Hacker News Reader

Comment by ramshanker | Hacker News Reader

I am of the opinion that instead of letting legaleses define good/bad patent, we should let the economics work it out. Mandate patent filers to declare a public flat license fee . Along with annual recurring patent-maintenance tax to government as some X% of that declared fee.

Market will decide which patents are worth maintaining for 20 years, and government get proportional revenue as well.

View on HN

Your solution seems to be addressed at solving a "there are too many patents because they're cheap to acquire" problem, but that's not a problem. Patents are expensive to acquire already. Bad patents don't lack value. Rather, they offer insane value for little effort if you guess the right thing to patent.

Look at the EFF's example. Say it's 1987, you've been playing with AOL for a bit, and you realize that someone might want to use a computer to do stuff like ordering pizza deliveries. You file a patent where your "invention" is "okay it's like pizza ordering on the phone but you do it with a computer instead." Then you pass it through a lawyer who draws up some diagrams to make it look like it's an invention, and boom, patent US4797818A is granted and you now control the entire online food delivery ecosystem until 2007. Oh, but then someone else says "wait, what if the customer does that but USES A PHONE." Well, now they own that for 20 years, so now you both can arbitrarily limit this space. Great. Oh, but now swomeone says "oh, but what if you do this and also the food is displayed IN CATEGORIES!" Well now he's got a patent, too.

It's madness. The problem isn't that it's too cheap to acquire and maintain these. The problem is that these are so obvious that they are no longer inventions.

One issue is that you don't have to have a prototype to get a patent. Except perpetual motion machines, too many people tried.

For the example given though, it doesn't seem too hard to prototype a "order pizza using phone" app. It might add a 10k expense or whatever to the patent, but that's small relative to how much money you can extract from all pizza chains in the US.

My assumption is if you had ro have a prototype it would be more likely to discover prior art.

That puts all the power over patents in the hands of those with deep pockets. They would be able to directly pay to prevent competition. It's insane to eliminate limits on what could be patented while doing that.

They could fund the keep-away of basic things like "ordering x on your phone" with the proceeds from being the only actor allowed to do that. Fundamentally exploitable system incentives.

Or we could just abolish the concept of patents entirely. All they do is stifle progress so that somebody can make money at an unfair advantage to everyone else.

>All they do is stifle progress so that somebody can make money at an unfair advantage to everyone else.

You don't see any problem with pouring huge sums of money into developing something, only for a competitor to copy your idea and undercut you? As an inventor, why would I bother putting in all that effort knowing that would happen?

Same reason you would do anything else. Because regardless of whether you have a monopoly, you can still make money by solving people's problems

If you're going to invest in making something possible, you need to recover at least the value of the resources you expended in doing so. Otherwise, you'll have no resources left for the next investment. Copyright and patents are capitalism's attempt to make investments possible in a world where copying is cheap, and while they're deeply flawed, that doesn't change the fact that investments do need to be returned.

(Don't get me started on the stock market: that has nothing to do with actual investment.)

>(Don't get me started on the stock market: that has nothing to do with actual investment.)

Secondary markets support "actual investment" by allowing early investors to cash out. An investment is far more attractive to you if you know you can offload it to someone else and get the net present value immediately, rather than waiting years/decades for it to pay out. For instance, if a nuclear plant takes will last 80 years but will take 50 years to break even, why would a 40 year old bother to invest? By the time it breaks even you'd be long dead.

At the cost of so deeply skewing what an "investment" is. How much destructive nonsense has been driven by the stock market? The South Sea Bubble, and every bubble since. Short-sighted decisions to meet quarterly revenue targets, at the expense of all else. What TeMPOraL calls "fracking": https://news.ycombinator.com/item?id=34297430. I would not call any of this investment, but the stock market would.

If they require disclosure of enough information for the technology to be (in principle) independently reproduced, patents are worth it. I'd even argue that patent terms could be extended, in some domains, where that's necessary to ensure a return on investment. Copyright is, in principle, worth it, though the term is currently far too long and the laws can be abused for unrelated things (e.g. DMCA ink cartridges).

