When in doubt, get permission from the author to use his code. But then, how would one know if the "author" didn't copy the code from someone else? Even if he slapped an MIT license on it, how do I know someone else won't claim that they wrote the original code under a more restricted license, which requires royalties to be paid?
Which is why the default being MIT makes perfect sense, and IMHO, limiting other licenses makes even better sense. Then there are no worries. We do what we've been doing for all this time.
Yes, the default of MIT is what I would suggest.
Parallax was nice enough to do an early release of the P2 on the FPQA so that we could get a jump start and get some work done on tools and documentation.
I don't think the intent was so people could stake claim to the usage of new features and lock them down with non-MIT licensing.
We normally have very open discussions and "what's discussed in the forum is open to the forum" seems to be the norm.
We have the thread on LMM for the Prop 2 where Ariba, Chip, and others have been working with Bill on possible versions of LMM2.
It seems that a lot of the pipeline work that led Bill to his byte code engine is the result of that thread and his later optimization likely was the result of Chip's suggestion for improvement.
I was pretty pumped up about the prop 2, and the community effort that seemed to be building around LMM2 was pretty cool, not so much after today.
Yes, and it would likely marginalize potential harm on all sides, encouraging people to use common sense remedies too.
We've had almost no instances of alternative license code posted here. Before that gets messy, it would be very good to set default expectations. Dropping code here that people cannot view and use without worries isn't well aligned with the core purpose and value of this forum.
Let's assume your idea is correct, people should post code to the forum, and no one should make use of said code without seeking permission from the person that posted it.
The Prop 2 isn't even released yet, there is very little code and documentation.
Let's say I hurry up and write out some code snippets for initializing the counters and other really common activities.
It's a good bet that since there isn't much of a code base yet I'll have some unique, albeit obvious code.
By your standard I can become the gateway to usage of the counters...Do we really want that?
Let's say I hurry up and write out some code snippets for initializing the counters and other really common activities.
It's a good bet that since there isn't much of a code base yet I'll have some unique, albeit obvious code.
By your standard I can become the gateway to usage of the counters...Do we really want that?
C.W., I have pondered such things on numerous occassions, particularly pertaining to items such as source code, and I would imagine the courts have heard numerous cases pertaining to simple programming issues. However, for the sake of discussion, the site requirement for a license would in no way hinder this type of activity. For instance, as in your example, I could simply create the code and post it to my own web site or any other web site, thus protecting it. Would such trivial items be enforced in a court of law? I have no idea, but I have definitely pondered such things.
EDIT: Which is one of the reasons why product documentation is so very critical. The manufacture should provide this type of information at the time a product is released.
Yes, but if that code were here, it could then be discussed in the way we do now. That's the point. There is no intent to control everybody and all code, just preserve the forum as the open exchange of ideas it currently is.
There are all sorts of things possible. The discussion isn't about that. It's about the practical and what best serves people here. MIT or No License at all, as Jazzed mentioned does that. Every single one of us has gained so much from how we do things. It's entirely reasonable to question changing that, which several of us are. There won't be as much benefit if we continue to allow other licenses in the discussion here.
Please don't misunderstand me.... I am all for a licensing standard... I just don't think all code should require a license. I normally just state that people are free to do what they want with my code, which is basically open season. However with the exception of that, I believe there should be one license for the OBEX and such things.
While I do not believe Wikipedia is the ultimate authority, I will concede that a "use of a copyright notice" is optional now and not required to assert copyrights. This is useful to know - the issue comes up in music all the time.
I understand the desire to be pure and simple but people operating like that have been taken for a ride big time! Really, if someone is wanting to make money then it's their responsibility to follow the regulations.
Also, bare in mind that ideas are not copyrighted only the verbatim text. I suspect this is why we never got a single hint of how that retired professor was going to implement AI with his wacky multi-noded memory extensions on the Prop1 ... anyway ...
Usually, any code of significance, ie: not just rehashed existing code, is going to evolve into something more substantial/complete before it becomes usable and presumably filed with an MIT license attached. Until then, it's copyrighted. That gives those that are forming the finished code the protection they fully deserve.
If it is just rehashed, with no attached license, then you can go get the original source, which is presumably, in the case of Props, MIT licensed.
PS: The content of whole online posts are also copyrighted as far as I know.
