Comment by throwaway2048 | Hacker News Reader

Comment by throwaway2048 | Hacker News Reader

I have a feeling that eventually we are going to be facing basically a completely paralyzed tech industry because of what basically amounts to a patent singularity.

As more and more formerly innovating and profitable companies that invented and patented key technologies either start to struggle, get bought out, or go out of business, there will be more and more pressure to monetize their huge patent warchests, (and competing companies are even more driven to build up their own warchests as a result) we are going to see wave after wave of massive industry wide patent lawsuits like this that are going to completely disrupt things.

We absolutely need massive patent reform in the USA before things reach this point and any incentive to bring new products to market is destroyed by the tidal wave of patent suits that it is virtually guaranteed to attract.

This stuff has very extreme chilling effects, for instance a technology that has almost limitless potential (forward error correction) receives very little attention because of how crazy the patent situation surrounding it is.

Since the technology moves on so fast, shortening the patent expiration to eg. 1year would solve most of the problems.

Companies get 1 year head start (which is a lot), and after 1 year, we all benefit.

Wouldn't the slowness of the courts and the lengths of processing lead to virtually no protection at all? If going through the courts takes sometimes 3+ years but parents expire after 1 year, every final court ruling would be retroactive.

So if you prevail you get payed retroactively for the damages. Whats wrong with that?

You can get six years back damages, and often you have a substantial portion of a patent's 20-year term in front of you. And frequently cases settle well before you are three years into a suit.

If you're suggesting a 1 year patent period for all utility patents, you might as well just get rid of patents. In the industry I work in (Biotech - DNA sequencing) it takes maybe 5 years to mature an approach and get it to market.

Once your patent is published, there isn't much to stop others developing the same method [1] (they just would run into trouble if they sold it commercially). So at the same time that the patent owner is developing their platform, a competitor can be developing theirs and launch at the same time or nearly the same time as you.

Certain areas, like software patents, I'd possibly agree. I don't think software/algorithm patents are a good idea for a number of reasons however...

[1] General consensus seems to be patents don't cover academic work. I've also heard that they don't cover commercial research, and that in any case it's hard to provide damages. http://softlib.rice.edu/ques.html#2

> If you're suggesting a 1 year patent period for all utility patents, you might as well just get rid of patents. In the industry I work in (Biotech - DNA sequencing) it takes maybe 5 years to mature an approach and get it to market.

So have a two phase validity. Once it's in the market you have a year.

You’d want some limitations on the first phase too, so that companies don’t hold onto their patents indefinitely while waiting for the perfect market conditions for their one-year monopoly.

A real solution is don't apply for patent protection. The down side is if someone else also figured it out, you don't get anything. Makes sense to me.

Yes, I agree. I'm not a fan of patents. While I understand the desire to protect ones investment, it doesn't make sense to me that it should be done at the expense of someone else's investment, since multiple people coming up with similar ideas around the same time isn't particularly uncommon and unfortunately there's no easy way to differentiate those from plagiarism.

This encourages companies to withhold products from the market. In many, cases they will withhold products for the maximal allowable period before releasing them into the market. This will give them the maximum period of patent coverage, and prevent other players from releasing products. Market leaders in particular will want to extend their coverage using this technique.

Aside from this, what is "on the market" most novel scientific equipment undergoes early access. This period often lasts up to a year, usually equipment is sold at a loss or near cost. The product is validated. So... they recover none of their R&D during the patent period (making the patent largely useless).

So... perhaps you structure things such that you give this patent covered product away for free during early access. But you use the product to recognize revenue elsewhere (consumables, reagents... whatever)... Thereby extending your coverage period.

So... how will all this be policed/structured so companies can't find ways round it?

If you can find a way of enforcing it, it still doesn't work out for the companies. In my experience a scientific startup needs at least 5 years in R&D.. an then at least 5 years on the market to cover those R&D expenses.

If this is going to get funded by a VC it then needs a high enough return to cover all the VCs investments. This is then looking like a minimum of 10 years of patent coverage, and something like 20 years to make it worthwhile for VCs.

Yea, you can get rid of patents all together. Probably the investment/funding sources will need to change and that might not be a bad thing. But I can't see that what you're suggesting is workable.

Patent validity should not be one size fits all like it currently is. It's hard to imagine but the patent system was actually designed to encourage growth, not slow it. The validity period should be part of the application process and awarded based on things like time-to-market and the length of the research effort involved. Pharma is an example of a long patent period and a software patent like two finger zooming should be shorter (maybe five years).

> It's hard to imagine but the patent system was actually designed to encourage growth, not slow it.

