A Patent-Holding Software Engineer Explains Why Software Patents Harm Innovation
from the the-system-at-work dept
It's no surprise that many technologists and engineers dislike software patents -- even as their company's execs and lawyers push them to get more patents. Stephen Kinsella highlights an anonymous comment from a software engineer who clearly works for IBM (though he doesn't come out and say that directly), where he explains how IBM actively encourage engineers to file for as many patents as possible (it rewards them with monetary bonuses). This is not new of course -- an awful lot of companies do this. However, the guy goes on to explain why even though he holds patents and believes very strongly that his company makes and sells nuermous innovative products, he believes all this patenting is damaging to the process of innovation:Speaking from my own experience, corporations (such as the one I work for) spend a lot of money to innovate. However, I would "press that button" and get rid of IP law immediately, given the chance. I agree completely with the arguements made in the article - as such, I'll just bring up a few other issues:Well said.
I think IP law is incredibly damaging to innovation and competition. In the case of software patents, moreso in that they take resources (primarily money which gets redirected to legal teams) from firms who are forced to research existing patens, and also defend themselves against IP lawsuits.
Many software patents are particularly silly. Many of these are issued for algorithms - the vast majority of the time, these algorithms are only available outside the company via patent! That is, when they are shipped externally, it is in a form that is not readable (object code). Sure - this can be reverse engineered. But for a particularly complex program or operating system, this in itself would be a colossal endeavor. Yet, a patent is issued for it - and the patent describes exactly what the algorithm does!
Another firm could look at the patent and use the invention. In most cases, it would be impossible to tell that they've "stolen" anything. Here they are counterproductive.
I should also mention the obvious - the corporation which holds the patent already has a huge advantage! They will ship a product with these innovations before any other corporation can ship its' product. Quite frankly it will generally be a significant period of time before another product can be shipped which contains these innovations - even if the innovation was immediately obvious and known. This will not generally be the case.
Then you have the patents for user interface - these are just silly. I've seen patents issued (granted, this was a long time ago) for using a particular color on a "dummy" terminal.
Anyway, I hope I do not sound like a hypocrite (because I hold IP patents). As I said, it is a part of my job. I also cannot fault my company for taking advantage of whatever silly laws are created. I simply view this as another case of the state interfering with the market, and the market adjusting to exploit the foolishness of the laws.
Thank you for reading this Techdirt post. With so many things competing for everyone’s attention these days, we really appreciate you giving us your time. We work hard every day to put quality content out there for our community.
Techdirt is one of the few remaining truly independent media outlets. We do not have a giant corporation behind us, and we rely heavily on our community to support us, in an age when advertisers are increasingly uninterested in sponsoring small, independent sites — especially a site like ours that is unwilling to pull punches in its reporting and analysis.
While other websites have resorted to paywalls, registration requirements, and increasingly annoying/intrusive advertising, we have always kept Techdirt open and available to anyone. But in order to continue doing so, we need your support. We offer a variety of ways for our readers to support us, from direct donations to special subscriptions and cool merchandise — and every little bit helps. Thank you.
–The Techdirt Team
Filed Under: engineer, patents, software patents
Companies: ibm
Reader Comments
Subscribe: RSS
View by: Time | Thread
Eloquence
Eloquently stated.
[ link to this | view in chronology ]
Re: Eloquence
Eloquent? Perhaps. But wrong in my opinion. "The state" didn't decide one day to twist the original intent of the patent system and encourage patent hoarding. The government may have been a willing participant, but it was the companies that saw this as a means to protect their own business.
While it may have worked for the established companies in the shorter term, this new mentality has ended up hurting companies of all sizes (and their customers). Case in point: IBM may gain quite a bit of revenue from "silly" ideas that were patented, but they do so at the expense of maintaining a huge legal department to handle all of their patent issues.
As I see it, the fact that companies feel like they have to create and hoard patents is an example of "you made your bed, now lie in it." The companies, not the state, brought it on themselves.
[ link to this | view in chronology ]
Re: Re: Eloquence
[ link to this | view in chronology ]
Re: Re: Eloquence
[ link to this | view in chronology ]
Re: Re: Re: Eloquence
This may be true now, but as more and more bad patents are granted, the cost of litigation will go up as will the cost of maintaining your evergrowing staff of patent attorneys. Some would say this has already happened, but you could get to a point where nobody could claim any "new" invention without getting a barage of lawsuits that would bring even the big companies like Apple to a standstill.
