Modifications can be licensed differently but that takes extra work. If I release a project with the MIT license at the top of each file and you download my project and make a 1-line change which you then redistribute, you need to explicitly mark that line as having a different license from the rest of the file otherwise it could be interpreted as also being MIT licensed.

You also could not legally remove the MIT license from those files and distribute with all rights reserved. My original granting of permission to modify and redistribute continues downstream.

Citation?

On the contrary: https://opensource.org/osd

What exactly does "open" mean when used as a qualifier for "source"?

The fact is that your claim "“open source” consists of two words which have meaning, but somehow doesn’t mean ==>that<== when combined into one phrase" is simply false, as there is no "that".

> Same with free software, in a way.

This is a much more supportable argument, but note the change in wording: "free software" is not the same as "free source". The latter suggests that one doesn't have to pay for the source, but says nothing about what one can do with the source or one's rights to software built from that source.

As for "free [as in freedom] software", I think there would have been less contention if RMS/FSF had called it "freed software" or "liberated software", and it would have been more consistent with their stated goals.

> Programmers really are terrible at naming things.

This is silly sophism based on one anecdote that you didn't even get right. Naming things well is hard, and names in software have conditions that don't exist in more casual circumstances. The reality is that good programmers put a lot of effort into choosing names and generally are better at it than the population at large.

Even without a specific definition for "open source", I wouldn't consider source code with a restrictive license that doesn't allow you to do much with it to be "open".

I don't think this is a case of programmers being bad at things (although I get that you said that as a joke), I think it's much worse than that: This is some kind of weird mind-over-matter "if we believe it hard enough it'll come true" thing. Sort of an "if we beat everyone who says the emperor has no clothes, we can redefine 'clothes' to include 'the emperor's birthday suit'". Note that these people who are downvoting anyone who dares to say that "open source" isn't synonymous with the OSI definition never concede an inch to the notion that the words have a common-sense meaning and the OSI didn't invent the term (provable via internet archive). Because it's not about being right it's about changing reality to match what they wish were true.

This isn't that uncommon:

* If a country doesn't have "closed borders" then many foreigners can visit if they follow certain rules around visas, purpose, and length of stay. If instead anyone can enter and live there with minimal restrictions we say it has "open borders".

* If a journal isn't "closed access" it is free to read. If you additionally have permissions to redistribute, reuse, etc then it's "open access".

* If an organization doesn't practice "closed meetings" then outsiders can attend meetings to observe. If it additionally provides advance notice, allows public attendance without permission, and records or publishes minutes, then it has “open meetings.”

* A club that doesn't have "closed membership" is open to admitting members. Anyone can join provided they meet relevant criteria (if any) then it's "open membership".

EDIT: expanded this into a post: https://www.jefftk.com/p/open-source-is-a-normal-term

Who says it isn't? "closed source" doesn't have a formal definition, but can be arbitrarily defined as the antonym of open source, and when people use the term that's usually what they mean.

And that has nothing to do with whether someone can be "blamed" for ignoring the actual meaning of a term with a formal definition.