Hacker News new | past | comments | ask | show | jobs | submit login
Red Hat's new patent counterclaim: GPL violation (itworld.com)
126 points by esolyt on Sept 16, 2012 | hide | past | web | favorite | 67 comments

I think the blog title is a bit sensational here. 'Patent trolls', at least as I understand the term, do not produce products. That's the whole point: they're trolls because they do nothing but sue people, without adding any value themselves. Twin Peaks does produce products, so they aren't a troll. Red Hat has just found an interesting way to counter-sue in a fairly ordinary patent dispute.

The fact that Twin Peaks clearly waited until Gluster had been acquired by Red Hat before filing doesn't make it a troll.

That and it's a one-time counter claim for this specific situation. From the title I presumed that they had somehow made a 'generic' weapon that could be used to shut down any patient troll.

Suing a patent troll for GPL violations is impossible by definition. A patent troll is a non-practicing entity, so they don't produce anything (but lawsuits) and thus they don't distribute software and conversely can't be sued for GPL violations.

If whatever company that is suing on patent grounds is also producing software (like this current case), then they are not a patent troll by the most common definition of the term but just plain aholes.

Nah, a patent troll is anyone who trolls with patents. That doesn't need to be the only thing they do. The classic example, SCO, had actual products which weren't related in any way to their patent suits.

No, a patent troll is a non-practicing entity (NPE). The specific reason they're dangerous is because they aren't subject to counter claims. You can't come along and redefine a word simply because you don't understand it.

The term is fairly new and as such the common definition is still changing, but Wikipedia agrees with regularfry

Patent troll is a pejorative term used for a person or company who enforces patents against one or more alleged infringers in a manner considered aggressive or opportunistic with no intention to manufacture or market the patented invention.*


(this article is referenced: Alexander Poltorak. "On 'Patent Trolls' and Injunctive Relief"., ipfrontline.com, May 12, 2006)

Odd summation. The Etymology and definition section of the same wikipedia article leans towards the non-practicing entity notion of a patent troll.

Just being a jerk with a patent dosen't make you a patent troll.

> You can't come along and redefine a word simply because you don't understand it.

Physician, heal thyself.

non-practicing in the area of the patent.

Read again, he wrote patient troll, not patent.

That was unintentional sorry: I use the word patient far more than patent, didn't notice.

Yeah, I thought so, but the play on word was nice, given that they waited for Red Hat to acquire the tech.

It is the examples of prior art that make it patent trolling.

Nope, not patent troll (https://en.wikipedia.org/wiki/Patent_troll):

Patent troll is a pejorative term used for a person or company who enforces patents against one or more alleged infringers in a manner considered aggressive or opportunistic with no intention to manufacture or market the patented invention.

I reject that definition and suggest that it should instead mean, "a pejorative term used for a person or company who enforces one or more patent(s) that do not fullfil the requirement of being non-obvious and inventive against one or more alleged infringers in any manner, especially if considered aggressive or opportunistic, or with no intention to manufacture or market the patented invention".

When I say patent troll, I mean someone using patents to delay innovation instead of using them to encourage innovation (as they were intended). In this case, there's clear prior art and they are suing after the company responsible was acquired by Red Hat. I'm not going to accept that someone is not a patent troll if they happen to have an application or framework that, e.g., redirects users to a online store for purchase of the full version of an app after tapping a link. Everyone in the field, the companies starting the lawsuits included, know that these patents should not have been issued.

Care to expand on that?

The patent for the Mirror File System was applied for in 2001, years after Novell Netware had exactly this sort of thing built into the operating system. That is just one example of dozens that will surely come to light. Distributed, replicated filesystems have been around for decades.

"Intellectual property" as a socially and legally recognized concept is nonsensical and harmful, even in instances like the GPL where it's intended to increase freedom.

I've started using the older term "monopoly" instead of "property". Respect for property rights is so deeply ingrained in our culture that it's really hard to argue against it without sounding like some crazed hippie. Arguing against monopolies is much more respectable.

Usually this argument is summed up more succinctly by adding two words to the phrase: "is neither"

If we could live off of a handful of ideas and had no want of food, shelter, or clothing, I'd be in complete agreement.

