Page 2 of 2
Re: open source
Posted: Thu Mar 01, 2012 7:07 am
by Jake
KevinW wrote:
Thanks for the links, Jake, I'll take a look.
You're welcome. I also did a podcast interview a while back with the author, Stephan Kinsella about his book. Here's the link
http://thevoluntarylife.blogspot.com/20 ... la-on.html
Re: open source
Posted: Thu Mar 01, 2012 9:45 pm
by AgAuMoney
stone wrote:
the author doesn't seem to grasp that time is something that has scarcity for every individual. So if someone spends ten years writing a series of books, they have forfeited something just as tangible as if they had sold the family farm. Saying that copying those books without attribution causes no conflict because there is no scarcity seems a bit iffy to me. Compare an author charging royalties from selling a book to a fisherman selling fish he caught in the sea. Both are simply demanding payment for their prior labour.
If it is the author's scarcity we are protecting, then libraries and lending and the 1st sale doctrine all need to be voided. In fact, you shouldn't be able to read a book twice without paying twice.
Instead we have decided that the author deserves something, for a limited time.
Intellectual property is very different from tangible property, where if I take your tangible property I deprive you of the use of that property. Taking intellectual property does not deprive anyone of the use of that property.
Some will argue that protection and ownership of ideas is what got us the huge advancement starting with the industrial revolution. From my perspective as a software engineer I do not believe that. Instead I attribute the success to communication of ideas rather than ownership of ideas. This allowed more people to build on the ideas of others rather than starting from scratch.
Would companies still invest money if they could own their ideas? Yes, they would and do because they do so today. Very few of the dollars spent in research and development result in a patentable idea, and very few patentable ideas end up with any economic value in and of themselves. In the tech world today patents are primarily used as a "war chest" to have something to trade when a competitor uses his war chest to limit your ability to compete, or visa versa.
Re: open source
Posted: Fri Mar 02, 2012 8:46 am
by stone
If copyright and patenting is wrong, is it fair to say that some other model is needed to ensure authors and inventors get rewarded? I suppose part of why patenting was brought in was so as to ensure early disclosure of the invention and rapid commercialization. I suppose they wanted to avoid people sitting on secret inventions. In a world without copyright or patenting, it seems hard to imagine how there could be professional authors or a private sector pharma industry. Maybe that doesn't matter?
I've wondered whether in the case of the pharma industry whether the best approach would be to pay them not via drug sales but instead by awarding prizes for inventing drugs that provide a benefit. Once the prize was awarded, anyone could make and give the drug with no commercial restriction. That would disconnect drug discovery from drug manufacture and marketing. To my mind drugs should not be marketed. The best medical practice should be chosen by physicians based on trial data not influenced by marketing.
Perhaps music and written work could be freely available and some sort of fund used to pay creators based on how prevalent reading or listening to their work was? Again it gets messy when work heavily samples previous work.
Re: open source
Posted: Fri Mar 02, 2012 12:36 pm
by KevinW
I'll take a listen, and check out the rest of your blog.
AgAuMoney wrote:
Intellectual property is very different from tangible property, where if I take your tangible property I deprive you of the use of that property. Taking intellectual property does not deprive anyone of the use of that property.
Yes, in economic terms, information is a nonrival good:
http://en.wikipedia.org/wiki/Rivalry_(economics)
When an individual can access useful information they are enriched. In aggregate, a society that fosters a free flow of useful information becomes wealthier. Since this flow does no harm to the person with the first copy (nonrivality), it's a win-win. I agree with Richard Stallman's point that prohibiting this kind of beneficial sharing is morally wrong and people on the side of prohibiting it are the ones that have the burden of justifying their position.
stone wrote:
If copyright and patenting is wrong, is it fair to say that some other model is needed to ensure authors and inventors get rewarded?
Yes. There are arrangements that compensate creators without resorting to IP law that prohibits beneficial sharing.
There's the patron model, where benefactors commission works. That was responsible for producing classical music which is all public domain. Performance artists can charge for live performances/showings and merchandise but give away recordings (e.g. Radiohead, Phish, many indie bands). Creators of important open source software can become gurus with many employers bidding on their services (e.g. the creators of Linux, Python, Apache, and others). There's the honor/tip-jar system --- if you download it please pay me $5 (Lewis CK, Radiohead). If you make a name for yourself people will buy your stuff even though it can be copied; this works in high fashion, architecture, and restaurants where there is no IP law (or very little) yet consumers still pay a premium for name-brand goods.
IMO the argument "if IP law goes away then no one will want to create anything" comes from the lobby of the current batch of media companies. You don't hear it very often from actual writers, musicians, software engineers, etc.
Re: open source
Posted: Fri Mar 02, 2012 1:21 pm
by stone
Kevin W
Yes. There are arrangements that compensate creators without resorting to IP law that prohibits beneficial sharing.