And this says nothing about the cost to the customer who is deprived of innovations because they're locked up in a lawsuit or never even brought to market for fear of a lawsuit.
[ link to this | view in chronology ]
Re: Re: Eloquence
[ link to this | view in chronology ]
Re: Re: Eloquence
More than just our laws get affected adversely for The People: campaign donations from companies should be outlawed as well. Gifts for law makers should be expressly forbidden.
We need to give the power back to the People. Its a shame it will have to be TAKEN since corporations would fight tooth and nail to prevent such actions from taking place.
[ link to this | view in chronology ]
Re: Re: Eloquence
But the original intent (or at least intended means) of the patent system IS to interfere with the market, by granting an artificial monopoly. And the state decided that patents could be obtained for software, thus interfering with even more markets.
[ link to this | view in chronology ]
Re: Eloquence
[ link to this | view in chronology ]
i love when insiders speak out
this is why i like cory doctorow's writing. as a sci-fi writer, doctorow has much to gain from IP law, and yet he, like the anonymous software engineer, would 'would "press that button" and get rid of IP law immediately, given the chance.'
[ link to this | view in chronology ]
Translation
Patents are the sign of a decaying, materialistic society. Imagine if whoever invented the first wheel patented it! Free will and freedom to invent are one of our best qualities but patents and IP law run against that grain!
[ link to this | view in chronology ]
So are many patents in every other industry..
"Sure - this can be reverse engineered. But for a particularly complex program or operating system, this in itself would be a colossal endeavor."
So you probably oppose patents for drugs because of the colossal endeavor in reverse engineering. The truth however is that it doesn't take much of reverse engineering to arrive at a high level understanding of how a software algorithm works
[ link to this | view in chronology ]
Re:
[ link to this | view in chronology ]
Re:
Try decompiling Windows. See how "easy" it is to reverse engineer.
[ link to this | view in chronology ]
oops
Oh please.
[ link to this | view in chronology ]
wellll
Ok off the soapbox, IBM is a bit of a white knight. It has some of the most obvious and ubiquitously infringed upon (read totally undefinable) patents that you could imagine. Like accessing a database form a network (Seriously).. And it has (to the best of my knowledge) proven itself to be a champion of reason, and on occasion the little guy. Thus.. if I worked at IBM, I would gladly pad their portfolio, ONLY because patents are VERY dangerous in the wrong hands (Troll firms).
site: http://www.bespacific.com/mt/archives/007272.html
[ link to this | view in chronology ]
Re: wellll
Mudak
patents are property
[ link to this | view in chronology ]
well put, Sir!
Precisely! That is how patents encourage innovation. It's all about dissemination of information.
[ link to this | view in chronology ]
patents at IBM
Patenting also gave them protection for their own inventions if they were key to a performance, efficiency or marketing edge on the competition.
Most of the technical staff was encouraged to file disclosures so that the patent department (through peer review) could determine which patents were most likely to provide both protection and leverage in the market place.
Disclosures that weren't patented were "published" in IBM's internal disclosure bulletins. These bulletins were also housed at a few key libraries and were considered as prior art so that no other company could file patents on these "inventions".
Previously, IBM's patent department/process was considered a cost-center and IBM rarely filed for any patent that was not likely to be granted. For the time period that I'm aware of, the patent filing limit resulted in an average of about 600 patents granted per year and reached a maximum of 1085 in 1993. They weren't using the patents to collect royalties.
More recently (over the last 16 years), the current patent strategy appears to have both a direct revenue and protection component. This is based on the number of patents IBM filed and were granted each year. This started at approx the time that Lou Gerstner became CEO and continues today.
As a by-product of their strategy, more is know about IBM's technology than any other technology company and IBM is still a very profitable company.
http://www.ibm.com/news/us/en/2009/01/14/e714183t64858z03.html
http://www.industrywee k.com/ReadArticle.aspx?ArticleID=1228
[ link to this | view in chronology ]
Why people don't look for 'ideas' in patents
Due to this MOST companies discourage R&D from reviewing any patents, so that they can avoid the 'willful' infringement part when they make the next widgit that 'accesses a database from a network'.
[ link to this | view in chronology ]
he wouldn't if they were his!
[ link to this | view in chronology ]
another anti-patent tripe from Mikey
Any competent engineer who attended conferences and spoke to those IBM or MShit R&D folks (and read their shitty patents too) knows what they think about their own patents: SHIIIIIIIIITTTT !!!!!!!!!!!!!!!!!!!!