And that was done by people completely oblivious to how real world economics work, and reinforced in that delusion by later generations of not much smarter statesmen.

It is not humanly possible to imagine how economic growth and efficiency is encouraged by intentional introduction of economic inefficiency through imposition of artificial scarcity (which is what patents are, literally a license on economic inefficiency)

It actually is quite humanly possible. The fundamental issue is that while replication of information is effectively free, generation of (valuable) information most certainly is not. Which takes us right back into limited resource allocation and value decision territory. And markets are the best tool we've found so far for societies to use to perform generalized efficient resource allocation towards societal goals. It's also often quite possible to keep information secret for a significant time period, but efforts to do so (and efforts in turn to gain that information) are themselves examples of the "economic inefficiency" you complain about. However, generalized efforts to restrict information require the cooperation of society overall and costs to society in a way that real physical property does not.

"Intellectual Property" is one possible way to try to answer all these points, a hack to try to couple market allocation to the information generation issue while also recognizing that it's not real property and is funded by and for the public. It offers a chance, not a certainty, to make money off an investment on information generation through convincing other parts of society to pay for it piecemeal during a limited pseudo-physical-property replication stage. In exchange, it shouldn't be secret, shouldn't cover basic facts of the world or certain other areas where additional generation incentives are unnecessary, and should have limited times taking into account the ongoing and ever increasing societal cost.

I think it's very foolish to mistake implementation issues with fundamental ones and throw out the baby with the bath water unless you've got something better to fund information generation equally effective. And I have yet to hear any proposed scheme by total-anti-IP folks to take its place, merely complaints about the current implementation. I think it'd be far more productive to first direct efforts towards fixing mismatches in historical IP in each of the 3 areas above vs modern times. On the secrecy issue for example, encrypted source code escrow should be a requirement for having copyright protection for everything that isn't open or available source already, and have-your-cake-and-eat-it Trade Secret laws should be eliminated. Patent protection for mathematics (all software patents) should be completely eliminated, copyright is more then enough already for software and better aligns incentives. And new, better ways to reflect ongoing and increasing societal costs, and the divergence in areas of generation, should be added rather than mere fixed times. Ongoing yearly renewal beyond an initial short period at non-linearly increasing cost (either fixed or an increasing percentage of revenue) would be one possible approach.

But all this seems far more useful than anything else I've seen. Total command and control at all levels of the economic stack hasn't proved any more effective for information generation than it has anywhere else in economics. IP has overall worked fairly effective, and it seems to me that the problems it has caused are more down to warts and rust that could be fixed.

> The fundamental issue is that while replication of information is effectively free, generation of (valuable) information most certainly is not.

This is a complete delusion. This is what about Goebbels said, "the bigger the lie, the easier it is to believe in."

People are so indoctrinated in that way of thinking, that some times I have to show this though experiment:

1. Set up a machine that will "destroy" the IP value of something by copying it.

2. Make it do so a billion times, and watch if the company whose IP being "infringed" goes bankrupt.

Now, will the company go bankrupt? Would they "run out" of their "IP generation resource?" Definitely not, that would be completely inconsequential to them. They nor lose nor gain anything. And then, all those guys run to make up some smart "economic theory" to explain that, often mutually contradicting, one sillier than another.

My idea is: if though process coming under IP generation is an unproductive process, that does not make money, why does the society goes nuts to encourage and facilitate that? A basic understanding of economic principles tells this is just like throwing money away.

See, US always brags of its informativeness. What was the last "big thing" an American company ever came with? I will say it was transistor, and it was more than 60 years ago, when no monstrous IP protection regime was in place.

It was not American companies who came up with: hoverboards, e ciggaretes, selfie sticks, stick computer form factor, an untold number of innovations in construction materials (MgO board is a Chinese invention, as well as FRPP piping, modern foam concrete manufacturing machinery,) an untold number of advances in materials science which allowed for $200 laptops being made from carbon fibre (A US made analogue will cost you $2000 just for the casing)

Just how many new products are coming from China in virtual anonymity and never given acknowledgement for their innovative nature are made without any consideration for getting remunerated for their "IP"

Now whose system make more innovation US or China's?

I clearly see, the moment IP protection began verging on extreme, US began stagnate in science and technology, and Chine went ahead.

>This is a complete delusion.

Um, no. It's objective fact. Whether it's a video game or R&D on drugs or semiconductors, it does in fact require the expending of limited resources. Are you for real? This stuff doesn't pop into existence out of thin air.

>This is what about Goebbels said, "the bigger the lie, the easier it is to believe in."

Ah yes going the Nazi quote angle, always the sign of someone who is definitely not a crackpot.

>1. Set up a machine that will "destroy" the IP value of something by copying it.

