November 23, 2004

Software Patents: IMHO

I was asked today about my opinion on software patents. This person has the opportunity to file some disclosures on products he has been working on, but he feels some discomfort about doing so because he is not sure whether or not he agrees with the idea of patenting those products. On the other hand he is being encouraged to file. There would be a financial award and it would be good for his career. My reply follows....

Well, my outlook is that I should do my part, and state my reasons for doing so to other people so they understand what I believe, and then maybe they will decide that patents are also bad. But it is not as though I will think someone is a bad person for believing in patents or filing them. Everyone is entitled to their own opinion on the matter.

I have a somewhat extreme view on patents. I've posted some of this on my personal blog, and also told my manager about it so she knows that although my career would benefit from it, I'm not willing to compromise. I think the issue is too important. If IBM ever decides patents are the only way they will value more anymore, or the only way I can advance, I will leave.

I am not 100% sure, but I believe I have met someone else in IBM who feels that their career has not moved forward because they do not have any patents. That could be true. But I also believe what one of the other IBMers I met said; that I should not be expected to or feel I need to compromise my morals or beliefs for the job.

Maybe it is easier for me to say that, since I'm a little younger. I don't worry about a job or stuff like that though; I've got enough people trying to hire me that I'm confident I could find something quick.

One of my CS professors told me that although he thinks the entire patent process is completely messed up, I should go for it because as one person I'm not going to fix it and it does mean he and I can get rich off the way things work. But I think that's exactly the problem with patents. People are naturally selfish and greedy. Shareholders are this way. Companies are this way. Employees are this way. Etc. That's why the system perpetuates itself.

In simple terms, I think of it this way. Where would our species be today if the first person to figure out paper patented it? Or fertilizer? Or the axe or screwdriver? The internal combustion engine or the wheel?

Of course, in your case, you don't have the option of sharing your ideas no matter what your decision. Work for hire means the company owns any IP. It's their idea and not yours. They will most likely get it patented somehow even if you don't have a hand in it. I am the passive protestor to this, I suppose. During Extreme Blue, there was a big push for filing disclosures. My team was the only one not to attempt any disclosures, as far as I know. If someone requests my help specifically on a patent application or anything related to a patent, I will refuse. I will not attempt to stop someone else, but I will not do anything to help.

There is also the argument that if your asset is your ideas, when why shouldn't you be able to patent those ideas and that will become the basis for your contribution to the species and how you make a living. You are the thinker, and someone else is the builder.

One of the problems I have with that mindset is that it is selfish and self-serving. Ideas are meant to be shared. That is how civilization has made progress for thousands of years. Everyone believes our children are our most valuable resource and that teachers are so important. But when it comes to making money, don't you dare teach! Keep it for yourself and milk it for all you can.

It also becomes a problem because ideas are not standalone things. All ideas are built on top of other ideas. At this point, it turns into that childish argument that goes something like this: I wouldn't have hit you if you hadn't broken my toy; I wouldn't have broken my toy if you didn't buy the toy; I wouldn't have bought the toy if you hadn't broken my other toy. Etc. It sounds crazy to apply to this patents, but due to the lifetimes of patents and copyright, it is applicable. A jet engine has so many patents on it that for someone else to build one would be the same way. You owe me for this and this and this and this and this.

The only reason we can write software today is because so much of it is coming out of academia where patents do not apply. Quicksort, arrays, try/catch, object-oriented languages, etc. We have competition for journaled filesystems today, but that would not have been possible if the idea came out of a company with a patent. Things that seem cutting edge today are the building blocks for the mundane tomorrow.

Crazy rant done.

I think for you, the two sides are the compensation+career, versus the internal discomfort of having your name associated with something you don't necessarily believe in. The majority of people you meet with will not feel that way but instead will congratulate you on your accomplishment. You'll just have to figure out which of those two sides is the one you would prefer to live with.

Posted by josuah at November 23, 2004 11:56 PM UTC+00:00

Trackback Pings

TrackBack URL for this entry:
http://www.wesman.net/cgi-bin/mt/mt-tb.cgi/56

Comments

Author Profile Pagehey dude says at November 27, 2004 1:36 AM

Do as the Romans do while in Rome. Follow the rules while playing the game. If you leave IBM because you refuse to file patents, maybe other companies would not like this type of employee either. Business exists by earning money so they can pay employee salaries. When a company's stock price increases, the cost of running its business is lowered. This is because the company debt will bear a lower interest rate when the stock share price is higher. That is why every public traded company needs to maximize share holder value by executing the business such that it will increase the stock share price. It is a matter of the livelihood of the business. That is just so simple. If the stock price does not increase, the workers' moral will be lower because a lot empolyees' wealth is tied to the company stock price. Every time IBM tries to raise money through the public doman, IBM will have to pay higher interest rate to lure poeple to buy IBM corporate bonds, hence higher business expenses.

Author Profile PageJosuah says at November 27, 2004 10:04 AM

And your comment is exactly why the system is flawed and society is hurting itself. And business does not need to monopolize ideas to make money. Service companies are an example. Half of IBM's revenue comes from its services business. Consumer goods have generic competitors.

One of the reasons processors are so cheap now is because AMD won in court the right to use what Intel claimed was intellectual property that AMD should not be able to incorporate in their chips. A monopoly on x86 chips and a very high Intel stock price would be at the cost of enormous productivity in our society.

Author Profile Pagehey dude says at November 27, 2004 3:45 PM

Your counter points are well taken.

Author Profile PageTim Pepper says at November 29, 2004 6:39 AM

One thing I can't get past is that as long as there's a DMCA and terror fears (ie: forever?) the patent issue is almost moot because one can invoke copyright's longer protection and do so with an embedded security system such that it's illegal to reverse engineer and create compatible works. Who needs a short lived patent to stifle creativity, when copyright is now perverted to do so better? Only a little over a month ago the courts ruled in the Lexmark DMCA case and rules against Lexmark. But the way I read the ruling (and IANAL) it seems like they spell out a recipe for how one might better go about using the existing legislation to lock out competitors, even if in this particular case the implementations were deamed legally reverse engineerable.

Part of the case hinges on some code not being copyrightable. I don't follow the specifics for the code in that case, but I think program code should be copyrightable (of course we need copyrights that expire in a reasonable amount of time and a solid fair use allowance and all the other things that would make copyright again a system of encouraging publication and creativity).

Plus we have an FBI that goes into toy stores and threatens to confiscate (legal, patent expired) replica Rubik's cubes in the name of homeland security. Compared to just after 9/11 things aren't so crazy, but the erosion of our civil liberties is only stalled some until the next attack. The politics of terror throw an interesting twist into the already tangled mess of business economics and technology around intellectual property.

I'm very curious to see the next version of the GPL and how it addresses these issues.

Author Profile PageJosuah says at November 29, 2004 9:25 PM

I don't think copyright being encrypted is a problem in and of itself. It's the restrictions on reverse engineering that is the biggest problem with the DCMA. Basically, figuring out how something works becomes illegal. That's just wrong. I shouldn't be able to punish you for figuring out how something works, just because I wish you hadn't. But I don't know that the GPL can do anything about that. It only applies if you decide your stuff is licensed under the GPL.

Maybe in the future, if you do figure out how something works, the punishment will be selective memory erasure.

Post a comment

July 2013
Sun Mon Tue Wed Thu Fri Sat
  1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29 30 31      

Search