Tue
Oct 24 2006
09:44:am
By: Michael C. Neel  shortURL

It's just started to hit the wire, but the news is IBM has sued Amazon over infringing patents.  I haven't had any luck tracking down the claims IBM submitted to a court in Texas yesterday, but it appears they cited 5 patents.  These patents are said to involve "Amazon's entire business" and cover ad targeting, storing data, and ordering from an electronic catalog.  The patents were filed in the 1980's though early 1990's.

When I get the text of the lawsuit, and the patents involved, I'll do a detailed breakdown of the case.  Right now however I can only comment on the issues this will present.  The biggest issue bar far is the "Software Patent" - many of the things patented under a software patent are merely computers doing a job that was once done without a computer.  One patent said to be in the lawsuit is IBM's "Placing an order on an electronic catalog" - which is nothing new if you take away the work "electronic."  A patent is supposed to be non-obvious to someone who works in the field of the patent, and I'm willing to bet most software patents do not meet that test.

There is also the issue of the purpose of a patent.  A patent is to allow an inventor the time to recoup his investment in developing a working version of his idea and get it to market.  Now I'm not for patents in general, but lets look at a drug patent.  It can take years to tens of years to get a drug though all the testing required by federal laws so it can be sold to the public, and that cost adds up.  Looking at software, it takes a moment to have the idea and a week to code it (most of the time in software development is spent on issues not related to the core logic of the program, the part under patent).  If we are talking about something that will fit into a website, like Amazon's "One-Click" patent, it can even be deployed within a month.  So there isn't much investment of inventor time even if we excuse the non-obvious requirement.

IBM suing Amazon is also a breakdown of these tow companies stand on patents.  In the past, IBM has applied for patents most for defensive purposes - to use when they were sued over a patent.  Amazon started out along evil lines, but since then Amazon founder Jeff Bezos has helped gain support for patent reform legislation.  Now that IBM has launched a full scale invasion, claiming Amazon owes royalties on every dollar earned, I expect gloves to come off and a counter suit from Amazon because I'm sure Amazon can find a few patents it holds to use against IBM.  And then we play Global Thermonuclear War (IBM would do so well at chess too).

Personally?  I say let the nukes fly.  A full scale war would scare any technology company into supporting patent reform, and maybe, just maybe, dropping the whole idea.

Metulj on a Public Terminal's picture

Keep it up

Some good tech blogging is always welcome. Thanks.

In other news, Microsoft patents the verb 'to be.'

Andy Axel's picture

In other news, Microsoft

In other news, Microsoft patents the verb 'to be.'

But doesn't Apple already have the iExist?

____________________________

Forget patriotism. Instapundit.com is the last refuge of scoundrels.

Socialist With A Gold Card's picture

Former patent examiner here

I worked as a patent examiner in the US Patent & Trademark Office for a number of years, so I feel a strange and irresistible compulsion to comment.

My tenure at the PTO was seemingly a lifetime ago, and quite a few things have changed since I left. The standard of "obviousness" is one of them, as is the standard of "what can be patented in the first place." Both standards are ancient in patent law, and both have changed significantly just in the last decade.

Historically, patents could not be granted on methods of doing business, nor could they be granted on any algorithm or process which merely represented a series of mental steps. If a process could be duplicated by merely thinking through a process, it was unpatentable. This is a standard that dates back to 17th-century English patent law.

Back in the day, software by itself was considered unpatentable unless it was in combination with some kind of physical device or physical process; this was due to the legal conclusion that software by itself was merely equivalent to a series of mental steps. Then, the standard evolved so that the "physical device" could include a computer modified by the presence of the software (in other words, the computer became a new and unobvious device once its state was modified by the unobvious software). This seems like splitting legal hairs, but it prevented violations of the old "mental steps" prohibition.

Because of a handful of court decisions from the Court of Appeals for the Federal Circuit, and because of the GATT agreement, those standards have loosened even more in recent years. Now, methods of doing business are patentable (for the first time in over 200 years of our patent system and 350 years of English law). Software by itself is also now patentable; a physical device is no longer required.

Alongside this (and for slightly different reasons), the standard of "obviousness" has loosened to a significant degree. This has occurred for all types of inventions, and not just software. Inventions which would have been unpatentable due to obviousness back in the day would now fly out the door with the Commissioner's seal on them.

For the first time since 1963, the US Supreme Court has just recently agreed to hear a patent case clarifying the standard of obviousness, since the whole thing has gotten so seriously out of whack. This Supreme Court term should be interesting for patent lawyers.

