Is this word referring specifically to the absolute value function of a real number (in which case it is a type of norm) or would it be used for any general case when a norm is desired (in which case, the exact type of norm may be vague but could probably usually be inferred from context or be elsewhere defined)? In either case, I think that the other definition merits a separate word explicitly addressing that meaning (since both meanings are important and specification and generalization/vagueness are both useful in different cases). I suggest that this word should be reserved for the general/vague meaning of "norm" (generic) and that we define a new word for specifically "absolute value of a real number". Doing so would be at the risk of confusion (clearly), since this definition says "absolute value" (we might need to edit that sometime), but it does make the shortest word that is immediately apparent (and available) useful in all appropriate contexts (even if it is not quite clear or as specific as could sometimes be desired); it also follows Lojban's typical tradition of being generic/general/vague and focusing/narrowing in on meanings with additions of some sort (in this case, probably syllables). Note that I have proposed cu'ai for another type of norm (actually, a set of types); I can think of at least a handful of others, including the p-adic norm and prime-distance, the complex magnitude/norm/modulus, word length norm (which is actually very general), sequence/function norms (amount of initial or total overlap, maximal pointwise distance, etc.; all of these tend to be incredibly general and useful in, for example, genetics), graph/network norms, various metrics, etc. I am not sure how to include all of these as words without flooding cmavo space. Certainly several deserve words and we might be able to consolidate a few others (although I feel slightly queasy about doing so), while some others probably are slightly to arcane/exotic/esoteric to merit (short cmavo). It all merits discussion.
|