Hacker News Reader

Comment by marknutter | Hacker News Reader

The entire patent system has become so absurd. You know what protects your profits? Delivering actual value to customers. It's something you can't patent, and you can't simply corner the market on. You have to fight for it every day and anybody can take those customers from you without recourse.

View on HN

Interesting quote in the article: "“We invest about 30 billion kronor ($3.51 billion) a year in research and development and license our patents to the whole industry. We expect fair returns to be able to continue to invest,” Mr. Alfalahi said."

This is how you do parallel technology development these days. A large company invests billions in research and patents their ideas, if you want to use their ideas in your product you license their patents already knowing that the technology will work, saving you billions.

Without patents there is no way for Ericcson to recover their investment in research, so they just don't do it, so you never get 4GLTE cellular signals you're stuck in some form of Analog. Or you get no roaming at all because everyone has their own proprietary radio infrastructure.

I agree that there is a lot wrong with the patent system, I also recognize that there is tremendous value in being able to fund research in this way.

What you describe is what the patent system was created for. It's pretty easy to see the benefits in encouraging the advancement of costly technology.

On the other hand, it's software and business process "patents" that I think are harder to defend. Those are better protected IMO through Copyright as opposed to patents. OpenSource licenses (Apache and GPL, Mozilla, Eclipse) have particular provisions set aside to grant licenses to patents based on usage (BSD and MIT make no claims on patents), but Copyright is maintained. Trolls don't "use" anything, so the OpenSource license doesn't protect you from them (maybe indirectly GPLv3).

I think it would better the industry if we properly talk about which types of patents we have issues with.

What I don't understand with software patent and business process is that we have basically given up the "non-obvious to people of the trade" requirement.

You only need to see once, the "slide left to right to unlock" feature to be able to reproduce it in your platform without having to look at any piece of code. The idea is not obvious, but the implementation very much is and that's what patent are supposed to protect, implementations. (I'm picking on Apple there, but that's just an example on top of my head, software patent are all like that, except exceptional algorithms that are back by math behind and are therefore ironically non patentable )

It is like the light bulb patent. If just seeing a light bulb would have allowed any electrical engineer to make one over the we in his garage, I don't think it would have been patentable.

"Not obvious" means "not obvious on the day before the patent was filed". The point of the patent is to explain it to the point that it is that one ordinarily skilled can now implement the idea.

And copyright protects implementations. Patents protect ideas.

Profit is not co-extensive with value created. Profit is a function of value created but also your ability to control and monetize the delivery of value to customers. Software is a great example. Creating a shrink-wrapped software product creates a lot of value, but it's almost impossible to monetize shrink-wrapped software these days. That's why everything is moving to SaaS--it's much easier to monetize.

The way the consumer electronics market is currently structured, what drives your ability to monetize created value is strong, vertically-integrated manufacturing capability. That's what Apple's and Samsung's profits are built on. Creating technology delivers value, but you can only monetize it to the extent you can build and deliver Chinese-made products into the hands of consumers at high efficiency.

A great example of this is Apple buying P.A. Semi. A7-A9 have created tremendous value for consumers. But Apple purchased P.A. Semi for a pittance. The value-creation potential in P.A. Semi was not monetizable without access to Apple's incredible manufacturing muscle.

Under capitalism, faced with competition, people find the easiest and cheapest ways to protect profit. Delivering actual value is neither easy, nor cheap.

> people find the easiest and cheapest ways to protect profit. Delivering actual value is neither easy, nor cheap.

Isn't using patents and other anticompetitive protections easier and cheaper than delivering value? Sounds like you are agreeing that the patent system is bad.

Parent's saying that absent the necessary-evil patent system, people will copy other people's stuff rather than innovating, because it's easier.

That could mean a few things. On the assumption that you mean people will simply copy existing products rather than making their own: It's true enough on the surface, and some degree of protection is warranted, but at the same time everything contains copying of something before it, either in whole or in part, and to allow large sections to be locked down effectively in perpetuity (all of us will be dead by the time patents rolling in today expire) is to stifle innovation in another direction. Both too great and too little protection are harmful.

Elizabeth Holmes of Theranos has a great quote on this, paraphrased by former Secretary of State George Shultz:

"...the best patent is making yourself obsolete. So the person who steals your patent steals yesterday’s technology..." [0]

[0] http://fortune.com/2014/06/12/theranos-blood-holmes/

That statement comes somewhat ironically from a founder that now does not appear to have much of a technology at all.

> That statement comes somewhat ironically

I don't see how. If their technology was already out of date or they were copying patented ideas in their own, maybe.

It's more that it didn't work, at least not like they made believe. There's not much to make obsolete.

> Delivering actual value to customers.

That's exactly what companies usually patent - things or methods that deliver value, because theoretically it's what patenting is all about.

Delivering value does not immediately follow filing a patent. It's great that you came up with something novel (chances are it's not, you just beat everyone else to the patent office), but it's not worth anything until you actually figure out how to get someone to pay you money for it. That's the hard part, and that's where the value comes from.

I think cellular radios are exactly the sort of thing the patent system works well for.

Without patent system, what would stop someone from just cloning the IPhone and coming out with a lower cost version which doesn't factor in all the r&d that apple put into the development of the phone?

Trademarks.

a giant head start, brand, trade secrets (hardware and software), strategic supplier & manufacturer relationships, strategic partnerships (app developers & carriers)

Yes, which is why Apple has such a market share lead over Android.

Being the "first mover" is more often than not a disadvantage, and being one entails taking huge risks. ("Copycats" by Oded Shenkar is supposed to be a good book about this. Google "first mover fast follower" and "copycat innovation" to find additional relevant references. )

> just cloning the IPhone

Because it's considerably harder to do than you're making it out to be. It's not just the iPhone hardware that makes it valuable to customers, it's the support infrastructure Apple has in place, the software they write for it, the ecosystem of other products, etc. And let's be honest here - companies basically are cloning the iPhone for all intents and purposes and Apple basically clones other companies' successful products as well. They sue the crap out of each other as a cost of doing business and a bunch of lawyers buy new BMWs and lake homes and we all pay a few bucks more for our electronics. It's absurd.

Reply on news.ycombinator.com