I agree with the first part of your argument regarding the silliness of most software patents; some software should be patentable (if it does something a human mind couldn't achieve in a reasonable amount of time, and performs some new, unobvious kind of task). However, I disagree with the third paragraph of your post; the amount of time necessary to develop an idea and the amount of time to create a prototype are and always have been irrelevant to the issue of patentability.

An inventor can see something in a momentary "flash of brilliance," and such inventions can often be revolutionary (in fact, the "flash of brilliance" was one of the tests employed by the Supreme Court in the 19th century). Also, patentability has never required that the invention be built or implemented; therefore, the time necessary to make it doesn't play a part. The issues are novelty ("Has it been done before?") and obviousness ("Is it an obvious modification of prior art?"). The time needed to make the thing is not relevant to either question.

I saw plenty of inventions that could be manufactured in a matter of hours or even minutes, but that had nothing to do with their novelty or unobviousness.

I also think the nukes should fly in the Amazon/IBM battle, but the end result should be a clarification of obviousness and a clarification of what's patentable in the first place. I don't think we should do away with patents entirely, but I think the policy of awarding patents on software deserves a serious and critical re-examination by the courts and by Congress.

--Socialist With A Gold Card


"I'm a socialist with a gold card. I firmly believe we need a revolution; I'm just concerned that I won't be able to get good moisturizer afterwards." --Brett Butler

 

Michael C. Neel's picture

IANAL, but...

Thanks for the insight - I have not worked in the patent office; the closest I've been was working for a company that applied for a patent on some software I wrote for the company. I recall being told by the lawyers we would have a year after the patent was approved to complete the "invention" but that may have them basing the advice on case law and not from the patent office.

I've found Jefferson's thoughts and history on our US patent office to be quite interesting. It's clear he was concerned about the very thing that's happening today (after I wrote this I saw SGI is after ATI over a graphics display patent). I also am surprised to know most of the founding father inventors never patented much of their work, many saying their inventions were not worthy of a patent.

Oh, well... this is probably going to end in a sealed settlement deal anyway...

Socialist With A Gold Card's picture

The lawyers

I recall being told by the lawyers we would have a year after the patent was approved to complete the "invention"

Michael, I think you may have that backwards. You have one year after the invention is divulged to someone else to file your application. If it's published or divulged and you wait even a year and a day to file your application, you can't get a patent. I had an inventor from someplace in Europe (I don't remember where) who had presented his idea at a conference in Russia. A year and a month later, he filed for a patent in the US. I couldn't give it to him because he had divulged it at that conference.

Jefferson was a visionary about many things; the potential for innovation to help cure society's ills was one of many of his passions. He was the first Commissioner of Patents (acting as Secretary of State), and his legendary inventive streak continues to inspire the staff of the PTO to this day. Seriously. Every patent-related conference or meeting I ever attended managed to work in a Jefferson reference somewhere. Hell, the paper copies of patents are still stored in drawers called "shoes," because Jefferson stored copies of the first patents in old shoe boxes.

I hope the Amazon/IBM thing works its way through the courts in a very public way. The issue of goofy software patents has to be addressed sooner or later, and the issue of obviousness is (as I mentioned above) going to be addressed by the Supremes this term. I hope Amazon and IBM slug it out in public, because the outcome of that fight could benefit innovators and the rest of us for a long time to come.

--Socialist With A Gold Card


"I'm a socialist with a gold card. I firmly believe we need a revolution; I'm just concerned that I won't be able to get good moisturizer afterwards." --Brett Butler

 

Michael C. Neel's picture

KnoxViews...

I just gotta say, getting a comment on my article by someone who's worked in the patent office is <peter griffin>frigg'n sweet</peter griffin>.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Post new comment

The content of this field is kept private and will not be shown publicly.
  • Lines and paragraphs break automatically.
  • Web page addresses and e-mail addresses turn into links automatically.

More information about formatting options

CAPTCHA
This question is used to make sure you are a human visitor and to prevent spam submissions.
Image CAPTCHA
Enter the characters shown in the image.

User login

TN Progressive

TN Politics

Local Media Blogs

Local Paper

Local TV News

State News

Wire Reports

Site Statistics

Last 7 days:
  • Posts: 33
  • Comments: 252
  • Visits: 8,514
  • Pageviews: 21,362
Last 30 days:
  • Posts: 120
  • Comments: 1201
  • Visits: 37,763
  • Pageviews: 91,565

TN Progressive

Nearby:

Beyond:

At large:

Police Scanner Feeds