This is the point where you demonstrate with evidence that a lack of recognition of IP would cause such problems, right?

As someone who gets paid for developing software that can be freely shared, I'm particularly interested.

If there's zero protection for "intellectual property", it becomes a lot harder for the majority of the software industry to even function.

Apple, as an enterprise, has been very successful by building a profitable operating system platform and channeling those profits back in to both hardware and software engineering. That kind of enterprise would not exist if Apple's software could be used by anyone for anything.

Any hardware vendor of any consequence could use that as they see fit, resulting in a stew of mis-matched standards and rampant confusion, not unlike the Android tablet market. While there are good products in this space, they are often difficult to differentiate from superficially similar yet technically inferior alternatives. This is not helping people on the whole.

Personally I like the choice between a premium product which is commercialized and free, open-source alternatives that are, for many applications, good enough. I would not want a market that's locked in to one model at the expense of the other.

We (the human race) would be worse off if we only had a handful of words or numbers, yet I believe that respecting words and numbers as property is nonsensical and harmful.

Your fallacious argument intrigues me. Fallacious because we can still live quite well without there being intellectual property. Could you expand upon your theme? Perhaps I'm not following.

But you are confounding physical property with "intellectual property", which is infinitely reproducible once discovered.

The only reason patents exist is to "promote the science and the useful arts", and it seems to be theres a growing body of evidence that this goal is not being achieved (at least, in the field of software - may be its working great for the manufacturing/pharma industries).

[citation needed]

That's great, except it would be totally ineffectual against a real troll. If you're not making anything, you're likely not violating the GPL.

You could use it against the patent troll's masters (eg. Mosaid)

This is not going to be effective unless Red Hat really pushes for damages. Shutting down Twin Peaks products is not a big deal, they will just release the source of their mount.mfs and/or rewrite the mount, and then push out a new version that doesn't violate mount's license.

Except that GPLv2 doesn't work like that. Once the license is violated you permanently lose the rights granted by the license. So even if Twin Peaks fixes their stuff and comes back into compliance, they do not get their rights back automatically -- the copyright holders have to explicitly give them back these rights.

This is one of the important differences between GPLv2 and GPLv3.

Wait, so when you're dealing with GPLv3 stuff, you do get the license for redistribution back automatically?

Where's this stated exactly in the license?

>Since both of Twin Peaks' products are released under a proprietary license, there was no source code of any changes it made to mount released, which is a violation of the GPLv2.

Seems like Twin Peaks could resolve this by just releasing the source code to their version of mount. Maybe Red Hat will be able to get damages out of it, but if Twin Peaks releases the code in a timely fashion, I doubt Red Hat will be able to get an injunction against their product sales. (IANAL.) Meanwhile, the patent suit goes on.

The GPL has a clause saying that if its terms are violated, all rights granted to the violator by the GPL are terminated. So if Twin Peaks did indeed copy Red Hat's mount code as alleged, they would never be allowed to distribute that code again (without specific permission).

Term 5? All that says is if you don't accept the terms, you don't get to use the software. There's no irrevocable termination of rights.

At worst, all they have to do is download a new copy of mount and presto, they have a new license.

[err, that's for gpl2. gpl3 has a more complicated termination procedure.]

ok, so in light of gpl3, they just need to go back and get a gpl2 copy of mount. It's not like mount has changed all that much, I'm sure the 2007 version will suffice. Or they could just write their own. How many lines of code are we talking about here?

Term 4 in GPLv2. "Any attempt otherwise to copy ... will automatically terminate your rights under this License."

Read right over that. But I suspect they can still just download another copy, as the second download would be a new offer, and they can then agree and comply with that one.

It's not that simple. Relevant reading in this context:


"Each kernel release is a different work; the chances that any given piece of code has been modified in a new release are pretty high. One could argue that each kernel release comes with its own license; the termination of one does not necessarily affect rights to other releases."

That is the argument I am making.

My only hope in this double-case is that if TPS offers an even settlement, RedHat refuses it.