My wife spent years writing for small press (back before the internet killed her gig) and I have sold photos and some short articles. There is both a legal and a practical side to copyright, both for copyright holders and for potential infringers.
There aren't copyright, patent, or trademark police who prowl the interwebs looking for infringements to bust. All those laws do is give you legal standing to sue for damages. This means, for the law to mean anything at all, (1) the copyright holder has to feel damaged, and (2) he has to feel he is damaged enough to justify mounting a lawsuit to recover those damages.
If there is no registration, then the maximum damages one can sue for are actual damages -- the fundamental carrot to get you to register your work is the right to sue for punitive damages.
It is probably safe to assume that nothing posted to this forum has been registered. (If it has, and isn't accompanied by a proper notice, shame on the author; their lack of compliance will make prevailing at trial very hard.) So, if you rip off someone's code here, what is the actual damage for which they could sue you? Unless you include it in your hot new smartphone design the answer is probably barely going to cross the threshold for small claims court. Even if Parallax or Sparkfun sold something based on your work you would have to establish what fraction of the profit from those sales is down to your work, and that's what you would theoretically be owed.
Copyright becomes important if your work is distributed to thousands, or better tens or hundreds of thousands, of people, and if it represents a significant fraction of the value of the sale (such as being a featured article in a magazine). The most infringy thing that's likely to happen here is that your code will end up in someone's homebuilt robot.
I have had this conversation with my own corporate overlords and they are cool with me posting code here even though I assume anything I post here will escape my control. I regard posting code here as an opportunity to seek help and/or advance the state of the art. Advancing the art helps us because a robust market for Propeller based stuff gives Parallax incentive to keep the chip in production and others incentive to create support stuff that might be useful to us. And when I posted my three-cog ROM font VGA driver, kuroneko tightened the inner loop and improved it. Win-win. Meanwhile, if someone else finds that driver useful it doesn't adversely affect me because there's lots more code in our actual commercial products which I'm not releasing.
Now, some of the stuff that's been done very recently (Bill's LMM experiments *cough*) is on a border between the world of copyright and the world of patent. Copyright is quite clear that you cannot copyright an idea, only the fixed expression of an idea. One could have a lively debate over whether something like a LMM kernel is a publication or invention more appropriate for patent. Something like a LMM kernel is more likely to be come the basis for something very widespread,and therefore likely to merit action if it is infringed. This probably makes it a good idea for Bill to stake his claim for such things if he wants to be reimbursed for their use in the future.
But that is a very unusual situation. For the most part, even for those of us who might be jealous of our code, it is very unlikely that anything will be done with code posted here that we would find worth exercising our rights over. If you're posting code you think might be that valuable, be prudent and post a license, and if someone goes to that effort respect their space; you can still use their code in your home robot or very limited run product without much worry but don't start marketing your own compiler based on it without permission. If you post a prototype (or finished) development system, as I might soon, do so with the expectation that people might use it to develop products; but if someone does that and puts a license or copyright notice on it, don't think it's OK to replace their name with yours in the attributes and start selling it for money.
It really isn't rocket surgery. Some of it, like the automatic copyright at creation, is a little counter-intuitive if you aren't used to it, but it's been around for a very long time. And while the continual Mickey Mouse term period increases are a travesty, a more reasonably termed copyright is actually a good thing that encourages content creation.
Gees, If someone used my code I posted here I would be honored!!!! If, however they sold it for a huge profit then I would expect to be compensated, but the code I write is nothing fancy so I doubt I will ever have to worry about it.
I guess any problem I have with Bill's "claims" boils down to these points:
1. Those who care to take advantage of the opportunity are very privileged to have Chip's advance FPGA code to work with. Staking a proprietary claim on any work they do with it would seem to be an abuse of that privilege.
2. Being first with something is no guarantee that it's non-obvious. I'm sure that whatever Bill did would easily occur to many of us, given the time and the need. Same applies to the basic LMM concept (sorry, Bill).
It's not my intent to demean Bill's enthusiasm or contributions to the state of the Propeller art by saying these things -- far from it. But to stick a flag in the sand this early -- before the P2 is even released -- then to label it a "Christmas gift" does seem a bit presumptuous. Moreover, it's just not clear to me what advantage there is to be gained by him or by the community at large in doing so.