There's the patron model, where benefactors commission works. That was responsible for producing classical music which is all public domain. Performance artists can charge for live performances/showings and merchandise but give away recordings (e.g. Radiohead, Phish, many indie bands). Creators of important open source software can become gurus with many employers bidding on their services (e.g. the creators of Linux, Python, Apache, and others). There's the honor/tip-jar system --- if you download it please pay me $5 (Lewis CK, Radiohead). If you make a name for yourself people will buy your stuff even though it can be copied; this works in high fashion, architecture, and restaurants where there is no IP law (or very little) yet consumers still pay a premium for name-brand goods.
IMO the argument "if IP law goes away then no one will want to create anything" comes from the lobby of the current batch of media companies. You don't hear it very often from actual writers, musicians, software engineers, etc
I guess the patron system in the past has been funded by a very few people who sometimes had distinctly dubious sources of income (eg the Medici selling indulgences or whatever). I guess though in principle it could expand to include a broad swathe of the population.
Would pharma be fundable in that way? Would it work if people only voluntarily gave money to drug discoverers?
Re: open source
Posted: Fri Mar 02, 2012 2:49 pm
by KevinW
stone wrote:
I guess the patron system in the past has been funded by a very few people who sometimes had distinctly dubious sources of income (eg the Medici selling indulgences or whatever). I guess though in principle it could expand to include a broad swathe of the population.
That's true, but I think those examples are more of an indictment of the Medicis themselves and the inequitable system of socio-economic justice they operated within, than the patron system itself.
Kickstarter is a more modern, distributed, transparent take on patronage:
http://www.kickstarter.com/
stone wrote:
Would pharma be fundable in that way? Would it work if people only voluntarily gave money to drug discoverers?
"Necessity is the mother of invention" and I think, if actually pressed, pharma would come up with a way of funding R&D. As was already noted, the automotive industry, which is similarly dependent on R&D capital, has operated under an IP cease fire for about a century. Perhaps one solution would be to use the system of competitive grants that government research agencies use to semi-privatize basic research. E.g. people who want heart disease drugs (patients, American Heart Association, health insurance companies, etc.) could pool their money and offer up a $100M (or whatever) contracts to improve the state of the art, and award them to whoever has the most promising proposal. That'd probably lead to a division between speculative R&D firms and retail-oriented distributors. Which works fine in other industries (e.g. electronics).
Re: open source
Posted: Fri Mar 02, 2012 4:08 pm
by stone
Kevin W
I think, if actually pressed, pharma would come up with a way of funding R&D. As was already noted, the automotive industry, which is similarly dependent on R&D capital, has operated under an IP cease fire for about a century. Perhaps one solution would be to use the system of competitive grants that government research agencies use to semi-privatize basic research. E.g. people who want heart disease drugs (patients, American Heart Association, health insurance companies, etc.) could pool their money and offer up a $100M (or whatever) contracts to improve the state of the art, and award them to whoever has the most promising proposal. That'd probably lead to a division between speculative R&D firms and retail-oriented distributors. Which works fine in other industries (e.g. electronics).
Would that approach be a bit like the "prize" way of funding that I was wondering about:
I've wondered whether in the case of the pharma industry whether the best approach would be to pay them not via drug sales but instead by awarding prizes for inventing drugs that provide a benefit. Once the prize was awarded, anyone could make and give the drug with no commercial restriction. That would disconnect drug discovery from drug manufacture and marketing. To my mind drugs should not be marketed. The best medical practice should be chosen by physicians based on trial data not influenced by marketing.
To my mind it is very important that funding comes AFTER drugs are discovered and shown to be effective. If instead of "prizes", funding was in the form of "research grants" then I think that would be a disaster. I think capitalism needs to always be on a "stupid is as stupid does" basis. What matters is what actually gets achieved. Otherwise you just run into the danger of croneyism however well intentioned everyone involved may be. To my mind for some endevours (especially mundane but vital ones) real personal wealth needs to be on the line to ensure things get done effectively.
Re: open source
Posted: Fri Mar 02, 2012 7:10 pm
by KevinW
Prizes work too, but are less economically efficient. You wind up with multiple teams competing for the same prize, and only one gets paid. I think that's well suited to research that's relatively quick, and where the next step is unclear and it helps to have many teams trying a diversity of avant-garde approaches all at the same time. E.g. the DARPA Grand Challenge. For longer-term incremental research, prizes seem like too big a waste and too big a temptation for shenanigans. Imagine if three rival companies had each spent $3 billion and 5 years working on the same project and only one of them gets the $4 billion prize...that's $6 billion wasted and there's a $4 billion bounty on influencing the selection process.
The accountability mechanism for grants is that the outcomes of past grants are considered in new applications. So if a researcher fails miserably they get blackballed from future grants. If they misuse funds they're liable for damages as a breach of contract. Not perfect but IMO sufficient to make abuse very rare.
Re: open source
Posted: Sat Mar 03, 2012 3:10 am
by stone
KevinW, I'm not really meaning malicious abuse but rather simply inadvertant ineffectiveness. There is a lot of controversy about funding accedemic research via research grants for specific proposals. Howard Hughes Medical Institute (HHMI) has decided not to do that and instead to fund "people not projects" because they have the humility to recognize that the proposal appraisal system is unable to pick winners. To my mind it would be impossible to extend that HHMI approach to commercial pharma with funding for "corporations not projects". Drug discovery is a very extreme example of human endevour because it is so unlike say building a road. It entails a tiny chance of making a massive hit. It is more extreme than buying lottery tickets and yet is not just luck. If it is done wrong there is zero chance of winning however much is staked.
