Summary
Why don't open source developers just put their code into the public domain?
Advertisement
I am a supporter of public domain code, but I am not a big fan of open source licenses.
The GPL is thankfully becoming outmoded, but some of the current popular licences which are considered "commercially friendly" such as the Boost License and the BSD License are still scary and hard to work with. What license do I use? How do I combine the licenses? What does "derived work" mean? Am I in danger of being sued?
It is becoming increasingly common to see source files containing more warnings and copyright information than actual code. What purpose do the commercially friendly open-source licenses really serve anyway?
Most open-source licences legally obligate attribution, but any fair and ethical use of source code would naturally provide attribution. Only a small fraction of people would pass public domain code off as their own, but so what?
Theoretically the licenses are supposed to protect authours from liability, but the notion of holding an authour of public domain code liable for damages seems to me ludicrous. We don't write physics equations or chemistry formulas with disclaimers, so why do we need disclaimers for source code?
I believe that if more developers started putting code and algorithms in the public domain and without disclaimers, it could lead to the technology of software development progressing at the pace it deserves and help overcome the fear seeded in our hearts by overly litigious corporate entities.
I could be wrong, but I think that if you release something into the public domain without any liability limiting statement, well, you're still liable if something goes wrong.
If you throw a rake on the sidewalk and tell people you don't it, it doesn't make you any less liable if someone trips over it.
> Why don't you place your blog entries in the public domain > instead of copyrighting them and reserving all rights?
I don't really have control over the copyright notice since I am using the Artima system. Honestly though I don't care if my blog entries are public domain or not. I could perhaps make this an official declaration that all of my blog entries are public domain.
If you want better examples of my hypocracy however just go to http://www.OOTL.org . Most everything there is currently released under the Boost license. The problem is that the OOTL adapts portions of boost code.
> I could be wrong, but I think that if you release > something into the public domain without any liability > limiting statement, well, you're still liable if something > goes wrong.
I know that this is a commonly held belief. I am not familiar with the US legal system (assuming that is where you are based) so I can't comment on the actual legality, but logically it is an absurd notion.
Code is something that is self-describing. It does what it does, there is no room for interpretation. If one person assumes that source code does something other than what it does how can the authour be held accountable for the misconceptions of the user?
> If you throw a rake on the sidewalk and tell people you > don't it, it doesn't make you any less liable if someone > trips over it.
I think this is not a good analogy. In order for public domain code to cause harm it would have to be copied, compiled, and executed. This requires several active steps on behalf of the injured party.
Code released into the public domain should be treated like text, or equations. It is completely harmless in of itself, but of great potential benefit or injury depending on how it is used.
I tend to agree with you. Working with open source is getting tougher and tougher with the plethora of licenses out there. It feels like you need a lawyer before you start to code.
> I tend to agree with you. Working with open source is > getting tougher and tougher with the plethora of licenses > out there. It feels like you need a lawyer before you > start to code.
> I know that this is a commonly held belief. I am not > familiar with the US legal system (assuming that is where > you are based) so I can't comment on the actual legality, > but logically it is an absurd notion. > > Code is something that is self-describing. It does what it > does, there is no room for interpretation. If one person > assumes that source code does something other than what it > does how can the authour be held accountable for the > misconceptions of the user?
I hear you on the logic, but with all respect, you're expecting the legal process to be rational. Why would it possibly be rational? Ladders are self-describing too. :)
> > If you throw a rake on the sidewalk and tell people you > > don't it, it doesn't make you any less liable if > someone > > trips over it. > > I think this is not a good analogy. In order for public > domain code to cause harm it would have to be copied, > compiled, and executed. This requires several active steps > on behalf of the injured party. > > Code released into the public domain should be treated > like text, or equations. It is completely harmless in of > itself, but of great potential benefit or injury depending > on how it is used.
And, there's the rub.
The fact is, we don't know the answer. Without consulting a lawyer, I'd definitely add the disclaimer. If five lines of text has the potential to prevent an adverse judgement, to me that's a win by any measure.
> Without consulting > a lawyer, I'd definitely add the disclaimer. If five > lines of text has the potential to prevent an adverse > judgement, to me that's a win by any measure.
I respect your point of view but I personally would prefer to take a stand for common sense and logic, and fight back against the tyranny of the threat of litigation, which I feel poses a significant threat to the progress of computer science.
1) people would publish their code more if there weren't so many licences
2) the licences are often chosen very deliberately, and dismissing the multitude of licences as silly actually insults the people who use such licences and their choices
Some people believe very strongly that they do not want their code to be usurped by large companies (Mircosoft, Sun, Google, etc.) and choose licenses accordingly.
Others, like the Boost folks, want their code to be usable in all companies, and write their licences accordingly.
Is it that hard to fathom that people don't trust large corporations to act "fair and ethically"?
And, as others mentioned, there are the liability issues.
> I think it's rather naive to assume that: > > 1) people would publish their code more if there weren't > so many licences
I am not suggesting that the number of licenses should be reduced. I was trying to say that sometimes people who are using open-source licenses, should consider releasing code into the public domain instead. If they did so, there might be a higher instance of usage.
> 2) the licences are often chosen very deliberately, and > dismissing the multitude of licences as silly actually > insults the people who use such licences and their > choices
I don't mean to insult anyone. I do realize that a lot of careful deliberation goes into the drafting and choosing of licenses. I will change my wording.