But the stock market is not worth it. The invisible hand of the market can remain irrational far, far longer than you can remain healthy and fed with a roof over your head, surrounded by friends and beautiful things. Let us not sacrifice what is good at the altar of the invisible hand.

The real estate market is similar: housing is for living in, not for speculative investment. The market incentivises real estate owners to ensure there aren't enough houses to go round, driving up house prices – but they usually forget that they can't actually sell their only property, so this only benefits landlords. (Irrational behaviour, sure, but whoever said humans were rational was a 20th-century economist who never read Adam Smith.)

Large infrastructure developments used to be handled at the community level, and in many places, they still are. For something as large as a nuclear power plant, that'd be your regional council or your national government. Hundred-year investments have to be made at that scale, anyway: that's what your secondary market proposal is about, after all. The stock market is only one (highly-flawed) way of organising things.

People do not build sheds in their gardens because they expect to be compensated: they build a shed because they want a shed. People do not work because they expect time and labour to be returned to them: they work because they get other things out of it. We should not assume that investment of resources yields fruits in kind – but the stock market cannot but make that assumption!

>If they require disclosure of enough information for the technology to be (in principle) independently reproduced, patents are worth it.

How does this work for "obvious" patents like "[x] but on [y]" that the sibling was talking about[1]? I think that's the patents that attract the most ire, rather than pharmaceutical patents that don't allow you fully reproduce the drug or whatever.

[1] https://news.ycombinator.com/item?id=39763707

>But the stock market is not worth it. The invisible hand of the market can remain irrational far, far longer than you can remain healthy and fed with a roof over your head, surrounded by friends and beautiful things. Let us not sacrifice what is good at the altar of the invisible hand.

So basically what you're saying is "people can't be trusted to price things correctly, therefore we should ban them from being able to engage in exchange of assets?" This is, to put it mildly, "throwing the baby out with the bathwater".

>Large infrastructure developments used to be handled at the community level, and in many places, they still are. For something as large as a nuclear power plant, that'd be your regional council or your national government. Hundred-year investments have to be made at that scale, anyway: that's what your secondary market proposal is about, after all. The stock market is only one (highly-flawed) way of organising things.

This is eventually going to break down to "central planning vs free markets", which has been litigated to death elsewhere and probably isn't going get resolved by two HN commenters.

>People do not build sheds in their gardens because they expect to be compensated: they build a shed because they want a shed. People do not work because they expect time and labour to be returned to them: they work because they get other things out of it. We should not assume that investment of resources yields fruits in kind – but the stock market cannot but make that assumption!

The shed example actually shows a lot of problems with your line of thinking. Sure, for a project like a shed in the typical case, you can be reasonably assured that whatever utility can get out of the shed (eg. 15 years of storage) exceeds the cost you put in (eg. $1000 and a weekend's worth of work). However, that's not guaranteed to be the case. For instance, you might not be able to use the shed for the full 15 years, eg. if you're planning to start a family and need to move in 2-3 years to a bigger home[2]. If that's the case, the economy (as a whole) would still presumably benefit from the shed getting constructed. Future occupiers of the property presumably would have storage needs as well, for instance. You might not be able to use the shed enough to recoup all the resources you put in, but other people will. However, the shed doesn't end up getting constructed because the person paying for the construction can't capture all the utility. In a world with markets this wouldn't be a problem, because if the shed is something that other people would appreciate, it will boost the value of the home, which would allow the current owner to capture the full value of the shed even if he moves out in a year.

[2] you might argue that sheds are such a slam dunk proposition that even using it for 1 year would justify the construction cost. Even if we grant that's the case, not every investment opportunity in the economy are slam dunks, and arguably we shouldn't restrict ourselves to investing in only slam dunks.

If you are unable to make efficiently products based on your ideas, for the entire society it is much better if someone else is able to do efficient production.

The reason why the Western countries have achieved their dominant position in the modern world is because at some point, a few centuries ago, the secrecy of the manufacturing guilds has been overridden by the new trend of publishing openly any new discovery in science and technology, while the countries like China, which have remained dominated by secrecy, have remained stagnant during the fast evolution of the European countries.

