Hacker News Reader

Comment by throwawaykf | Hacker News Reader

That's just, like, your opinion man.

Me, I prefer to see data. Fortunately, there are a number of studies coming out about patent trolls. Unfortunately, there is no clear consensus. Also, things like the impact of baseless demand letters simply cannot be studied because they cannot be tracked and there is no data. This will hopefully will change soon.

But I'd much rather prefer analytical study over uninformed vitriol. Ars Techica is usually good, but not when it comes to patent matters. This is the same blog that painted CSIRO, a well known Australian research lab, as a patent troll because they claimed to invent WiFi and sued manufacturers over it. The comments on that article (many from posters down under, naturally) kept calling out Ars' incorrect rewriting of history, but they would not back down. I wouldn't want my news from somebody who won't accept the facts because they don't agree with their biases.

View on HN

There isn't exact data and there will never be exact data, because many of the settlements extracted are confidential and there is no incentive for anyone to talk about them.

If someone is ambushing people in the street, hitting them over the head with a pipe, and robbing them, I don't need to see a double fucken' blind study to decide that this needs to stop.

You've reminded me of a meta-study published in some medical journal, analysing "parachute application to prevent trauma due to gravitational challenge", i.e. if it's worth wearing a parachute when jumping out of a plane. They found that little high-quality data exists to support parachute use, and recommended further study.

You don't need exact data, you just need enough. There's no doubt bad actors in the system, but do they characterize all of them?

The papers I read indicate that trolls actually often have "strong" patents, but like you said, most of those settle out of court, so you never hear about them. The ones that go to court are the weak ones, and those are the ones you hear about, which creates an image that all trolls assert bogus patents.

I mean even NewEgg, who've been spouting a lot of "We Never Settle" rhetoric admitted, in this very case, that they've licensed a majority of patents that were asserted against them. The coverage on Ars is a bit brief, but it sounds like out of 80 assertions NewEgg has taken it to court only a handful of times. To me, that sounds like the system is not imbalanced.

> If someone is ambushing people in the street, hitting them over the head with a pipe, and robbing them, I don't need to see a double fucken' blind study to decide that this needs to stop.

Or maybe the alleged victims claiming to be robbed are lying because they don't want to pay money owed to the alleged assailant and they're hoping he gets put away? Are you going to believe individual he-said/she-said cases or data?

BTW the parachute study sounds hilarious. Link?

The study is here: http://www.bmj.com/content/327/7429/1459. One of the authors archived it here, http://elucidation.free.fr/parachuteBMJ.pdf, since the British Medical Journal website is restriced-access. It actually is a pretty profound comment; I recommend it highly.

The fact that many of these idiot patents are legally strong is precisely the problem; that's why we need patent reform to kill them.

Wow, thanks for pointing out that paper.

> We think that everyone might benefit if the most radical protagonists of evidence based medicine organised and participated in a double blind, randomised, placebo controlled, crossover trial of the parachute.

When jumping out of a plane, parachutes are optional. Just make sure you put one on before you hit the ground:

http://www.youtube.com/watch?v=lDBrdl2sZWs

There have been studies. Here's one of them:

http://www.bu.edu/law/faculty/scholarship/workingpapers/Bess...

As for consensus...I don't need consensus to know what's ethical. I've been working in software all of my professional life. Software patents are a disaster for innovation.

Reply on news.ycombinator.com