On the other hand, there are PATENTS, usually owned by smaller companies and startups which DO teach some novel and unobvious stuff to the world
If no patent protection is available then everything will be a trade secret
Sofware field is no diferent from any other field, patentwise
It just irritates me when ignorant morons like Masnick write anout somthign they have no clue about
[ link to this | view in chronology ]
IBM "patents"
He concluded that while IBM technology is generally pretty good, their patents are of extremely poor quality, almost all of them not citing any non-patent prior art etc. etc.
I stidied a few IBM "patents" in my own field
They are all shit, only sligly better that MShit patents, but still shit not worthy of reading
BTW, the most prolific "inventor" was some IBM guy working on some semicionductor stuf, filing like couple patent apps each week (with his team)
Can anybody seriousluy belive that one can invent something patent-worthy couple times a week, year after year ?
Like I said, SHIIIIIIIIIIIIIIIIIIIIIIIIIIIITTTTT !!!!!!!!!!!!
[ link to this | view in chronology ]
Software Patents
[ link to this | view in chronology ]
Software Patents
I also congratulate the tone of the article. There are patents that can help greatly, and are good for us - the Founding Fathers recognized that in the US Constitution. The article properly limited itself to software.
Now, don't get me started on "non-invention" patents! There is a TON of those!
[ link to this | view in chronology ]
I would press the kill button
What is the rational behind, "I would press the kill button (on all IP laws) in a heartbeat"?
There's no logic behind this. Just a GI-Joe thrill fest.
If there were no IP laws, there would be great dis-incentive to invent and innovate. No need to be first. Just wait for the other guy to do it and then take his workproduct at zero cost. Don't tell me the market will compensate the first to invent. Quite the opposite. He bears all the costs and risks while the copyist is burdened by none of these.
[ link to this | view in chronology ]
Re: I would press the kill button
For someone who claims to be interested in logical explanations, you really ought to try harder... or at least do a little research.
You are simply wrong.
Historical evidence has shown time and time again that what you claim is not true. In societies without patent protection, innovators are still rewarded and followers don't do nearly as well.
Also, you seem to think that innovation is a static thing. It is not. It's a dynamic process. So, sure, a copyist comes along, but by the time he's copied your work, you're already on version 2.
And, in historical research after historical research that's exactly what happened. But why bother understanding when you think you know what would happen.
Sheesh.
[ link to this | view in chronology ]
NonObvious Algorithms
The founding fathers set up the patent system as an alternative to trade secrets. If someone where to come up with an incredibly powerful encryption or compression algorithm I'd rather have that available to the public, even if no one else can use it.
The problems that arise are ones of duration and obviousness. The original term for patents in the US was 14 years, but it's now up to 20. The software industry moves far too fast for any time period like that. A 2 year algorithm patent would give companies plenty of time to make their money while providing the public a new way of doing things.
The problem with obviousness arises from the USPTO being either unwilling or unable to discuss these patents with people who have experience in the field in question or do basic research. Another way to resolve the issue is to simply disallow broad/vague patents, and fine anyone who submits one.
If even my first suggestion was followed, then patents would be much less of an issue. The second suggestion would insure that only things that are of useful value and not just a tool to get money are accepted.
Anyways that's my $2 on the matter.
[ link to this | view in chronology ]
Different purposes
When profit margins were high and money flowed freely, I suspect a large part of this was just bragging rights on IBM's part. They could show they were among the leading tech companies by citing the large number of patents. Of course, this also works internally: Their large IP staff could cite their success in bringing IBM to a leadership position in patents to justify ... having a large IP staff. There's less of that sort of thing these days, because money is much tighter.
There *can* be practical uses for such otherwise-useless patents as defensive measures: Having the patent yourself makes it impossible for some troll to get it and come after you. Also, patent lawsuits among the big players are usually settled by cross-licensing, the result of a kind of "mutual assured destruction": If you're actually producing product, as opposed to just trolling, and you go after IBM, they'll go through their vast library and find *something* that you're using. Now - shall we fight over both of these patents, or shall we settle quietly?
By the way, the traditional advice to publish as a way of blocking patents has gotten dangerous as a result of misuse of provisional filings and continuation patents. I file a provisional filing, which is a brief, broad description that's supposed to be a place-holder. It, however, establishes a primacy date. Later, you publish - and I "clarify" my claims by saying they always included what you published. A provisional application is only good for a year, but by various creative games with vaguely drafter, overbroad claims, and continuation patents, one can stretch these kinds of things out. Improper, probably illegal in some cases - but people have gotten away with it.
[ link to this | view in chronology ]