7 Things To Remember About Copyright Before You Distribute That Cool Thing You Made!

[Janx]

Speak for yourself.

I believe I did.

I have a patent, too, and I don't agree
Here in Germany it's basically impossible to patent an algorithm. So, in order to patent my software it had to be applied to a particular device, in my case a PBX. It would have been applicable to all kinds of network devices, but that would have required separate patents. And since I came up with it working for a telephone complany they weren't interested in any other applications. Now, is that a good or a bad thing? It means that someone else can copy my approach, as long as it's used in a different kind of device without having to respect my patent. I'm totally fine with that.

That is a short-sightedness on your governments part. PBX's are dead. Now it's VOIP which is just IP packets traveling through switches, routers and firewalls. Everything is software and the devices are dumb without it.

Now it is fine if you guys just wanted a patent to cover your industry, but that's not necessarily the best restriction for all patent ideas.

Since you brought up Tesla: in his biography, he's cited as complaining about patent law: He would have liked to patent the principles for his motors but since that wasn't possible, he patented as many variations of the principle that he could think of (and afford...).

He was a patent whore as much as anybody. Now if his goal was to patent magnetic fields or something fundamental, that'd be as bogus as patenting a gene. Nature should not be patentable. If he wanted to patent the configuration of magnets and coils that cause an AC motor to spin, that's probably legit.

Really, I'm a firm believer that some things simply shouldn't be patentable. What would you have done if someone had patented conditional jumps? It would have crippled the entire field of software engineering. Patenting genetically manipulated plants or animals, medicine, etc. are all questionable applications of patenting to me.

It's like you didn't read my whole post. I too also agree some things should not be patentable. But I also see where some things WERE a clever idea (like Tapping) that were only obvious after the fact.

Conditional Jumps? You mean GOTOS, the "thou shalt not use these" of advanced programming languages?

in a lower level language, they are inherently obvious as in if you are following a set of sequential instructions (having to be sequential by nature of it's a Computer) and you write a comparison, it is inherently obvious that you'd need a GOTO command as part of that to branch out to other parts of code. Thus it would have failed the "non-obvious" test. Anybody inventing machine language/Assembly would have figured that out.

As I noted before, I don't think Nature should be patentable. Gene patents where the researcher simply identified a naturally occurring useful gene in a test subject is bogus. Michael Crichton's (sp) last book was pretty definitive on the problems with that.

 

[tomBitonti]

I didn't think that algorithms were patentable here in the US.

I did find this quote (which has an embedded quote, which is actually more specifically applicable to my response.)

The United States Patent and Trademark Office has granted patents that may be referred to as software patents since at least the early 1970s.[29] In Gottschalk v. Benson (1972), the United States Supreme Court ruled that a patent for a process should not be allowed if it would "wholly pre-empt the mathematical formula and in practical effect would be a patent on the algorithm itself"

What is obvious and non-obvious is a whole nasty kettle of fish. What is more important now-a-days is whether you can tie the invention to a value. That doesn't seem to fit the definition of "non-obvious", as in "non-obvious to a practitioner in the field", but nevertheless is what seems to be used. While that casts the question in a distinctly commercial light, perhaps this was always the case. I can't say, since I don't have enough history in the domain. It's important to keep in mind that patents are a reflection of the intersection of three areas: Intellect vs Commerce vs Law & Politics. Patents consider not just the cleverness of an invention. The value is important, too.

Thx!

TomB

 

[Janx]

I didn't think that algorithms were patentable here in the US.

I did find this quote (which has an embedded quote, which is actually more specifically applicable to my response.)



What is obvious and non-obvious is a whole nasty kettle of fish. What is more important now-a-days is whether you can tie the invention to a value. That doesn't seem to fit the definition of "non-obvious", as in "non-obvious to a practitioner in the field", but nevertheless is what seems to be used. While that casts the question in a distinctly commercial light, perhaps this was always the case. I can't say, since I don't have enough history in the domain. It's important to keep in mind that patents are a reflection of the intersection of three areas: Intellect vs Commerce vs Law & Politics. Patents consider not just the cleverness of an invention. The value is important, too.

Thx!

TomB

given the expense ($10K is common), anybody submitting to the process is likely certain of the value. In a large corporation, the expense is sort of higher. Patentable ideas are submitted to an authority group that reviews entries and decides what is worth pursuing. Those people cost money, as they are usually senior engineers/lawyers (one of my old co-workers joined that group and eventually became an lawyer through it). From there, the idea is submitted to a lawyer who drafts up a patent submission, which also gets reviewed with the originator. That lawyer costs time, plus the time during their review/editing stage. Then there's the sending to the USPTO. That costs money, plus about 5 years waiting for it to be approved.

Generally, you don't go through the process except on your best ideas that you can capitalize on. Value is therefore inherent in the process usually (except for patent trolls, those guys suck).

I can't speak to the court case TomB mentioned, but one semantic thing to be wary of is that a programmer will refer to their bit of code as "the algorithm" which in laymans terms really means the process and NOT usually specifically referring to a mathematical formula.

This may be something that muddies up cases when mixing legal speak with technical jargon.

 

[tomBitonti]

(Text omitted.)

