[davidfstr]

No license is going to satisfy everyone in all contexts.

For example I cannot use GPL at work (based on my own rules as team lead) but am open to it in personal projects.

And other folks would consider having something NOT GPL licensed to be suspect.

It is not possible to provide a uniform license that will be acceptable to all in all circumstances.

[mannerheim]

Even the hardest core Stallmanites aren't unwilling to use BSD/MIT-licensed software, nor Stallman himself. In contrast, these sorts of licenses that restrict what the software can be used for are unequivocally condemned by the FSF.

[jpmoral]

I don't think they meant that there should be One True Open-Source License but rather that open-source projects should pick one of the 'standard' ones (GPL, MIT, etc.) instead of writing their own.

[davidfstr]

Well, if by “open source” you mean broadly any software available whose code is viewable on the internet, you’re essentially advocating for no projects that use “nonstandard” licenses to exist.

That seems like a waste if there would be otherwise be projects around with a nonstandard license that was acceptable to a subset of users.

For example the CPython interpreter (the official interpreter for the Python language) does not use a standard license. Would you not want Python to exist?

[mannerheim]

Any permissive software license (which it seems CPython's is) is broadly interchangeable with any other, and can be used with both other permissive licenses and copyleft licenses like the GPL. That can't happen with any license that places restrictions on what it can be used for.