Phil: Being first with something is no guarantee that it's non-obvious.
Obviousness has nothing to do with copyright. That is a patent argument. Patents and copyrights are very, very different, even though there is such a thing as a software patent (and that a thing lots of people would argue was a mistake and should not exist).
The courts (in the US at least) have taken a dim view of trying to use copyright law in place of patent. Perhaps the most egregious excess was the printer manufacturers who put chips in their ink cartridges to enforce replacement and "correct" proprietary refills, and claimed copyright infringement when ink refillers reverse-engineered the comms from the cartridge to the printer to allow third party replacements. The courts shot that down toot sweet. Copyright does not exist to protect your monopoly on printer ink.
I seriously doubt a short snippet of code like a LMM kernel would be protected at all by copyright. The specific source code for an implementation might, but change up the comments and reindent it and it's not the same work. If an entire application were hijacked it would be different -- if you had tens of K of even object code that was identical and you could show that it would be very unlikely that another clean room architect would come to that very exact design, then you'd have some ground for argument. But twenty lines of code tightly constrained by pipeline mechanisms and instruction timings? That's not a copyright work. That's an invention.
Not that I blame Bill either for staking a claim on his considerable and innovative work. But that just isn't how IP law works. You can patent an idea, but you cannot copyright one. You can copyright a complex system like PropGCC or Tachyon or Catalina, because it's very unlikely that someone else embarking on the same creative path will come up with *exactly* the same result. (That's why we have four Forths now.) A LMM interpreter is more of a concept that anyone "skilled in the arts" is likely to converge on, given enough skill and effort. That can be patented if you get to it first, but it can't be copyrighted.
What is bad if any of You
> Will have little credit/recognition.
I live in "Nobl's country" -- So don't say me it is not important.
And I see like debate in TV, Newspaper every Year
> Who was first with what and like and I'm are very quibble on that.
Have people/humans that big problems give credits to others for theirs work.
Have people/humans that big problems give credits to others for theirs work.
Of course not! But yielding credit as a courtesy and laying claim to what might be an obvious concept and demanding credit for it as a condition of use are two entirely different things.
Of course not! But yielding credit as a courtesy and laying claim to what might be an obvious concept and demanding credit for it as a condition of use are two entirely different things.
Anyone interested enough in protecting their code to be considering anything other than a MIT-like license should probably get themselves a copy of Patent It Yourself from nolo.com:
Even if you're thinkng copyright or trademark instead of patent, it has an excellent rundown of what is covered by which body of law in the first few chapters. It will give you a clear idea whether your concept should be patented, copyrighted, or trademarked, and even if patent isn't the right outcome for you the guidance on that is better than I've seen from any other source.
"obvious concept" --- For that ones that You and me.
But for how many more ?
Patent law (for which this matters, unlike copyright) makes this pretty clear -- it's obvous to someone "skilled in the arts." Yeah that sounds mushy but the way it's usually interpreted it means that in Propellerland if Phil or Bill or Chip or I could figure it out easily and without help then it's "obvious." If it's a patent for a nuclear weapon feature then they don't mean obvious to Phil or Bill or Chip or me, they mean whether it's obvious to someone like Ed Teller who is conversant with the field.
One thing that's been left out of this discussion is the difference between an invention and a discovery. When probing the capabilities of a new chip like the P2, I would submit that many of the techniques that result from such probing should be categorized as discoveries, akin to first sighting the rings of Saturn or inferring the structure of DNA. Chip has provided a tool that makes such early discoveries possible. But it's a far intellectual stretch and more than the few weeks time since this tool has been available before such discoveries can lead to credible inventions. With the P2, those who are taking advantage of the FPGA preview must, by definition, still be in the "science" phase, not the "engineering" phase.
Patent law (for which this matters, unlike copyright) makes this pretty clear -- it's obvous to someone "skilled in the arts." Yeah that sounds mushy but the way it's usually interpreted it means that in Propellerland if Phil or Bill or Chip or I could figure it out easily and without help then it's "obvious." If it's a patent for a nuclear weapon feature then they don't mean obvious to Phil or Bill or Chip or me, they mean whether it's obvious to someone like Ed Teller who is conversant with the field.
One thing that's been left out of this discussion is the difference between an invention and a discovery.