> Some people believe very strongly that they do not want > their code to be usurped by large companies (Mircosoft, > Sun, Google, etc.) and choose licenses accordingly.
I am not sure what you mean specifically by usurped here, but I don't believe public domain code can really be usurped.
> Others, like the Boost folks, want their code to be usable > in all companies, and write their licences accordingly.
Not all companies can use Boost code. The legal ramification of the Boost license are not fully tested nor understood, and as such pose a threat to the future of a company.
Consider the following scenario: Company X builds a software package using a code library based on the Boost libraries which represents a signficant investment of time and energy. A disgrunted employee realizes that the code base is covered by the Boost license and can legally resell the code base to a competitor for a large sum of money without violating their non-disclosure agreement.
> Is it that hard to fathom that people don't trust large > corporations to act "fair and ethically"?
You are correct that you can't trust corporations to act fair and ethically. However the worse thing I can imagine a corporation doing with regards to public domain code, is to illegally claim the public domain code as their own, and threaten to sue people for using it. (AFAIK this currently carries no penalty under US law).
If they actually follow through on their threat, then the defendant can launch a counter-suit for wrongful litigation.
> And, as others mentioned, there are the liability issues.
I find it hard to believe that the average court would really hold a public domain code authour liable for any use of public domain code. Here is a C++ program which I officially declare as being public domain:
#include <cstdlib> int main() { system("format c:"); return 0; }
Now, if someones copies, compiles and executes that program, and then sues me for damages I think that I could probably launch a successful countersuit for filing a frivolous lawsuit.
> I think it's rather naive to assume that: > > 1) people would publish their code more if there weren't > so many licences
Why?
> 2) the licences are often chosen very deliberately, and > dismissing the multitude of licences as silly actually > insults the people who use such licences and their > choices
The multitude of licences is a symptom of their inherent inpenetrability, their unenforcability and the fact that they simply do not do the job they set out to do. The fact that lawyers are required for their interpretation and that no two lawyers will give the same interpretation are clear indicators that software licences are, basically and conceptually, no different to software patents in their usefulness or desirability.
> Some people believe very strongly that they do not want > their code to be usurped by large companies (Mircosoft, > Sun, Google, etc.) and choose licenses accordingly.
A licence whose aim is political, to specifically restrict a set of people from using the software simply beacause you don't like them, is a bad licence.
> Others, like the Boost folks, want their code to be usable > in all companies, and write their licences accordingly.
Public domain software, by definition, has no restrictions on usage. A licence that permits unrestricted usage is a non sequitur.
> Is it that hard to fathom that people don't trust large > corporations to act "fair and ethically"?
Most of this distrust arises from a cobbled together mixture of conspiracy theories, urban myths, uninformed and sensationalist news reporting and political opportunism. It is not the basis for deriving a sensible distribution licence.
> And, as others mentioned, there are the liability issues.
Warnings and disclaimers would cover liability issues. Liability and licencing are separate and unrelated issues.
The US legal system is based on English Common Law, just as Canada's. There are several notable differences, though.
Disclaimers are included in licenses and public domain notices because liability is a separate legal issue from copyright, ownership, or author's moral rights. Just as carrying acceptable car insurance is a separate legal issue from driving while intoxicated, which is separate from smog inspections or not fiddling with the Canadian governor chip.
Your argument that code is what it is sounds quite a bit like the legal argument coming from the fast food industry (http://news.bbc.co.uk/2/hi/americas/3500388.stm basically something like "fatty food, sweet food, etc. is what it is, it is not defective in the same way arsenic-laced food is; we should not be held liable for providing food that fit their requirements if the nature of that food led to obesity, health problems, etc."). It has a certain common sense ring to it.
My father is an attorney, and when he talks about lawsuits, he talks about winning. Assuming that the "it is what it is" argument would win the case, he'd like it. My father-in-law, on the other hand, is a general contractor. When he talks about lawsuits he talks about being sued, even if he won in the end (losing is just adding insult to injury). "It is what it is" just wouldn't be enough for him, even if he knew he could win with it. He'd want a disclaimer to avoid going to court, or to keep any lawsuits very short.
I don't know about Canada, but most US States follow the Uniform Commercial Code (http://straylight.law.cornell.edu/ucc/ucc.table.html), which states that certain warranties exist for anything traded in commerce (usually meaning "originating from a merchant"), UNLESS there is some written document specifically disclaiming those warranties CONSPICUOUSLY, usually in ALL CAPS (yes, I'm being cheeky). Those warranties are the two you always see (merchantability and fitness for a particular use -- sections 2-314 and 2-315.).
> I don't know about Canada, but most US States follow the > Uniform Commercial Code > (http://straylight.law.cornell.edu/ucc/ucc.table.html), > which states that certain warranties exist for anything > traded in commerce (usually meaning "originating from a > merchant"), UNLESS there is some written document > specifically disclaiming those warranties CONSPICUOUSLY, > usually in ALL CAPS (yes, I'm being cheeky). Those > warranties are the two you always see (merchantability and > fitness for a particular use -- sections 2-314 and 2-315.).
Me writing public domain code, and posting it on usenet, could not be reasonably considered something traded in commerce. There is no trade, no commerce, and no merchant. Of course, commercial trade of code (irregardless of the license) would require disclaimers.
Flat View: This topic has 19 replies
on 2 pages
[
12
|
»
]
Previous Topic
