"Murray Sargent" <murr...@exchange.microsoft.com> wrote: > Alex notes "Operands are not operators, e.g. in a+b, a and b are operands, > + is an operator."
Not always true, this depends on the domain of definition, see below, and all operators can also be themselves operands of another operator. The generic term for them should be "functional objects". (with the same assumption made in computer "functional languages" that use this formalism). All these functional objects can be considered as constants (or values) or as functions modifying the nature or value of the surrounding functional objects to create a new functional object as the result of their combination. Whever they are constant or not depend on the domain of definition (which is not always exposed in all formulas, and often implied by the context. > I'm sure Karl Williamson knows that, but the mathematical alphanumerics also > aren't operators and they nevertheless have the math property. We need to > change > the description of the math property to include all characters that are used > primarily for math and the EULER CONSTANT is such a character. The important word of your sentence is "primarily", which just means "mostly for maths" but it should be "for scientific formal notations", and not for written orthographies of humane language. Maths can use any character in the UCS it wants (or more exactly any grapheme cluster that can be built with UCS characters, plus a few specific combining characters). And it will continue to create new symbols, and there are certainly many of them that have still not be discovered in some books or scientific paper, that will be encoded later). Given that also the General category is stable (and also th fact that some humane orthography may choose to borrow some symbols currently encoded as Maths symbols within its alphabet, or a convenient abbreviation signs), this general category is the wrong tool for us. So we may need a custom property (but NOT subject to the stability policy) to reference characters that are CURRENTLY considered as NOT being used in humane languages, but mostly for mathematic/scientific notations, even if these lette-like symbols were created from a script for humane languages : they are used only for their symbolic value (and do not obey to the linguistic rules such as collation mappings, case mappings...). The hbar symbol is such a character. Such a property would be useful to exclude, in an implementation of a specific version of Unicode, these characters from normal linguistic processing, in order to protect them from alteration. And it would also be useful if ever, later, some humane language starts getting written using the symbol, and starts applying linguistic features such as collation mappings and case mappings : In that case the symbol should remain stable, and the linguisitic letters should be encoded separately, unless there's evidence that too many texts are already using the maths symbol directly (in which case, this symbol will be removed from the custom "scientific-only" category defined by the custom (non stable) property. Note that in maths, there's no really any distinction between operators and operands. They are just symbols having a functional behavior and an implied associativity (on left or right, or both, depending on the notation used). It's impossible to predict the associativity and use of any symbol without knowing the context of use, and without knowing the domain of definition of these "functional objects". Philippe.