The big thing that's been left out of this is the difference betwen patent, copyright, and (not mentioned yet, but it always gets dragged in) trademark.
If Bill has a claim on code he has posted here, unless he has registered for patent or trademark protection he is claiming copyright. Copyright doesn't give a hoot about obviousness. Copyright is nice because it issues, like magic, the moment you create the work. Cool! But it doesn't protect ideas or simple iconic images.
With regard to what the FPGA owners are doing, almost all of that is going to be classified as ideas. It really doesn't matter if they are "inventions" or "discoveries" because neither of those things can be copyrighted. Both can, under the right circumstances, be patented, but that's not a thing that happens automatically when you record the idea, it's something you have to pay to apply for and there's a pretty steep fee and lengthy process involved.
I still don't understand why it is so disturbing what type of License TEXT he used --->
Some people are purists about "freedom of information." The thing is, even if someone puts code hre with a notice that says WINGED MONKEYS WILL COME FOR YOU IF YOU USE THIS CODE, unless you believe they really have winged monkeys you're safe using it for your personal robot project.
I suspect what is going on is Bill has been advised (I've met the advisor) to explicitly put terms on the code he posts, especially when it involves a lot of effort. This is not a bad idea. But there are limits to what that gains Bill (not zero, but limited) and to what it means for you if you want to use his code (don't make it the basis for your smartphone or compiler without paying him the usage fee).
It's kind of silly because I don't think what Bill is trying to protect here can really be protected by copyright law, and everyone is getting all het up because he's asserting such rights as he might have, which if you look closely he probably doesn't have.
So substance of all debate here is
> Don't post anything even if that help people to understand programming on Propeller I/ Propeller II as that automatically give others free usage in theirs business for free without any type of recognition.
Ps. You will work --- Others will make MONEY for that!
Some people are purists about "freedom of information." The thing is, even if someone puts code hre with a notice that says WINGED MONKEYS WILL COME FOR YOU IF YOU USE THIS CODE, unless you believe they really have winged monkeys you're safe using it for your personal robot project.
I suspect what is going on is Bill has been advised (I've met the advisor) to explicitly put terms on the code he posts, especially when it involves a lot of effort. This is not a bad idea. But there are limits to what that gains Bill (not zero, but limited) and to what it means for you if you want to use his code (don't make it the basis for your smartphone or compiler without paying him the usage fee).
It's kind of silly because I don't think what Bill is trying to protect here can really be protected by copyright law, and everyone is getting all het up because he's asserting such rights as he might have, which if you look closely he probably doesn't have.
Before you proceed, you shoud realize that I am not a lawyer, this article is not to be construed as legal advice.
As far as I can tell, the reason for the disstress of some forum members is that I used an open source license that was not to their liking.
I don't even want to imagine the outcry if I had used a less permissive license.
For the record, my choice of license was to:
- preclude any entity from trying to patent the concept by clearly establishing prior art
- preclude any entiry from coming up with extremely similar code and posting it with a very restrictive license
- get credit for my work, which required a lot of thought and experimentation to come up with a nearly optimal byte code execution engine
In order to avoid any possible misunderstanding, Parallax and Catalina have been good about providing credit; as have the majority of other projects - but even one bad apple leaves a bad taste.
What did I learn from the reactions I received?
- There is a woeful lack of understanding of intellectual property laws
- there is a lack of respect for those trying to do good
- some of those complaining about "trying to force attributions" have no qualms about trying to force others to change their choice of license
Thankfully, a shining few have appreciated what I was trying to do, and understood why I chose the license I did, and even if they would have chosen otherwise, respected my right to do so. I will not name any of you here, but know that you have my thanks.
The compaints against the very permissive license I chose generally fell into the following broad categories:
1) you cannot copyright an idea - this is true, but you CAN copyright an expression of an idea, which becomes very useful if a patent troll tries to later patent the idea, as a posting on a public forum can be used to show prior art as some mentioned.
2) the idea is too simple / obvious / small to deserve protection. Where do we draw the line? 100 lines of trivial code? 10 lines of trivial code? 3-5 lines of clever code?
3) Using a CLUT/STACK to hold code and using it to generate a highly optimized byte code engine is obvious: To whom exactly? To the exact cycle count, with the exact instruction sequence? Which could not be determined until many cycle counting pipeline experiments were run? With the experiments being run with publicly available tools and information?
