Counterexample: the 4-clause BSD license (with the advertising clause) is Free Software, but not Open Source.

https://www.gnu.org/licenses/license-list.html#OriginalBSD

https://web.archive.org/web/20160116183019/https://opensourc...

I guess I stand corrected.

I never heard of the “Open Watcom” license before. But if it really does not allow modified copies to run locally then it is violating criteria 1, and 3 of the Open Source Definition. I am surprised the OSI approved it.

Here's the licence.

* http://perforce.openwatcom.org:4000/@md=d&c=GxS@//depot/open...

As you can see, it says the opposite. Per §2 you must publish your modified source if you deploy it per §1 in ways including "any and all internal use or distribution of Covered Code within Your business or organization except for R&D use and/or Personal Use".

Richard Stallman dislikes the idea of being forced to publish source code when the modified application is being published internally within a business. Ironically, and rather confusingly, being required to publish source in such additional circumstances he deems "too restrictive", right after calling open source licensing "looser" than free software. (Pick one direction for the metaphor and stick with it, M. Stallman. (-:)

This oddity comes about because freedom is also the freedom to not give one's software source code to other people. Whereas the Sybase licence makes it mandatory to publish modified source even when the compiled modified software is not being given to the public at large.

It's not hard, furthermore, to see why terms that require giving people the source code in more circumstances remain compatible with the principles of open source.

Thank you for taking the time to explain that.

I guess it really comes down to Criteria 1 of the open source definition being worded as "...shall not restrict any party from selling or giving away..."

But this is surely an oddity. It's almost like someone went out of their way to come up with a license to prove they're different.