|
|
|
|
|
by memefrog
1100 days ago
|
|
It is not "virtually closed source" to license something as GPLv3 and also under a proprietary license. In fact, it is something explicitly advocated for by Richard Stallman himself and the Free Software Foundation generally: https://www.fsf.org/blogs/rms/selling-exceptions He's got mixed feelings, but says "I consider selling exceptions an acceptable thing for a company to do, and I will suggest it where appropriate as a way to get programs freed." If of all people rms considers something acceptable, then whatever that thing is, it certainly can't be described as "virtually closed source". If you release something as open source software, then it is open source software. The end. That's what the term means. Also licensing it under a proprietary license doesn't change that. Free software licensing is not "try before you buy", and it is not "free usage for non-profits". Open source and "not for profit" are not synonyms de facto or de jure. You can use GPL'd software in commercial products and in fact many people do, frequently. You probably have a dozen copies of Linux in your house in various appliances that you don't even know are there. There are many other examples. |
|
If you publish a piece of software under two licenses, one open source and one commercial, why would anybody pick the commercial one?
You cannot say "corporations have to pick the commercial one". The only way to do that would be to include a term in the open source license so it cannot be used by corporations. But then it would no longer meet the open source definition. "Free for private use but payed for commercial use" is not open source!
But if the open source license is copyleft (e.g. GPL), any software that uses it would also have to be open source. So dual licensing allows corporations to pay to keep their own code a secret.
This implies that dual licensing is pointless for non-copyleft licenses like MIT, and also for developer tooling that is not directly used in the final product.