4) the idea may be very clever, but it is a direct derivation of LMM. Even if that were true, that would be irrelevant. Who came up with LMM?
5) "Bill discussed LMM on the forum, and received feedback from others, therefore he cannot claim any IP rights on P2BEE as it is derived from LMM." - I disagree.
6) "This will hurt Parallax, as this code can not be used in GCC without attribution, which is incompatible with the GPL." - My reading of the "GCC Runtime Library Exception, Version 3.1, 31 March 2009" is that this is not correct if I give them permission to include it.
7) "The MIT license already requires attribution" - mostly true for source distribution, assuming the licensed item meets the definiton of being significant enough. In whose opinion? Would you care to bet that someone who did not want to attribute would set as the bar for "significant enough" very high?
General observations
In my humble opinion, it is in Parallax's interest to foster interaction and posting of code in the forums, as long as the license terms used are not very onerous. Currently, there is no requirement that postings be placed under an MIT license, which is as it should be. Other non-restrictive licenses should be encouraged, so a wider range of individuals will feel comfortable posting their code, with whatever level legal protection they desire.
Those advocating "forcing" an MIT license should consider the questionable legality of such a tactic. Frankly, requiring MIT is already hampering the growth of Obex, and will continue to deter professional developers from contributing objects.
I was very dissapointed to see a few forum members descend to the level of setting up straw man arguments in an attempt to influence others to their point of view, or even worse, descending to thinly veiled ad-hominem attacks.
If any reader is uncomfortable with any license, exactly what stops the reader from politely asking - in the forum, via e-mail, or pm - the author if they might make work available to that reader under a license more amenable to the reader?
Final thoughts:
Given that I am assured by some forum members that my software innovations are simple and obvious, and that they - or someone else - would have independently come up with the same innovations without my contribution, I see little reason to post about any future software innovation I may come up with in this forum.
Additionally, I will no longer discuss licensing on this forum
Comments
Well now, that is a whole other debate.
Yes, the default of MIT is what I would suggest.
Parallax was nice enough to do an early release of the P2 on the FPQA so that we could get a jump start and get some work done on tools and documentation.
I don't think the intent was so people could stake claim to the usage of new features and lock them down with non-MIT licensing.
We normally have very open discussions and "what's discussed in the forum is open to the forum" seems to be the norm.
We have the thread on LMM for the Prop 2 where Ariba, Chip, and others have been working with Bill on possible versions of LMM2.
It seems that a lot of the pipeline work that led Bill to his byte code engine is the result of that thread and his later optimization likely was the result of Chip's suggestion for improvement.
I was pretty pumped up about the prop 2, and the community effort that seemed to be building around LMM2 was pretty cool, not so much after today.
C.W.
We've had almost no instances of alternative license code posted here. Before that gets messy, it would be very good to set default expectations. Dropping code here that people cannot view and use without worries isn't well aligned with the core purpose and value of this forum.
Let's assume your idea is correct, people should post code to the forum, and no one should make use of said code without seeking permission from the person that posted it.
The Prop 2 isn't even released yet, there is very little code and documentation.
Let's say I hurry up and write out some code snippets for initializing the counters and other really common activities.
It's a good bet that since there isn't much of a code base yet I'll have some unique, albeit obvious code.
By your standard I can become the gateway to usage of the counters...Do we really want that?
C.W.
C.W., I have pondered such things on numerous occassions, particularly pertaining to items such as source code, and I would imagine the courts have heard numerous cases pertaining to simple programming issues. However, for the sake of discussion, the site requirement for a license would in no way hinder this type of activity. For instance, as in your example, I could simply create the code and post it to my own web site or any other web site, thus protecting it. Would such trivial items be enforced in a court of law? I have no idea, but I have definitely pondered such things.
EDIT: Which is one of the reasons why product documentation is so very critical. The manufacture should provide this type of information at the time a product is released.
There are all sorts of things possible. The discussion isn't about that. It's about the practical and what best serves people here. MIT or No License at all, as Jazzed mentioned does that. Every single one of us has gained so much from how we do things. It's entirely reasonable to question changing that, which several of us are. There won't be as much benefit if we continue to allow other licenses in the discussion here.
