Ondrej Certik wrote: > If you are not interesting in licensing, you can skip this. > >> sorry, you're absolutely correct. I'm a long term debian user and I >> tend to say one or the other as a short hand which is a terrible habit >> because they are completely different things.... I'll check out the >> differences between the BSD and GPL (and subversions) and all that red >> tape required to coerce people into being reasonably well mannered,and >> decide when I come to release it. Basically I want a very strong >> license so that anyone who uses it has to be free and respectful about >> it...I don't want people putting a GUI on my masters project and then >> trying to claim it is all theirs and sell it or something, I also >> don't want some cheeky company taking my ideas and patenting them as >> if they were their own (and then possibly suing me for using 'their' >> idea some way down the line -- apparently this is happening) > > Feel free to discuss any licensing issues or concerns here. > > Basically there are very good arguments for GPL (you stated some) and > there are also very good arguments for BSD, see for example: > > http://www.scipy.org/License_Compatibility > > The only important thing from my side is that all SymPy developers and > users can agree on some common interests, so that we can produce > something useful together, instead each of us working on our own thing > only. > > I can also offer my own personal view (but other sympy developers > doesn't have to agree fully with me here): I generally prefer less > restrictions than more restricitons, so that the code I write can be > used by more people any way they like. I want them to acknowledge that > though and both BSD and GPL require that. > The thing I don't like on GPL is that it forces the user to also use a > GPL. So technically if you use GPL2, the user cannot license his > program as GPL3, because they are not compatible. You can use GPL3, > but then the user cannot use GPL2 programs. You can use GPL2 or later, > but then you are not stating the precise terms to copy/redistribute > your program, but rather giving the future of your program into the > hands of FSF, i.e. whatever license they create in the future could be > applied to your code as well. But see also below. > >> it...I don't want people putting a GUI on my masters project and then >> trying to claim it is all theirs and sell it or something, I also > > I think both BSD and GPL require them to give credit to you. Otherwise > I don't mind anyone building on top of my code and improving it. See > below about contributing back. > >> don't want some cheeky company taking my ideas and patenting them as >> if they were their own (and then possibly suing me for using 'their' >> idea some way down the line -- apparently this is happening) > > I am not a lawyer, but it seems to me it is not possible to patent a > code that someone else has written and released as BSD or GPL, is it? > > There is also a question which license is the best to encourage people > to contribute the code back to the project and to encourage people to > work together on one thing, instead of forking things and doing things > on their own. Linus says it's GPL2, but I think more important than > the license itself is the way the project is handled and how many > people use it. I think the motivation to work on one thing doesn't > stem from the license, but rather because it is more effective for > everyone involved. > > Nevertheless, I am not 100% convinced either way, I can see good > arguments for both ways.
IMHO the things are different depending on what the purpose of your code is. In the case of a fundamental library (in the sense of basic research), for example, a general symbolic engine, a finite element (FE) engine, a plotting library, the less restrictive license the better - here one wants a large user base, lots of developers/contributors, and thus also a mutual _compatibility_ of projects built on top of that library (free or not). This is exactly the case of sympy, by the way. On the other hand, when developing a special stuff (e.g. applying a general FE code to some particular deep problem) it is not so clear to me, all depends on what the author wants of her code to become. Just a historical remark: I have initially released SfePy (a FE code) under GPL, but following the discussions on Numpy/SciPy/Matplotlib/IPython mailing lists I have changed the license to BSD. Then Ondrej (and others) joined and things started to accelerate :) This may also put some weight on the balances - it seems to me that in the world of Python scientific libraries the BSD license is more usual (just a hunch, I have no numbers.) r. --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "sympy" group. To post to this group, send email to sympy@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sympy?hl=en -~----------~----~----~----~------~----~------~--~---
