As seen on Slashdot and other places there has been quite a lot of public discourse recently in wake of consortiums consisting of Microsoft and Apple among others buying the Novell and Nortel patents. This caused Google through its Senior Vice president, David Drummond, to call the use of these consortiums to buy patents anti-competitive. And he is absolutely correct. In the Zdnet article they point to a tweet by Microsofts Brad Smith who counters that they offered Google to bid with them. Well as I pointed out when blogging about the Novell patent sale, the problem here is that these patents are needed to deter lawsuits from Microsoft, and co-owning some patents from Microsoft doesn’t really achieve that.
Another horrid attempt at a rebuttal to Google is from John Gruber who starts by trying to equate the patent problem with Google undercutting Microsoft on price. Why even though he partially admits the US patent system is broken later on, he tries to say that Microsoft attacking Google with bogus patents is the moral equal to Google competing with Microsoft through undercutting them on price…/facepalm.
And he then tries to be clever and rhetorically ask “But what exactly does Google need to defend against, if not actual patents Android actually violates?”. Well the answer to that is that Google needs its own pool of bogus patents as it is the only way to protect yourself from other peoples bogus patents. But what John seems to forget is that two wrongs doesn’t make a right, and unless the US congress at some point decides to actually do something that helps the US economy and drop the stupidity that is software patents, then innovating companies will continue to need to waste money and time on software patents, so protect themselves form the attempts of the market incumbents to abuse the patent system to shut down their competitors.
For those wanting to read up on how damaging the current software patent regime is I recommend this article from the Economist called Patents against Prosperity and this blog by Craig Hockenberry called the Rise and Fall of the Independent Developer.
The sad part though is that by the time Google manages to build up their patent arsenal to protect themselves, I am sure they will have managers who decide that in order to protect the interest of Google shareholders, Google should start to favour software patents, just like Bill Gates turned coat on the issue once they realized that while patents lawsuits would cost them a fortune, they could use it to kill of a lot of potential competitors.
Thanks for this sane summary of the situation. It seems like a lot of websites have gone mad, confusing Novell with Nortel and basically missing the point entirely.
I hope you will be proven wrong on your last paragraph though. :)
Patents seem like an example of the broken window fallacy. Just because a giant pile of money moves around because of patents does not mean that they represent a net improvement to society.
Your analysis is exactly correct. There was an episode of This American Life recently that covered this very topic. http://www.thisamericanlife.org/radio-archives/episode/441/when-patents-attack
While I’m against software patents as much as the next guy, I do feel bad sometimes for Apple that the amount of technical and theoretical innovation that the iPhone represented has been so shamelessly copied. I have to imagine it’s disheartening at Apple to invent something as cool and intuitive and new as pinch-to-zoom and then see it copied and given away for free at no benefit for you or the product you’re working for.
I just wish there were a system to encourage innovation in software (as patents do very well in modern hardware) without the broad and terrible implications that modern software patents have caused. But I can’t think of a way to split the difference.
@Ethan: well one way to fix the system and still give true innovators a chance to profit from their ideas would be to reduce the length of software patents from 20 to 5 years. At that level I am sure most patent trolls (including Microsoft) would not find it worthwhile to file and maintain random patents, while really good and innovative ideas can be used by their inventors products exclusively for a little time.
@Ethan
Do you want to say Apple got too little profit for inventing iPhone? As far as I know they made loads of profit and are still making more. Patents are not needed, innovation already has an inherent benefit: the first-mover advantage.
Copying of innovations is a crucial element for economic efficiency, because it creates price competition among manufacturers. If copying is not allowed (through patents for example), monopolies are created, because the initial innovator later does not need to do any more significant innovation, as its competitors cannot enter the market. Additionally, patents forbid other companies to innovate on top of the original innovation, thus blocking further innovation.
The good thing about Google is that it’s not run by managers, but engineers. So no, I don’t think Google will start to favor software patents.
Google copied Java as Android w/o licensing. Shame on them!
@uhuu: If I invent a brand-new toaster that’s dramatically better and set up a business to sell it, would it be fair if a big company came in and made something identical and dramatically undercut me on price (even subsidizing the low cost of the toasters from something else) a few months later? What incentive do I have to invent new toasters if I can’t get a business from it for more than a few months?
Yes, patents create a limited monopoly, but they do it intentionally and for a reason; the ease of copying discourages first movers from developing something new in the first place–especially if the innovation is easy enough to copy that the first mover advantage is low.
At that point, why would anyone bother to create something new, other than for academic fun? No-one is arguing for the patent system as it stands re: software, but it exists for a reason.
@Christian: That wouldn’t stop the abuse of the system, though. There was a recent story about a company set up by two Indian developers suing virtually everyone for “a system to communicate online”, with a patent that has technically not yet been granted. The problem isn’t the term of patents, it’s the type.
@Ethan: Small inventors being able to defend their product using patents is the ideal, but the evidence both for and against it seems highly anecdotal. For every story I heard about a small inventor through history who have managed to do well with their patents, I have heard a story about how the big guys been able to use patents to shut them down, or simple use their market power to keep them of the market for 20 year. (not speaking of software patents here, as they seem to be almost all of the bogus kind). Read for instance this book talking about how Edison for instance got screwed over even with patents protecting his ideas.
I think competing with big established players is hard, but a small new company can do it if they manage to make good products and use their size to be fast and nimble. I am sure there are cases where having patents can help in such a scenario, but I think one will find almost just as many cases where there opposite is true. Big companies are slow and bureaucratic, thats their weakness, and small inventors can use their inventiveness against them, that is the motivation to innovate, but they need to keep innovating, not just rely on 1 good idea and hope that will make them the next big thing.
As for my idea for 5 years, I agree that ideally removing software patents all together would be a better solution, the problem is that we need big software companies on board if we ever want to hope to get enough pressure on the issue to change patents law, and I think we will find that even if they see the problem the concept of wiping billions worth of patents from their balance sheet overnight will scare them away. So a 5 year patent duration for software patents might give them enough transition to be able to support it, and while there might still be abuse, at least it will be a lot less of it.
@Ethan: Where your argument breaks down is that software development has flourished just as fast before patents came into play. Even without patents, software companies bothered to come up with new stuff.
One of the first, if not *the* first, so-called “software” patents was granted to AT&T in 1988, after 4 years of concerted legal wrangling. That patent and all subsequent “software” patents are really system, process, or method patents. These types of patents have a long history in manufacturing – like say a novel process for producing soap using no heat thus saving energy. This protection was denied software based processes until 1988, and the industry largely relied on trade-secret and to a lesser extent copyright law.
There were a few software patents filed over the next few years, but filings really took off, not surprisingly with the internet “boom” starting about 1996 or so. This was essentially a land-grab, caused precisely because the patent office allowed a vacuum to exist when it had not allowed software patents for the first 40 years of the computer industry. Some of those early software patents have expired, and even the ones filed in the late 1990’s are approaching 10-15 years in age and will start expiring in a couple of years.
While the patent process is flawed, there is also a lot right about it. My point is that we are approaching something of an equilibrium point after 23 years of allowing the first software patent, that exists in other . The problem was not that we allowed software patents, but that for so long we did not. Gutting patents will only play into the hands of those with great resources.