Wiktionary talk:Votes/pl-2013-09/CFI and trimming the Idiomaticity section

Rationale
Action 1: I don't believe we treat open compounds (e.g. "car accident") as generally idiomatic regardless of whether they are semantic sum of parts. On the other hand, if "compound" is meant to mean closed compound (e.g. "headache"), then any analysis of the ease of getting the meaning of the whole from the meanings of the parts is irrelevant, as a closed compound is a single word, which at least in English is undisputed.

Action 2: "megastar" is not to be analyzed for idiomaticity, no more than "headache". "headache" is idiomatic because it is a single word, a closed compound.

In general, let us remove from WT:CFI things that contradict common practice, are not supported by consensus, are misleading, or have low or unclear added value. --Dan Polansky (talk) 09:09, 8 September 2013 (UTC)
 * I would say that these sections are relevant, but only to English. We certainly do not want this kind of logic to be applied to languages like Chinese or Sanskrit, which use different rules for writing words together. As such, I am in favor of moving these sections to About English. -- Liliana • 09:11, 8 September 2013 (UTC)
 * There is no actual difference in speech between open and closed compounds, because you can't pronounce a space. So the rationale for this difference in treatment is really only the "anything without punctuation is always idiomatic" rule. Is that even in CFI? 11:34, 8 September 2013 (UTC)


 * I agree with both removals.  Re: first removal: Even if we feel that we should keep all open compounds — and I don't think there's consensus for that — the section claims that there is exactly one reason for considering compounds idiomatic, and I think that most editors disagree with that claim.   Re: second removal:  is a prefix, or at most a combining form, so  isn't really a "compound", and really isn't relevant (especially since we have WT:COALMINE). —Ruakh TALK 23:31, 8 September 2013 (UTC)
 * My mistake in talking about closed compounds before talking about prefixed words in relation to action 2. Indeed, the discussion of "megastar" suggests that the likes of "postprocess", "postwar", "kilojoule", "millimeter", and "microsecond" should be checked for idiomaticity in terms of their parts. --Dan Polansky (talk) 18:16, 9 September 2013 (UTC)
 * The mistake is not yours, but WT:CFI's. —Ruakh TALK 22:17, 9 September 2013 (UTC)

Action 3: I realized that the paragraph now proposed for removal under action 3 (starting with "This rule must be applied carefully ..." is a follow-up on the paragraphs in action 1 and action 2. So this may be removed as well. It seems that "This rule" in "This rule must be applied carefully ..." refers to this: "The reason is that the parts often have several possible senses, but the compound is often restricted to only some combinations of them." The paragraph under action 3 discusses how to deal with "bank parking lot" given that "bank" has several senses that can be combined with "parking lot". --Dan Polansky (talk) 18:36, 10 September 2013 (UTC)

Introduction
The discussed parts and one other part of CFI#Idiomaticity were introduces in the following edits: --Dan Polansky (talk) 14:03, 15 September 2013 (UTC)
 * , 17 January 2007, +"Compounds are generally idiomatic ..."; +"For example, mega- can denote either a million ...", user:Dmh
 * , 23 January 2007, +"This criterion is also referred to as the fried egg test, ...", user:Dmh
 * , 24 January 2007, +"This rule must be applied carefully and is somewhat subjective. For example bank has several senses and parking lot has an idiomatic sense of "large traffic jam".", user:Dmh