Autocannons are not machine guns
But the point of the entry is to show what people might mean when using the word "cannon". One of those things is what is more accurately described as an autocannon.
That's exactly what I said, that "cannon" is often short for "autocannon" when discussing aircraft. So why remove it?
I'll go ahead and put it back then since it seems to have been a misunderstanding.
We're talking about two different things. I removed the "large-bore machine gun" definition and simultaneously added the autocannons on aircraft line. I figure whoever added the large-bore machine gun line was most likely thinking of autocannons.
My understanding was that you said that my usage comment about autocannons on aircraft makes the machine gun definition correct, which led to my first comment above saying that autocannons are not machine guns. There is no version of a cannon that is a machine gun, or vice versa, so the definition of "a large-bore machine gun" is completely incorrect. If you want to say that some cannons are autocannons, that would be correct, but I think the autocannons on aircraft line under usage covers it.
Ok, but then why didn't you replace the machine gun definition with one about autocannons? Usage notes aren't meant to add more definitions to a word, only to give information about how the listed definitions are used in more detail.
Because a cannon isn't a type of autocannon. An autocannon is a type of cannon. So how would I add autocannon as a definition of "cannon"? Or, do you want me to list shorthand as a definition? I mean, I could say that "cannon" is sometimes used as short for "autocannon" under definitions, but that doesn't seem right.
I'm not sure. I'm fairly certain that there are other examples where a more generic term is widely used to refer to a more specific term, to the point where using the specific term is thought of as overspecifying. "computer" is another example; most people use it specifically for a desktop personal computer. An egg is assumed to be from a chicken, and milk is assumed to come from a cow. Maybe this should be brought up at the Tea Room as there may be an existing practice concerning cases like this already.
Yes, but all of your examples are usage, not definitions.
Usage is just an example of how the word is sometimes used. A definition is what the word actually means.
Are you pulling my leg?
One definition of "car" could be "a vehicle used for transportation" while usage could be "Jane drove the car to the store."
Of course, but when you say that sentence, you're using the word "car" and intend it to mean something specific. That's its definition; it's what people mean when they say it.
In this case, when people say "cannon", they sometimes mean something that could also be described by the word "autocannon". So in that sense they are synonyms. Therefore, one of the definitions of "cannon" is "autocannon".
If you want to say that one of the definitions of "autocannon" is a "cannon" with certain particular characteristics, then I'm ok with that. I'm not ok with saying that a "cannon" is an "autocannon" because that would be a lie.
I've decided I can live with adding "sometimes used as shorthand for "autocannon", when mounted on aircraft" if it's genuinely important to you. Emphasis placed on shorthand.
But CodeCat's examples are unlike either of yours- I think you're reanalyzing things without realizing it. A definition is just a tool for presenting a meaning- it doesn't determine meaning, unless it's adopted by users of the language. As for the meanings themselves, they can't exist independently of usage. Any language is a negotiation between those who use it: everyone may have a different idea of what a term means, but only those meanings that are understood by the one producing and the one receiving will work: I may be convinced that German means extraterrestrials with three ears, but if I use it that way, no one will understand me, and if I try to assign that meaning to what others say, I won't understand them. The way this negotiation works is through usage. At one time, an earlier form of hound was a general term for any dog, but at some point, it became restricted to a particular type of dog, and the earlier form of dog became the general term. The kind of things CodeCat is talking about are hidden parts of the meaning that may, at some point, become explicit and thus part of what you would call the definition. Even now, people tend to use qualifiers for any type of milk that's not from a cow: goat's milk, mother's milk, etc. That's a sign that things are already heading in that direction.
Chuck, you wouldn't list "short for goat's milk" as a definition for the word "milk". You could, on the other hand, list "milk that comes from a goat" as a definition for "goat's milk". So, you could go to "desktop personal computer" (if it was one word) and add a definition that it's a certain type of computer, but you wouldn't list desktop personal computer as a definition of computer.