I recently ran across my first software patent.

Looking through the code it dawned on me after a few minutes that what they had done was to patent the idea of a feedback loop with process dead time calculations added in.

I really was unimpressed.

 

I would be most impressed to see _one_ truly valid software patent. Since most practitioners seem quite capable of sniffing out bad patents, one wonders how the USPO can possibly do so poorly. It seems as if they have a policy of "grant first, ask questions later." I sincerely doubt that a position of not granting software patents could possibly do as much harm as those they have granted. After all, it's supposed to be a bargain in the public interest.

Regards

cww

 

That idea is so generic it's called a "Smith" predictor! ;-)

--
Steve Myres, PE
Automation Solutions
(480) 813-1145

 

The patent that is really obsurd is the one regarding hooking into a spreadsheet from a PLC. Some Law firm bought it from Modicon is making a fortune from Fortune 100 Companies. Leave it to Lawyers to see a way to make money on something that is irrelavant. The bigger question is, are they splitting the rewards with Modicon. Comments are surely welcomed. It would be nice to see how many people have heard about this lawsuit.

 

Software patents especially in the automation industry is a tricky subject and open to interpretation. Who does own the software? Is it the code writer since they are assumed to be his thoughts and original ideas to implement the desired process control? Is it the customer since they have paid for the field controllers and probably the programming software license for future support? Does this mean that if one company writes the original software, goes out of business and than a few years later the softare needs to be updated does someone new require written permission from the previous company to make changes? My thoughts are that in the end the customer owns the software and has the freedom to edit/change/modify any or all parts of the logic. The way PLCs are to me is an open architecture system. You don't necessarily require a source code decompiler to change anything. What are some of other people's thoughts on this?

 

I thought Software is copyrighted which gives the author 50 years of protection vs. 17 years for utility patent. Is SW patents new and how long is the protection good for?

oj

 

The notion of a software patent is so vague, I can’t even conceive of the idea. Isn’t a software license agreement sufficient? What is one protecting when he or she patents software, syntax, results of an operation, mechanics of an operation, or the concept? It sounds to me like just another way for the ones with the deep pockets to intimidate the ones with the shallow pockets.

 

Sadly, I think part of the logic is:

1) if USPO denies a "valid" patent, it could lead to a lawsuit against USPO (possible?) or someone giving the President a $1 million "political donation" & adversly affecting certain people at USPO.

2) if USPO accepts a "bogus" patent, then 2 external companies end up settling the issue in court - which I think is how must bureaucrats think the USA "free market" is SUPPOSED to work.

So I think a large part of this problem is just "cover they butt" by the USPO people. They don't get fired for doing #2, but might by doing #1.

- LynnL

 

Correct me if I'm wrong, but aren't patents different from ownership? I own many objects that are patented and I make modifications to them all the time. The issue of patents and copyrights involves distribution and use. I can download someone's white paper off the web and make modifications to it for my own use. Once I try to distribute the document or pass it off as my own without permission, then I am infringing on the copyright. Without any specific licensing agreement (read the fine print that comes with any software package today, they often include many restrictions including copying, modifying, or reverse engineering), you can modify the software any way you like as long as you don't try to re-sell or distribute the software or modify it such that it included patented or copyrighted material from someone else. I believe that the restrictions that are pertinent to an end user are the licensing restrictions provided by the supplier. Any legal eagles out there that can validate my interpretation?

 

In my humble opinion:
A particular software implementation (actual code) can be copyrighted.
A particular process expressed as descriptive "claims" in Patent-lingo
(involving software or not) can be patentable.
Meir

 

In most countries software is a grey area. However, there are three distinct legal mechanisms that must be considered. These are patents, copyright, and license.

To begin with, most countries do not allow patents on software. In those that do, you may own a patent without having any actual software which implements it. That is, you may hold a patent which companies who actually write software must license from you. It is probably impossible to write a major software application without unknowingly violating someone's patent.

Copyright law controls copying and distribution of software. However, there is a special twist on this which applies to software. In many countries it is judged that when an end user copies the software onto their computer (e.g. installs it) or copies the software from their hard drive into RAM in order to execute it, they are "copying" the software and this is therefor violation of copyright if the software was not purchased in an authorised manner. This is different from for example a book, where the owner of a copy of a book cannot be held responsible for copyright violations by the author or publisher.

Finally, there is license. This comes under contract law, and is therefor different for each country. For mass market software this is usually implemented by the infamous "click-through" licenses which appear when you install software. Patches for bug fixes or security holes may contain new license restrictions in addition to the ones in the original package.

Many people are of the opinion that these "licenses" have no foundation in law and are utterly worthless. Others believe that these licenses may in fact have some bite to them. The validity of these licenses is not clearly established, and this may in turn vary between countries.

Most people don't read the terms of the "click-through" licenses that they "agree to" when they install software. This is just as well, since they probably wouldn't be willing to accept the terms if they took them seriously. The "click-through" licenses will often constrain you from reverse engineering or modifying the software. Some bar you from benchmarking it or otherwise comparing the software to its competitors.

The above is general comment based on what I have read and should not be construed as legal advice. If you want a valid legal opinion, then talk to a lawyer.

--

************************
Michael Griffin
London, Ont. Canada
************************