To my mind the vital thing is that as and when someone gets a notion about how to create an effective drug, that is when it should be able to take off. I don't think it is something that can be solved simply by throwing money at the problem in a narrowly directed way from the outset. The right pieces of the jigsaw have to be at hand. If you say that a $50B grant is available to "find a cure for cancer" then ofcourse proposals will be submitted, staff will be employed and reagents will get mixed together and go to landfill. To actually get a cure, I think a wide base of non-applied exploratory research and knowledge infrastructure (eg genome sequences, databases etc) is needed along with a source of finance for following any realistic drug discovery leads if and when they crop up. I was thinking that if prizes were available for drug discovery with say a certain award for a certain "quality life year" gain then that would entice finance for realistic drug discovery leads to be taken forward. I agree that if two similar drug discovery leads are brought forward together, then it is hard on whoever finishes second but that is true for everything and does help to keep things moving. The economic efficiency argument you put forward perhaps should be viewed against the experience of command economies such as the USSR. Efficiency seems to require a lot of individuals to each have the freedom to try and get things done in their own way.
There is fascinating look at the drug industry in
http://www.bbc.co.uk/radio4/features/in ... /20101209/
The then boss of GSK says, "The critical mass of discovery is the size of one human brain, and I really believe it".
Re: open source
Posted: Sat Mar 03, 2012 1:21 pm
by KevinW
From the outside looking in, it seems kind of ridiculous that the only way of finding drugs is through massive trial and error. That's not really how any other science-based product development works. You don't make a bridge by building 1000 that fall down before you stumble on the 1 that stays up.
Maybe instead of doing more and more trial and error they ought to be working on turning the process into more of a linear design process. Or if trial and error is inescapable, maybe they could figure out how to automate it. Trial and error is cheap when a computer or robot is doing it.
Anyway you might be right that paying after works better than before. I guess I still don't think it'd make a huge difference in practice.
But I think all this strengthens my original claim, that there are viable ways of funding pharma even in the absence of IP law.
Re: open source
Posted: Sun Mar 04, 2012 3:12 am
by stone
KevinW, I'm also a baffled outsider looking in. I've some familiarity with it because my better half works in the molecular diagnostics industry which is connected to the drug discovery industry and I'm a biologist so it seems less exotic to me than say software does. The pharma industry does use lots of robots etc but its still expensive.
I think say a road bridge can be designed rationally because all of the forces and materials are well understood. Biology is so full of unknowns that linear design is typically a shot in the dark. I totally agree though that it is vital that people look out for openings where a straightforward approach could work rather than blindly hammering away at brute force screening.
I've used trial and error based approaches to produce lab reagents and it can work nicely in the right circumstances. It can be possible to select the one functional molecule out of the 10 000 000 "best candidates" in a situation where no one really has any idea why that one molecule works any better than all of the other "best candidates". I suppose life itself came about by such a trial and error but obviously on a much grander scale. Is that approach not used much by software engineers? I looked at the wikipedia site
http://en.wikipedia.org/wiki/Evolutionary_algorithm but it doesn't seem to give examples of practical examples made that way. I sort of presumed that more tricky software such a those self driving cars would be made that way but I have no idea.
KevinW wrote:
From the outside looking in, it seems kind of ridiculous that the only way of finding drugs is through massive trial and error. That's not really how any other science-based product development works. You don't make a bridge by building 1000 that fall down before you stumble on the 1 that stays up.
Maybe instead of doing more and more trial and error they ought to be working on turning the process into more of a linear design process. Or if trial and error is inescapable, maybe they could figure out how to automate it. Trial and error is cheap when a computer or robot is doing it.
Anyway you might be right that paying after works better than before. I guess I still don't think it'd make a huge difference in practice.
But I think all this strengthens my original claim, that there are viable ways of funding pharma even in the absence of IP law.
Re: open source
Posted: Mon Mar 05, 2012 11:37 pm
by KevinW
stone wrote:
I suppose life itself came about by such a trial and error but obviously on a much grander scale. Is that approach not used much by software engineers? I looked at the wikipedia site
http://en.wikipedia.org/wiki/Evolutionary_algorithm but it doesn't seem to give examples of practical examples made that way. I sort of presumed that more tricky software such a those self driving cars would be made that way but I have no idea.
In software it's necessary to distinguish between the steps made by the human software engineer ("process") and the steps performed by the computer ("algorithm"). I've never heard of a successful use of brute force as a software design process, but many practical algorithms are based on brute force:
http://en.wikipedia.org/wiki/Brute-force_search
In other words the human designers decide to use brute force and implement it, which is a straightforward task taking perhaps one person-week of work. And then the computer evaluates all the candidate solutions, which might take a matter of minutes on modern CPUs which execute billions of instructions per second.