>2. Make it do so a billion times, and watch if the company whose IP being "infringed" goes bankrupt.

Did you, you know, actually READ anything I wrote, including the very first sentence that you yourself quoted? Because what you just described is information replication, which as I wrote in, again, that very sentence you quoted, is effectively free. It does not cover the step 0 of GENERATING that information to be copied in the first place. That's the problem part, which IP attempts to solve by hacking on synthetic limits on the unlimited part of the equation to handle allocation to the limited part.

>What was the last "big thing" an American company ever came with?

Basically the entire semiconductor and software industry? Most modern medicines? Much of modern genetic technology? Reusable rockets? Like, are you for real?

>and it was more than 60 years ago, when no monstrous IP protection regime was in place.

Uh, yeah when it comes to patents yes, IP is about the same as it was 60 years ago, beyond the Republicans continually trying to cut the budget of the PTO. That's part of the problem in fact.

I keep wondering if you're being sarcastic or engaging in satire here yet you seem to be entirely serious, but it's utter nonsense stuff.

> Um, no. It's objective fact. Whether it's a video game or R&D on drugs or semiconductors, it does in fact require the expending of limited resources. Are you for real? This stuff doesn't pop into existence out of thin air.

You can do a lot of normally valueless things for money, if somebody gives them to you for that, whether by legal compulsion or own volition. Not denying that. The question is whether giving money for that is a rational thing to do.

>It does not cover the step 0 of GENERATING that information to be copied in the first place.

I put that example exactly to demonstrate that it does not influence whether the person/business in question has his capacity "of GENERATING that information" diminished.

No, that person or a business is 100% capable continuing doing so in the same capacity.

The capacity "of GENERATING that information" gets diminished if it gets regulatory limited though IP monopoly (I copyrighted this mathematic formula/program, thus nobody else can use it in their mathematic formulas/computer programs)

China for example has its own non-insignificant pharmaceutical industry making original drugs, low profile semiconductor industry which nevertheless makes own developments, and gigantic videogame companies all with close to none IP enforcement.

About encouraging growth or not. I'm not sure, it would be interesting to see what would happen without patents. However, I'm reasonably sure scientific startups wouldn't get VC funding as easily.

Don't get me wrong, it's totally possible that this is a good thing, and it might make it easier for companies to bootstrap if they don't need to compete with VC funded startups. But it's a very different model than we have now.

But...I don't think it's obvious that patents don't encourage growth. It's possible that they do. They do encourage disclosure to an extent (I've learned a lot reading patents). Without patents, you might find companies relying on trade secrets more.

Do you trust the patent examiners to set the validity period? I don't particularly to be honest...

I would think that setting a fast increasing tax (say doubling each year) on patents would solve lots of issues. Once you do not want to pay the tax, the patent moves to public domain. Same tax should apply also to copyrighted work.

Shortening patents (when it can take years to gain traction) or making them more expensive only plays into the hands of large companies

As someone who researches at a western company, this would actually get rid of a lot of research, because business moves so slow to commercialize due to the highly regulated environment.

Regulation and legal arbitrage, where companies (particularly in Asia) can take what is newly invented on a silver platter and sell to the global market due to lack of environmental protections etc. would be all that is left.

» As someone who researches at a western company, this would actually get rid of a lot of research, because business moves so slow to commercialize due to the highly regulated environment.

I don't know about you but I'm perfectly ok with that.

I like technological progress, but I would also be okay digging ditches in case our economy crashed towards a commodities-only enterprise, so I’m strange in that regards :)

Some people think patents are required for innovation. They are mistaken. It is human nature to innovate.

I agree, but it is not human nature to take their innovation to the world. Post WW2 the US govt had tens of thousands of technology patents but no one to bring it to the world.

Patents are risk mitigation tools for funding entities to actually support commercialization. Innovation will happen always, but most of it stays on a dusty shelf a la an academic library. It’s not necessary that anything interesting and innovative (and sufficiently complex) gets sold.

Patents exist to prevent the sale of innovations. The notion that I can't make a buck without a patent because someone else will undercut me may be true, but it also says that innovations will get to market without patents.

Yes, the enforcement is the manifestation of their use as a risk mitigation tool. Some innovations will get to market without patents, complex innovations requiring some threshold dollar amount will not find funding and will not (this was learned by the world of govt. science policy over the last century).

It is, and people would keep coming up with ideas.

It would get a lot harder to put together a few hundred million dollars in funding to build the vacuum chambers, ovens, lithography machines, and other things you need to actually test your idea and get it working, though.

I'm sure a national emergency declaration would resolve an issue like this.

Reply on news.ycombinator.com