That post had me confused until I took a look at Bill's link and found this:
Also, bare in mind that ideas are not copyrighted only the verbatim text. I suspect this is why we never got a single hint of how that retired professor was going to implement AI with his wacky multi-noded memory extensions on the Prop1 ... anyway ...
Usually, any code of significance, ie: not just rehashed existing code, is going to evolve into something more substantial/complete before it becomes usable and presumably filed with an MIT license attached. Until then, it's copyrighted. That gives those that are forming the finished code the protection they fully deserve.
If it is just rehashed, with no attached license, then you can go get the original source, which is presumably, in the case of Props, MIT licensed.
PS: The content of whole online posts are also copyrighted as far as I know.
There aren't copyright, patent, or trademark police who prowl the interwebs looking for infringements to bust. All those laws do is give you legal standing to sue for damages. This means, for the law to mean anything at all, (1) the copyright holder has to feel damaged, and (2) he has to feel he is damaged enough to justify mounting a lawsuit to recover those damages.
If there is no registration, then the maximum damages one can sue for are actual damages -- the fundamental carrot to get you to register your work is the right to sue for punitive damages.
It is probably safe to assume that nothing posted to this forum has been registered. (If it has, and isn't accompanied by a proper notice, shame on the author; their lack of compliance will make prevailing at trial very hard.) So, if you rip off someone's code here, what is the actual damage for which they could sue you? Unless you include it in your hot new smartphone design the answer is probably barely going to cross the threshold for small claims court. Even if Parallax or Sparkfun sold something based on your work you would have to establish what fraction of the profit from those sales is down to your work, and that's what you would theoretically be owed.
Copyright becomes important if your work is distributed to thousands, or better tens or hundreds of thousands, of people, and if it represents a significant fraction of the value of the sale (such as being a featured article in a magazine). The most infringy thing that's likely to happen here is that your code will end up in someone's homebuilt robot.
I have had this conversation with my own corporate overlords and they are cool with me posting code here even though I assume anything I post here will escape my control. I regard posting code here as an opportunity to seek help and/or advance the state of the art. Advancing the art helps us because a robust market for Propeller based stuff gives Parallax incentive to keep the chip in production and others incentive to create support stuff that might be useful to us. And when I posted my three-cog ROM font VGA driver, kuroneko tightened the inner loop and improved it. Win-win. Meanwhile, if someone else finds that driver useful it doesn't adversely affect me because there's lots more code in our actual commercial products which I'm not releasing.
Now, some of the stuff that's been done very recently (Bill's LMM experiments *cough*) is on a border between the world of copyright and the world of patent. Copyright is quite clear that you cannot copyright an idea, only the fixed expression of an idea. One could have a lively debate over whether something like a LMM kernel is a publication or invention more appropriate for patent. Something like a LMM kernel is more likely to be come the basis for something very widespread,and therefore likely to merit action if it is infringed. This probably makes it a good idea for Bill to stake his claim for such things if he wants to be reimbursed for their use in the future.
But that is a very unusual situation. For the most part, even for those of us who might be jealous of our code, it is very unlikely that anything will be done with code posted here that we would find worth exercising our rights over. If you're posting code you think might be that valuable, be prudent and post a license, and if someone goes to that effort respect their space; you can still use their code in your home robot or very limited run product without much worry but don't start marketing your own compiler based on it without permission. If you post a prototype (or finished) development system, as I might soon, do so with the expectation that people might use it to develop products; but if someone does that and puts a license or copyright notice on it, don't think it's OK to replace their name with yours in the attributes and start selling it for money.
It really isn't rocket surgery. Some of it, like the automatic copyright at creation, is a little counter-intuitive if you aren't used to it, but it's been around for a very long time. And while the continual Mickey Mouse term period increases are a travesty, a more reasonably termed copyright is actually a good thing that encourages content creation.
EDIT: OK, So who is Bill and why all the ranting?
2. Being first with something is no guarantee that it's non-obvious. I'm sure that whatever Bill did would easily occur to many of us, given the time and the need. Same applies to the basic LMM concept (sorry, Bill).
It's not my intent to demean Bill's enthusiasm or contributions to the state of the Propeller art by saying these things -- far from it. But to stick a flag in the sand this early -- before the P2 is even released -- then to label it a "Christmas gift" does seem a bit presumptuous. Moreover, it's just not clear to me what advantage there is to be gained by him or by the community at large in doing so.
