Not logged in
Log in now
Create an account
Subscribe to LWN
LWN.net Weekly Edition for May 16, 2013
A look at the PyPy 2.0 release
PostgreSQL 9.3 beta: Federated databases and more
LWN.net Weekly Edition for May 9, 2013
(Nearly) full tickless operation in 3.10
This is why large corporations support patents even when they are the ones getting sued most often.
Patent lawsuits are less of a threat to their profits then free and open markets.
Mueller: Apple's iterative approach to FRAND abuse is not for the faint of heart
Posted Feb 6, 2012 21:12 UTC (Mon) by hitmark (guest, #34609)
It also demonstrates how much of a pipe dream the idea of a unregulated market really is.
Posted Feb 6, 2012 22:11 UTC (Mon) by jcorgan (subscriber, #47213)
Heh, imagine that--someone thinking that patents, which are government-granted monopolies, are part of an "unregulated market."
Posted Feb 7, 2012 0:33 UTC (Tue) by dgm (subscriber, #49227)
Posted Feb 7, 2012 1:01 UTC (Tue) by bojan (subscriber, #14302)
Posted Feb 7, 2012 1:36 UTC (Tue) by drag (subscriber, #31333)
No it is not. Being able to buy and sell and change prices based on demand is just the definition of 'market'. It does not indicate whether or not there is regulation present. You seem to be ignoring the definition of your adjective (at least in the context), or are using it in a bizarre manner.
> and the fact that they are government-granted is irrelevant.
No... the fact that they are 'government-granted' means that it's regulated. It is _extremely_ relevant.
Saying that government granted monopolies on the ability to distribute and use software in a marketplace is not regulation on software marketplaces because you have the ability buy and sell these monopoly rights just doesn't make a lot of sense.
I am very sorry, but the level of absurdity here makes it difficult to even refute. One has a hard time even figuring out were to start!
Posted Feb 7, 2012 4:06 UTC (Tue) by ghane (subscriber, #1805)
Das ist nicht nur nicht richtig, es ist nicht einmal falsch! -- http://en.wikipedia.org/wiki/Not_even_wrong
Posted Feb 8, 2012 10:37 UTC (Wed) by alecs1 (guest, #46699)
Patents are a form of regulation
Posted Feb 6, 2012 22:15 UTC (Mon) by david.a.wheeler (guest, #72896)
Posted Feb 7, 2012 0:21 UTC (Tue) by drag (subscriber, #31333)
Yes. That's absolutely correct.
Patents are, by definition, monopolistic legal instruments. The central government hands out special privileges based on the fact that a person is able to be the first one to pay their fees and navigate the government bureaucracy. It's a money game and the person who pays the most money, has the most lawyers, is able to get a huge advantage.
In fact it is even less legit now then it is in the past. President Obama, relatively recently, signed a bill* that got rid of the first-to-invent system of deciding patent awards and replaced it with a pure first-to-file system. This streamlines the system (and in a legal sense it is a superior system), but all it really amounts too is that even if you did invent something first you can get your sued if you didn't happen to think that it wasn't important to pay the government off.
*(this bill was signed in 2011 and will go into effect March 2013)
Free and open software proves that it is not necessary to have market intervention in this way. In fact it probably proves that it is damaging. Every time free software progress bumps into this type of regulation it retards progress and wastes valuable resources as significant effort must be expended to work around this or that road block or jump this or that hoop to keep the software free.
Posted Feb 7, 2012 1:40 UTC (Tue) by xaoc (guest, #54140)
Except if you have proof of your prior art it becomes just that 'prior art'.
Posted Feb 7, 2012 1:59 UTC (Tue) by dlang (✭ supporter ✭, #313)
the old patent system required that prior art be at least two years before the filing date. I haven't looked in to what these changes do to this.
Posted Feb 9, 2012 17:05 UTC (Thu) by Wol (guest, #4433)
I gather the new law didn't alter the prior date in the US, so their new patent system now combines the worst bits of prior art from their legislation and elsewhere.
"First to file" is far better than first to invent. It gets rid of a lot of argument and fraud about when something was invented. And in most instances I believe trade secrets are protected - if somebody else comes along and patents what I've been doing for years, they can't sue me. But if I didn't publish they do have a valid patent and can sue anyone else.
The whole point about "first to file" is not that the first person gets a patent. It's that if there is NO PRIOR ART then it assumes that the person who gets his application in first is the true inventor. Any previous publication (including by the applicant!) is classed as prior art and will block the patent.
Most patents are drag on the innovation
Posted Feb 7, 2012 0:22 UTC (Tue) by khim (subscriber, #9252)
Actually most patents fail to promote the Progress of Science and useful Arts. The only kind of patents which actually make sense are pharmaceutical patents - and this is because there are yet another government-regulated institution: drug certification. Companies spent hundreds of millions (and sometimes billions) to get certificate - and without patent there will be no way to recoup these costs.
Few will argue that unregulated sales of just-invented drugs will be beneficial to society and that means that pharmaceutical patents make sense (even there it's not clear if it's the best way to solve certification issue, but at least it's clear where money you need to later recoup with patent license go to). All other patents just help incumbents in fight with newcomers. More information in a book.