Fighting GPL violations is the "good" side, patent trolling the "bad".

>Since free software is so pervasive these days, it might not be very hard to find something within a company's product portfolio that depends greatly on free or open source software. Once discovered, all it takes is a claim of license violation like this one to shut down that software

That paragraph actually makes it sound scary to even use open source software when they actually mean only GPL'ed software and that too only when modified and distributed/sold.

> That paragraph actually makes it sound scary to even use open source software when they actually mean only GPL'ed software and that too only when modified and distributed/sold.

No, you have to comply with the GPL regardless of whether you've modified the source that you're distributing. (As in this case: I'd be surprised if Twin Peaks modified mount -- I don't see why they'd need to.)

They are a file system vendor, they probably added some mount options for their file system to mount.

If I remember correctly, each file system has its own mount.fsname command, the mount command itself just routes to those individual commands (based on the -t argument). You could surely add tools for a filesystem without modifying the general mount tool itself.

Ah yes you are right. So presumably they copied one of the other ones as a template. Its not much code and easy to rectify then.

Unfortunately, all shipped products already willingly violated RH's copyright. Doesn't look too good.

You completely missed the point.

There's a lot of open source code that isn't GPL. Many licenses don't require the code to be distributed, even if it's changed.

The sentence from the article makes it sound like all open source code is GPL.

Yes, but the poster 'recoiledsnake' made a separate mistake, stating that modification to GPL code is required for it to be vulnerable to lawsuits. As cjbprime stated, that isn't true; using the code is enough for it to be covered.

Distribution not use is required for the GPL to restrict you. But you are right that modification is irrelevant.

thanks, funny that I also made a mistake

Could we (as in, everyone) please stop referring to programs in general as "apps"? It's retarded. An "app" is a specific kind of program for a smartphone. mount isn't such a program.

Sorry, this just bugged me.

>An "app" is a specific kind of program for a smartphone.

No, "app" is short for "application". The term "killer app" has been around long before there was such a thing as a smartphone.

Stop being bothered by such trivial matters, moreso when you don't know what you're talking about.

Could we (as in, everyone) please stop referring to something we find stupid or dumb in general as "retarded"? It's offensive.

Sorry, this just bugged me.

If your best argument for why someone should stop doing something is that "it's offensive," I will have trouble taking your side.

"it's retarded" appears to take on the place in a paragraph where an actual argument or reason would go, but it isn't. It's just a childish phrase used to indicate dislike for something. If you don't like the "it's offensive to use a phrase indicating some form of developmental disorder as a pejorative", then please consider the argument that it accomplishes and communicates nothing except the fact that the speaker / writer lacks the ability or motivation to articulate his objections.

When given a choice between two otherwise equivalent options, most non-sociopaths will choose the option that is less likely to offend their audience.

Fuck the audience. If they are so easily offended they should get therapy instead of calling others sociopaths. Offence is taken, not given.

However, if you're attempting to get a point across, you generally do better by keeping social norms in mind. Offending your audience instantly makes them think about whatever "horrible" thing you just said and not on whatever point you're trying to communicate.

Put another way, if nobody is getting your point, perhaps the problem isn't with the audience :)

(I am, however, going to steal that "offense is taken, not given" line for future (ab)use)

> However, if you're attempting to get a point across, you generally do better by keeping social norms in mind.

And if everyone followed that principle, society would never change. The perfect fourth would still sound dissonant, rape victims would be under even more pressure to keep the crime secret, etc.

You know, i did say generally. Did you just miss that or..?

The term "app" was in use before smartphones even existed, so no.

I'm with wikipedia on this one: "Application software, also known as an application or an app, is computer software designed to help the user to perform specific tasks." From http://en.wikipedia.org/wiki/Software_application

Application software is such a vague and undescriptive word that it's better to just use the terms "program" and "software" as they're impossible to misuse or argue about.

Wikipedia uses the term "utility" to refer to programs like mount, although again, just call it a program and be done with it.

(this is just a general remark, not a reply)

"App" doesn't imply smartphones. It has been used this way since at least the 80s:


Here's one of those retarded people:


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