-Phil
Obviousness has nothing to do with copyright. That is a patent argument. Patents and copyrights are very, very different, even though there is such a thing as a software patent (and that a thing lots of people would argue was a mistake and should not exist).
The courts (in the US at least) have taken a dim view of trying to use copyright law in place of patent. Perhaps the most egregious excess was the printer manufacturers who put chips in their ink cartridges to enforce replacement and "correct" proprietary refills, and claimed copyright infringement when ink refillers reverse-engineered the comms from the cartridge to the printer to allow third party replacements. The courts shot that down toot sweet. Copyright does not exist to protect your monopoly on printer ink.
I seriously doubt a short snippet of code like a LMM kernel would be protected at all by copyright. The specific source code for an implementation might, but change up the comments and reindent it and it's not the same work. If an entire application were hijacked it would be different -- if you had tens of K of even object code that was identical and you could show that it would be very unlikely that another clean room architect would come to that very exact design, then you'd have some ground for argument. But twenty lines of code tightly constrained by pipeline mechanisms and instruction timings? That's not a copyright work. That's an invention.
Not that I blame Bill either for staking a claim on his considerable and innovative work. But that just isn't how IP law works. You can patent an idea, but you cannot copyright one. You can copyright a complex system like PropGCC or Tachyon or Catalina, because it's very unlikely that someone else embarking on the same creative path will come up with *exactly* the same result. (That's why we have four Forths now.) A LMM interpreter is more of a concept that anyone "skilled in the arts" is likely to converge on, given enough skill and effort. That can be patented if you get to it first, but it can't be copyrighted.
What is bad if any of You
> Will have little credit/recognition.
I live in "Nobl's country" -- So don't say me it is not important.
And I see like debate in TV, Newspaper every Year
> Who was first with what and like and I'm are very quibble on that.
Have people/humans that big problems give credits to others for theirs work.
-Phil
"obvious concept" --- For that ones that You and me.
But for how many more ?
http://www.nolo.com/products/patent-it-yourself-pat.html
Even if you're thinkng copyright or trademark instead of patent, it has an excellent rundown of what is covered by which body of law in the first few chapters. It will give you a clear idea whether your concept should be patented, copyrighted, or trademarked, and even if patent isn't the right outcome for you the guidance on that is better than I've seen from any other source.
Patent law (for which this matters, unlike copyright) makes this pretty clear -- it's obvous to someone "skilled in the arts." Yeah that sounds mushy but the way it's usually interpreted it means that in Propellerland if Phil or Bill or Chip or I could figure it out easily and without help then it's "obvious." If it's a patent for a nuclear weapon feature then they don't mean obvious to Phil or Bill or Chip or me, they mean whether it's obvious to someone like Ed Teller who is conversant with the field.
-Phil
I still don't understand why it is so disturbing what type of License TEXT he used --->
He don't said anyone need PAY for it -- Only recognise his work.
If anyone have Problems with that -- So write its own routines --> What are problem.
The big thing that's been left out of this is the difference betwen patent, copyright, and (not mentioned yet, but it always gets dragged in) trademark.
If Bill has a claim on code he has posted here, unless he has registered for patent or trademark protection he is claiming copyright. Copyright doesn't give a hoot about obviousness. Copyright is nice because it issues, like magic, the moment you create the work. Cool! But it doesn't protect ideas or simple iconic images.
With regard to what the FPGA owners are doing, almost all of that is going to be classified as ideas. It really doesn't matter if they are "inventions" or "discoveries" because neither of those things can be copyrighted. Both can, under the right circumstances, be patented, but that's not a thing that happens automatically when you record the idea, it's something you have to pay to apply for and there's a pretty steep fee and lengthy process involved.
" those who are taking advantage of the FPGA preview must, by definition, still be in the "science" phase, not the "engineering" phase"
Chip posted FPGA code on Forum --- Not to any of us specifically so all have same chance to that work.
Some people are purists about "freedom of information." The thing is, even if someone puts code hre with a notice that says WINGED MONKEYS WILL COME FOR YOU IF YOU USE THIS CODE, unless you believe they really have winged monkeys you're safe using it for your personal robot project.