Posted Feb 7, 2012 3:09 UTC (Tue) by wahern (subscriber, #37304)
For example, in the book you cited, the authors note that they believe it would be more efficient for the government to simply pay for tests directly, rather than using an indirect system of taxation, viz patents. But you could have that independent of drug controls. The only regulation you'd need is for clear labeling, such as labels which say "FDA tested and approved for XYZ condition" and the like.
That would dramatically bring down the cost of pharmaceutical treatments. Theoretically the overall incidence of harm in the population could go up; but my guess is it that it would go down. It's hard to put numbers on such things anyhow, because of hidden abuses in the current system, as well as coming up with reliable and credible counter factuals.
Posted Feb 7, 2012 13:43 UTC (Tue) by Cyberax (✭ supporter ✭, #52523)
Drug design and testing is HARD. It's not uncommon for a very promising drug to fail the first or the second phase of the clinical trials, even though drug designers sincerely believed that it should work. No amount of transparency can fix this, alas.
Posted Feb 8, 2012 12:16 UTC (Wed) by nix (subscriber, #2304)
We know this because we see it happening in countries where regulation of pharmaceutical sales is relatively unregulated. Heck, if it happens with *alcohol*, whose imbibing is never a life-or-death matter, and which is a heck of a lot easier to manufacture ("no open flames, keep the methanol out" being the only real human-safety rules), it sure as anything would happen with pharmaceuticals too.
Software routinely critical for life and limb needs regulation too, and is regulated (e.g. medical device software, avionics software).
Posted Feb 9, 2012 1:17 UTC (Thu) by rahvin (subscriber, #16953)
And that's AFTER computer models, rat studies, primate studies and early human testing, a process that typically takes a decade to even get to the phase 3. Current estimates are that including the Phase 3 clinical trial a drug from discovery to post phase 3 costs $250 million (yes a quarter billion dollars). This doesn't include factory certification, production line certification, ongoing inspections and quality control or long term studies (becoming far more common due to the Vioxx scare).
And that's just the US. Then you do the same thing for Europe, Japan and pretty much every other country in the world if you want to sell the drug there.
Getting a drug to market is HARD. Last numbers I saw said less than 1% make it to market. So on top of that $250 million dollars you probably have another billion or two in failures that also has to be recouped by the successful drug.
As others have said, drug patents make sense. They make a LOT of sense given the government regulation. Now if you want to suggest that companies can drug test on the unsuspecting population I suggest you line up first.
Posted Feb 9, 2012 18:54 UTC (Thu) by dmarti (subscriber, #11625)
There's no clear-cut division between drugs that work and drugs that create risks. Example: Tegaserod (Zelnorm®).
Yes, it might kill you, and it's risky enough that the FDA had it pulled from the market. But for some people it was the only thing that ever let them go anywhere without an immediately available toilet. Would you take it? Would you give it to someone without informed consent? Should an informed adult be able to take it?
Posted Feb 9, 2012 19:46 UTC (Thu) by Wol (guest, #4433)
When they manufactured thalidomide for testing purposes, they got pure L-thalidomide. Which is safe. When they scaled up, the process made 50/50 L- and R-thalidomide (which are chemically identical). It was the R-thalidomide that did the damage.
So if you gave me a "bottle of thalidomide, perfectly safe for pregnant women" it's quite possible you're telling the truth ...
Posted Feb 9, 2012 22:46 UTC (Thu) by Cyberax (✭ supporter ✭, #52523)
However, it's a great and simple example of reasons beyond the extensive drug testing which begins with the toxicity tests, and with the compound which will later be certified and sold.
Posted Feb 10, 2012 13:23 UTC (Fri) by Wol (guest, #4433)
But it's normal to get a pure L or R isomer or a racemic mix, depending how you produce it. And it's quite likely, similar to what you assert, that when they produced commercial quantities one of the production steps contained a catalyst that converted from "all L" to "racemic mix". Which they wouldn't have picked up unless they were actively looking for it.
Posted Feb 10, 2012 13:41 UTC (Fri) by Cyberax (✭ supporter ✭, #52523)
Well, it does interconvert at least in-vivo: http://www.ncbi.nlm.nih.gov/pubmed/15080764
Posted Feb 13, 2012 11:54 UTC (Mon) by nye (guest, #51576)
Thalidomide will racemize in the body. From http://www.ncbi.nlm.nih.gov/pubmed/15080764: "Thalidomide interconverts between the (R)- and (S)-enantiomers in plasma"
I think it's quite common for enantiomers to flip. For example, get some fresh caraway, and note that it smells *very slightly* like spearmint. Then leave it in a jar for a year or two, and note that it now smells rather *strongly* of spearmint as some of the s-carvone has become r-carvone.
Posted Feb 13, 2012 11:55 UTC (Mon) by nye (guest, #51576)
Posted Feb 8, 2012 22:43 UTC (Wed) by oever (subscriber, #987)
If the government did that, then pharmaceutical companies could outsource all high-throughput experiments to the government.
The government does not do the tests themselves, because they are expensive and the outcome is unsure. Private investors are more motivated to minimize the costs for testing. The government needs to specify the requirements that the drugs should meet.
Posted Feb 8, 2012 13:12 UTC (Wed) by niner (subscriber, #26151)
Could we please stop reiterating this completely unproven assumption?
Posted Feb 7, 2012 23:40 UTC (Tue) by kunitz (subscriber, #3965)
Copyright © 2013, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds