Hacker News Reader

Comment by wtallis | Hacker News Reader

Is there any way for IBM to notify someone of potential infringement that you wouldn't call a threat?

There can be different objectives of informing someone of potential infringement:

1) making someone aware but saying right away that it's fine (that's what IBM as a self-declared friend and protector of open source should have done in this context)

2) making someone aware and offering a license deal on fair, reasonable and non-discriminatory (FRAND) terms; that would have made sense in this case, too, since TurboHercules stated in its first letter that it wanted to reach an agreement with IBM on such terms

3) making someone aware while simultaneously pursuing exclusionary, anticompetitive objectives

I like the first possibility, I can accept the second one if the offered terms are truly FRAND, but I despise the third option.

How about option

4) making someone aware and giving them the opportunity to cease and desist from commercial infringing activity

Is that a threat?

It's one thing to tolerate infringement from a community open-source project. It's another thing to let someone commercialize that technology and bundle it with hardware when a fully proprietary competitor would reasonably expect to be sued in to bankruptcy in no time flat. Should the fact that part of your business is "open" give the whole business immunity from patent infringement claims?

Also, IBM isn't "a self-declared friend and protector of open source". They only made a limited non-aggression pledge. They aren't (and aren't trying to be) in the same category as the FSF and EFF, or even RedHat.

Your point 4 is just a different way to phrase my point 3, the worst option of all.

Similarly, your distinction between a "community open-source project" and someone bundling it with hardware makes no sense. Companies bundling open source software with hardware or commercializing it in other ways are simply part of the community as long as they comply with open source terms.

So any use of software patents is unduly "exclusionary [and] anticompetitive", especially when applied to commercial users of open-source software?

I hadn't really picked up a strong "they're all wrong" vibe from what I've seen of your coverage of the smartphone mess.

You're now trying, not for the first time in this discussion, quite desperately to pretend not to understand what I say.

You previously just phrased my 3rd option differently to artifically create a 4th option.

Now you don't take into account that I explained 3 different approaches: grant a free license, grant a FRAND license, and prohibit any implementation of the patented invention only to foreclose competition, which is exactly what IBM (a monopolist in the context we're discussing) tried. Fortunately, the European Commission is investigating IBM for suspected violations of EU competition law.

You're implying that any use of patents to stifle competition is probably illegal, and should certainly be investigated. That rather thoroughly contradicts the definition of "patent".

Further, IBM's decade of tolerating Hercules development suggests that they do see a difference between community infringement and commercial infringement.

Your first paragraph is, once again, a complete strawman as opposed to what I said or implied. Not granting a license to a patent on FRAND terms is certainly anticompetitive, but not necessarily illegal. If, however, a monopolist does so in order to foreclose competition, then it raises serious antitrust issues and can, as the result of a regulatory process or lawsuit, be deemed illegal.

Your second paragraph doesn't make sense to me given that the patents asserted by IBM in that correspondence were related to the Hercules open source emulator, not specific to any commercial offering. The maintainer of the Hercules project explained on his blog why this was an attack on the Hercules project as a whole: http://ibmvshercules.com/

Reply on news.ycombinator.com