I suspect what is going on is Bill has been advised (I've met the advisor) to explicitly put terms on the code he posts, especially when it involves a lot of effort. This is not a bad idea. But there are limits to what that gains Bill (not zero, but limited) and to what it means for you if you want to use his code (don't make it the basis for your smartphone or compiler without paying him the usage fee).
It's kind of silly because I don't think what Bill is trying to protect here can really be protected by copyright law, and everyone is getting all het up because he's asserting such rights as he might have, which if you look closely he probably doesn't have.
So substance of all debate here is
> Don't post anything even if that help people to understand programming on Propeller I/ Propeller II as that automatically give others free usage in theirs business for free without any type of recognition.
Ps. You will work --- Others will make MONEY for that!
As far as I can tell, the reason for the disstress of some forum members is that I used an open source license that was not to their liking.
I don't even want to imagine the outcry if I had used a less permissive license.
For the record, my choice of license was to:
- preclude any entity from trying to patent the concept by clearly establishing prior art
- preclude any entiry from coming up with extremely similar code and posting it with a very restrictive license
- get credit for my work, which required a lot of thought and experimentation to come up with a nearly optimal byte code execution engine
In order to avoid any possible misunderstanding, Parallax and Catalina have been good about providing credit; as have the majority of other projects - but even one bad apple leaves a bad taste.
What did I learn from the reactions I received?
- There is a woeful lack of understanding of intellectual property laws
- there is a lack of respect for those trying to do good
- some of those complaining about "trying to force attributions" have no qualms about trying to force others to change their choice of license
Thankfully, a shining few have appreciated what I was trying to do, and understood why I chose the license I did, and even if they would have chosen otherwise, respected my right to do so. I will not name any of you here, but know that you have my thanks.
The compaints against the very permissive license I chose generally fell into the following broad categories:
1) you cannot copyright an idea - this is true, but you CAN copyright an expression of an idea, which becomes very useful if a patent troll tries to later patent the idea, as a posting on a public forum can be used to show prior art as some mentioned.
2) the idea is too simple / obvious / small to deserve protection. Where do we draw the line? 100 lines of trivial code? 10 lines of trivial code? 3-5 lines of clever code?
3) Using a CLUT/STACK to hold code and using it to generate a highly optimized byte code engine is obvious: To whom exactly? To the exact cycle count, with the exact instruction sequence? Which could not be determined until many cycle counting pipeline experiments were run? With the experiments being run with publicly available tools and information?
4) the idea may be very clever, but it is a direct derivation of LMM. Even if that were true, that would be irrelevant. Who came up with LMM?
5) "Bill discussed LMM on the forum, and received feedback from others, therefore he cannot claim any IP rights on P2BEE as it is derived from LMM." - I disagree.
6) "This will hurt Parallax, as this code can not be used in GCC without attribution, which is incompatible with the GPL." - My reading of the "GCC Runtime Library Exception, Version 3.1, 31 March 2009" is that this is not correct if I give them permission to include it.
7) "The MIT license already requires attribution" - mostly true for source distribution, assuming the licensed item meets the definiton of being significant enough. In whose opinion? Would you care to bet that someone who did not want to attribute would set as the bar for "significant enough" very high?
General observations
In my humble opinion, it is in Parallax's interest to foster interaction and posting of code in the forums, as long as the license terms used are not very onerous. Currently, there is no requirement that postings be placed under an MIT license, which is as it should be. Other non-restrictive licenses should be encouraged, so a wider range of individuals will feel comfortable posting their code, with whatever level legal protection they desire.
Those advocating "forcing" an MIT license should consider the questionable legality of such a tactic. Frankly, requiring MIT is already hampering the growth of Obex, and will continue to deter professional developers from contributing objects.
I was very dissapointed to see a few forum members descend to the level of setting up straw man arguments in an attempt to influence others to their point of view, or even worse, descending to thinly veiled ad-hominem attacks.
If any reader is uncomfortable with any license, exactly what stops the reader from politely asking - in the forum, via e-mail, or pm - the author if they might make work available to that reader under a license more amenable to the reader?
Final thoughts:
Given that I am assured by some forum members that my software innovations are simple and obvious, and that they - or someone else - would have independently come up with the same innovations without my contribution, I see little reason to post about any future software innovation I may come up with in this forum.
Additionally, I will no longer discuss licensing on this forum