In my career, 90%+ of the code I've written is Free Software (specifically, LGPL or AGPL licensed). That said, most of it is not publicly available - it's custom code written for clients, who could - but for the most part don't - release it publicly. That complies with rms' philosophy, but is it enough for you?
If not, then you have fewer choices - mostly a position on a company like Mozilla, Canonical, Sentry, etc. Finally, some people live on patronage (by companies and/or individuals), but this is generally hard to achieve.
> In my career, 90%+ of the code I've written is Free Software (specifically, LGPL or AGPL licensed). That said, most of it is not publicly available - it's custom code written for clients, who could - but for the most part don't - release it publicly. That complies with rms' philosophy, but is it enough for you?
Hmm, yeah interesting point! I think I would put this in one of those gray areas I was talking about. I've written code for clients, but in those cases the client decided whether and how to license the code, not me.
I'd probably suggest that if neither you nor your clients release the code then it doesn't really fit Stallman's definition of Free Software. Business decisions to not release it publicly are exactly what he's talking about when he claims the code is limiting freedom of the users, without source they can't modify the code.
The piece feels a bit like Stallman against the world though, he's not only alienating most of the people who would like to write Free Software while by day getting paid to write code for their employer, but he's also pitting himself against many of the people who freely release their source too.
> If not, then you have fewer choices..., but this is generally hard to achieve.
Yeah, exactly right. Those few companies have managed to eke out the funding model that perhaps depends on B2B relationships or support contracts, and not direct software sales.
I'd probably suggest that if neither you nor your clients release the code then it doesn't really fit Stallman's definition of Free Software. Business decisions to not release it publicly are exactly what he's talking about when he claims the code is limiting freedom of the users, without source they can't modify the code.
They don't release the code nor the software itself, as it is for internal use, not for redistribution. That's in accordance with the philosophy:
"Distributing a program to users without freedom mistreats those users; however, choosing not to distribute the program does not mistreat anyone. If you write a program and use it privately, that does no wrong to others."
Well, the user in this case is the client who ordered the code. If you become the user, you have all the rights listed in the (A/L/)GPL. I think that's more to the point of Free Software, not to liberate all code that has ever been written for someone but to liberate the code that you have to run for some reason.
I'm trying to think of how this situation works. And the only one I could come up with is that you're writing software that your clients use internally. Or I guess the software could produce something that your clients would end up selling to their clients.
The biggest "risk" is where you interpret the user of the software. At what level does the rights to view, modify and redistribute go away? Does your client have to make source available to its clients?
This is one of the holes closed by Affero version of the GPL. Where internal GPL code is used as a server and uses a documented non-GPL protocol and is not distributed.
In my career, 90%+ of the code I've written is Free Software (specifically, LGPL or AGPL licensed). That said, most of it is not publicly available - it's custom code written for clients, who could - but for the most part don't - release it publicly. That complies with rms' philosophy, but is it enough for you?
If not, then you have fewer choices - mostly a position on a company like Mozilla, Canonical, Sentry, etc. Finally, some people live on patronage (by companies and/or individuals), but this is generally hard to achieve.