I can't speak to the court case TomB mentioned, but one semantic thing to be wary of is that a programmer will refer to their bit of code as "the algorithm" which in laymans terms really means the process and NOT usually specifically referring to a mathematical formula.

This may be something that muddies up cases when mixing legal speak with technical jargon.

I was wondering about that. The difference between "process" and "algorithm" seems small, in everyday speech.

I've been through the patent process at a corporate level several times (google my name). I rather cannot say anything further here in that space.

Thx!

TomB

 

[Andor]

Something worth noting as well are Trademarks. Copyrights are ironclad. Patents might require litigation to defend but cannot be lost. Trademarks however are weird. You can steal a Trademark from someone without even intending to. They can also have local prior uses which ignore your global trademark. I think there is a small town somewhere where Burger King has to call the whopper something else because a local dinner already had a burger called the whopper. Like I said weird, and I am NOT an expert. However you can skirt Trademark infringement issues by (oddly enough) nothing that you are using the (whatever) without permission.

I recall once printing an image on a catalogue where an image that was supposed to be a generic starship trooper ended up looking too much like a GW Space Marine. We got permission from GW to print it with a note saying we printed it without permission. Yes, it makes that much sense.

 

[prosfilaes]

However you can skirt Trademark infringement issues by (oddly enough) nothing that you are using the (whatever) without permission.

The point of trademark is let a company build a brand, so if you buy a Ford, you know that it's backed by the Ford company (for better or worse). But they wanted third parties to be able to make parts compatible with a Ford F150 or Windows XP or D&D, so you can use the name on your product, but you can't trade on their quality; you make or sell a part for the Ford F150, but not Ford's part for the F150. Of course, companies often don't like that, so they use technological means (NES and Sega Genesis) or licenses that make things available (like the OGL) to stop people from using the trademark or making compatible parts. (Or lawsuits, like TSR vs. Mayfair, Sega vs. Accolade or Lexmark Int'l v. Static Control Components, which the courts frowned on but could often produce settlements.)

I recall once printing an image on a catalogue where an image that was supposed to be a generic starship trooper ended up looking too much like a GW Space Marine. We got permission from GW to print it with a note saying we printed it without permission. Yes, it makes that much sense.

I think that was as much GW as the law. A Beanie Baby price guide was sued, and the judge's ruling sounds slightly annoyed that the license for the licensed price guides insisted they say "used without permission".

 

[Jhaelen]

But I also see where some things WERE a clever idea (like Tapping) that were only obvious after the fact.

I was disagreeing with your example of 'Tapping' as something that was a clever idea. If you've read WotC's patent, you'll have noticed how it tries to patent not only 'tapping' (i.e. turning a card sideways) but _all kinds_ of ways to represent that a card's effect has been used, e.g. using counters, etc. You may think of 'tapping' as being a clever idea, but using _some_ kind of indicating you've used a card _is_ obvious. E.g. what I sometimes still do is to place cards that I've used to my right, and placing unused cards to my left.

Conditional Jumps? You mean GOTOS, the "thou shalt not use these" of advanced programming languages?

No, that would be an unconditional jump. I was referring to the assembly instruction underlying an IF... THEN... ELSE... statement.

Thus it would have failed the "non-obvious" test. Anybody inventing machine language/Assembly would have figured that out.

That's funny. Because that's something I'd have considered an example of something that was only obvious after the fact. I guess, that serves to show it can be hard to decide what is obvious and what isn't. What's even funnier is that googling it, I actually found a related U.S. patent.

As I noted before, I don't think Nature should be patentable. Gene patents where the researcher simply identified a naturally occurring useful gene in a test subject is bogus. Michael Crichton's (sp) last book was pretty definitive on the problems with that.

Yup, looks like we're in agreement here!

I didn't think that algorithms were patentable here in the US.

I did find this quote (which has an embedded quote, which is actually more specifically applicable to my response.)

Ah, thanks for the quote! This is precisely what I thought was missing from US patent law.

 

[jeffh]

No, that would be an unconditional jump. I was referring to the assembly instruction underlying an IF... THEN... ELSE... statement.

That's funny. Because that's something I'd have considered an example of something that was only obvious after the fact. I guess, that serves to show it can be hard to decide what is obvious and what isn't. What's even funnier is that googling it, I actually found a related U.S. patent.

Computer languages are, at the most basic level, implementations of the rules of formal logic. "If...then" has been a mainstay of those for centuries. So I'd definitely come down on the "obvious, and not just after the fact" side there.

 

[tomBitonti]

What is obvious or not seems to be very difficult to determine.

My understanding is that logic was, since the time of the ancient Greeks, based on syllogisms, as opposed to a the little matrices which show the several different ways of defining binary operators. You would think that the binary operators and the little grids were obvious, but for a long time they weren't.

The idea of using punch cards to perform instructions, and then feeding the cards back into the computation unit in a loop was considered a big deal when it was first done, but it seems rather trivial today.

I do think that you have to couple an idea with a valuable application: Not just the sequence for genes which are associated with an increased risk, but also some application (a value) of that sequence.

You have to be careful about compositions, which used to be the basis for a lot of patents, but aren't much anymore based on a court ruling.

Sometimes, applying and old idea in a new domain is not considered to be much, but sometimes it's considered to be quite clever and patentable.

Thx!

TomB

 

[Morrus]

Just bumping this as I think it’s relevant.