The Silicon Valley has become what it is because in the early days of the electronics industry the concerns about "IP" were minimal, every new discovery was published and it spread quickly among all companies located there.

Then gradually the "IP" protection has increased, the competition has diminished more and more and the pace of innovation has slowed down proportionally.

It is extremely rare that those who have the original idea for a patent are also able to discover the follow-up innovations that are necessary to convert an idea into a successful product.

There are very numerous examples of patents which have been almost never used in anything useful as long as they were valid, but immediately after expiration many companies were able to make good products based on them.

> If you are unable to make efficiently products based on your ideas, for the entire society it is much better if someone else is able to do efficient production.

What happens when the idea in question takes a non-trivial amount money to develop, and you want to avoid a prisoner's dilemma situation where nobody wants to put in the investment?

This is such egocentric/selfish thinking. I wish to live in a society which “pours” adequate sums of money into developing things for the benefit of all, not leaving it up to individuals like you who are in it only to benefit themselves.

I declare that I patented human breathing. Every human who is breathing is to pay me the, really very cheap, fee of 1 cent per day. O there is a tax? So now I only get half a cent per day per living being.

You probably joking, but that wouldn't work because of https://en.wikipedia.org/wiki/Prior_art. All the "bad" patents were probably novel at the time they filed. That's not the issue. The issue is that they're for arguably obvious stuff like "[x] but on internet/phone" that the sibling commenter is talking about.

Read the comment that I am responding to.

'letting legalese define good/bad patent, we should let the economics work it out.'

If 'legalese' cannot define good/bad patent anymore, there is nothing stopping such a silly patent that I propose.

But some companies evolve on top of their patents and use them to obtain a competitive moat over a period of time to outcompete larger players. If the patents were available for a flat license fee, you’d penalize patent innovation by smaller entities as bigger entities could pay the flat license fee and copy the patent. So smaller players will either just keep it a trade secret and hope that no one finds out (which theoretically hurts society due to lack of published secrets) or have to hope they can monetize their patent super fast & find someone with really deep pockets to fund that work.

And as mentioned, patents are super expensive. From talking to some people it varies significantly depending on your details but probably starting is ~10-20k per patent.

While that's an interesting thought experiment and there are definitely good reasons to make it more difficult to sit on an unused patent, it's already difficult enough for lone inventors and small organizations to get and keep a patent. This would require everyone to either have a lot of money up front to pay the maintenance tax or keep the license fees low thus ensuring that they will never have a lot of money.

Make sure keeping a patent is prohibitively expensive. Start with some trivial price y, like say a dollar. Then make every year cost equal to `(y*x)!` where x is the year/month/time period and `!` is the factorial function.

Any patent worth the upkeep for first 10 years is probably worth keeping it.

How does this work for patents like drugs or whatever, which are very expensive to develop/bring to market?

Medical r&d should be done by the government anyway. There's no reason to allow pharma companies to extract untold billions of dollars in profit from the economy every year when the involved "research" is either trivial (such as with the insulin racket) or financed by the government anyway.

>when the involved "research" is either trivial (such as with the insulin racket)

source that "the insulin racket" was "trivial" to research? Yes, insulin was discovered more than a century ago, but that's not the thing that's patented today. All the expensive insulin are enhanced versions that are better in some way (eg. lasts longer, reacts sooner). If you want the old, crappy, and non-patented kind, you can get it at walmart for less than $25.

>financed by the government anyway.

Basic research that the government finances doesn't cover the billions of dollars needed to bring the drug to market.

Then it works as intended, the pharma company at first enjoys a cheap patent, but gatekeeping the drug for years, until way after the r&d costs are recouped from sales, becomes prohibitively expensive.

If the drug doesn't sell a lot and is never profitable, then the patent gets dropped earlier to avoid sinking more costs.

If the drug cures cancer and makes millions, the company can easily recoup its cost up until the upkeep is too much, at which point surely it would be better for society that the drug be open for everyone?

Pharma companies are massively profitable, the argument of "think of the poor pharma companies doing r&d and their costs uwu" isn't really valid.

Reply on news.ycombinator.com