Comment by throwawaykf | Hacker News Reader

Comment by throwawaykf | Hacker News Reader

Parent Full thread

Sorry, I should have said "tech media". Every "tech journalist" and their grandmother out there likes to write about patent trolls, and calling for the government to make changes.

And here we have a study by a branch of the US Congress, the one entity that, based on the constitution, can make sweeping changes to the patent system. The Congress' investigative branch went off to study the problem and essentially said trolls are not a big problem.

But as far as tech media patent-related reporting goes, this one has gone nigh unmentioned. Even though software patents are specifically called out. Why?

My suspicion is that the lack of attention is mostly because the study does not gel with the perception the tech media is trying to sell, that trolls are a huge burden on the economy. And typically, these complaints are made by practicing companies who, if one were to be uncharitable, one could say are actually doing most of the litigation [1].

To me, the signs point to a submarine [2]. And if that seems far-fetched, consider this:

You ever hear of a "study" that claimed trolls cost the economy something like 29 billion (pinkie to corner of mouth) dollars a year? A number that gets mentioned in any article about trolls or patents in general? Well, last year multiple sources poked holes all over it [3, 4, 5, 6].

Would you like to guess how many tech media outlets reported about that?

Now, would you also like to guess how many tech media outlets continue to harp on about the debunked 29 billion number every chance they get?

1. http://seattletimes.com/html/businesstechnology/2021696607_g...

2. http://www.paulgraham.com/submarine.html

3. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2117421

4. http://blog.patentology.com.au/2012/06/29-billion-us-troll-t...

5. (Pay-/Registration-walled) http://www.iam-magazine.com/blog/Detail.aspx?g=454c1adc-52c3...

6. (A recap of the previous 3 references, as well as highlighting the sources of funding for the authors of the "study") http://gametimeip.com/2012/07/30/patent-scholars-challenge-b...

I don't buy it. Who the fuck wants patents on not just well researched algorithms but something as simple as the doubly linked list? Even though I haven't seen that used in court, for good reason, I still can't see that and other signs as symptomatic of something that needs to be taken behind a shed.

Who profits from software patents? Who gets held back?

Why does the narrative matter at all? Don't you remember the fucking shit throwing contest over pinch to zoom?

What drugs are you taking. Are they legal?

> Who profits from software patents? Who gets held back?

That is the only question that matters (beyond just "software" patents,) And there is a lot of work by a lot of people that try to answer the question. Search ssrn for work by authors like Risch, Merges, Allison, Lemley, Mann, Zeidonis, Etro (just for starters) and you'll see that the issue is much more nuanced and complex than what the common narrative tells you.

So does the narrative matter? Well, let's look at the rest of your comment.

You mention a "doubly linked list" patent. That patent is not the greatest ever, but it is not a doubly linked list: http://patents.stackexchange.com/questions/738/prior-art-for...

(Note that even the highest rated answers there get it slightly off.)

Then, pinch to zoom; I'm not sure what you refer to by that, but the last time Apple's patent was mentioned on HN, its scope was also grossly misrepresented.

Your reply makes two things clear:

1) you have strong emotions about this topic; and

2) you don't really understand how these things work.

And seeing other comments on HN over many years, I can safely extrapolate another thing:

3) your condition is common to the vast majority of commenters here.

At least to me it seems clear why the narrative matters.

Why can't an astroturfer at least use a good pseudonym.

a). A data structure is patented. I'll give you that it's not a doubly linked-list, but seriously. If we are going to start claiming modifications are unique enough for a patent I guess I should probably look into copying some n-array.

b). http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sect2=H... That patent is for zooming and making the UI 'bounce' as though somehow that is an amazing and ingenious idea. No, it's not. Just because I'm too stupid too invent an iPhone does not mean I couldn't have invented that in less than a week.

I do have strong emotions about this topic! When a company infringes on the hard work of others I think it's a real fucking big deal. When tripe shit gets media attention and takes years to settle, it just fucks over everyone with a real patent.

Listen, I agree with the Rearden steel narrative. But if we are going to let any fucker who can fill out a patent application get dibs on a goddamn data structure, well you can go fuck yourself. I'd rather be a damn hermit at that point.

Apologies for late reply. Who would I be astroturfing for, exactly?

a. Can you make a logical argument why data structures should not be patented, as opposed to, say, mechanical or electronic structures? Try to differentiate them using concrete, technical reasons rather than handwavy "software is math" (aka "mechanisms are just physics" or "electronics are just flowing particles") arguments.

Also, modifications have always been patentable ("a better mousetrap" etc.) The thing is, this patent is not even a straightforward modification; I still have not figured out a solid reason why anyone would use this, a single multiply-ordered linked list, over multiple differently-ordered linked lists.

b. That's not the "rubber banding" patent you linked, that's the "pinch-to-zoom-if-touch-input-is-multi-point-else-scroll" patent. Anyway. You could argue both of those are bogus patents, but you have the benefit of hindsight. Hindsight is a very powerful bias, which is precisely why US (and most other places) require examiners to provide concrete proof, in the form of dated, published documents, of non-novelty or obviousness to deny a claim. This is, as far as I can tell, the only way to truly objectively quantify a claim's contribution, and this is why you see the patents you see.

Not saying it's a perfect system, but once you decide that patents should be a thing, I personally can't think of a more objective one.

Also, if you think it's as easy as "fill out a patent application to get dibs on something", I strongly encourage you to give it a shot. Assuming a micro-entity status in the US, it should cost you a couple hundred bucks.

Reply on news.ycombinator.com