understand at some point GPL code will be shipped - it is too tempting. The real goal is to ensure that when it happens they can convince the courts it is a rouge employee doing something he wasn't supposed to, as then the penalty is a slap on the wrist and a bunch of developers emergency switched to rid our stuff of GPL. If the courts decide the company didn't do enough to prevent infringement then the court will decide that it was company policy to make their product open source and the courts will force the release of source code. This is why all developers I work with have to take open source training, we have someone assigned to audit all our code, and we have bought tools that look for potential open source code, it all builds a case before the court. To my knowledge the above as kept us from infringing in the first place, which is the real goal, but since all tools have holes eventually we can assume it won't.


> There will be a market for programmers who do systems programming (which is currently not a big market, but it will get bigger.)

What‽ The market for systems programmers is enormous! You think "smart" things program themselves? What about cars or industrial systems or traffic lights or well, literally everything that isn't running on a full blown computer?

Systems programmers don't get much play on open source sites like Github but they're probably the majority of software developers world-wide because there's vastly more embedded systems than there are desktop and servers.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: