Re: Change “~” to “!”

I do not, myself, much care which symbol to use, so long as we are consistent.

However, our responsibility in acting as a committee means that I have to represent not only my own concerns, but those of the user base.

I think Norm is right that dismissing the experience of users is not in our best interests; I think it is also ignoring our duty to those users.

Particularly when the issue has prompted them to contribute to the list.

It feels like there are more arguments being presented to use `!` than there are to use `~`.

Tom

_________________
Tomos Hillman
eXpertML Ltd
+44 7793 242058
On 18 Apr 2022, 16:09 +0100, Bethan Tovey-Walsh <accounts@bethan.wales>, wrote:
> I assume we can expect most ixml users to be, to some extent, familiar with some flavour of computer programming. I would note that “!” or the keyword “not” are overwhelmingly the most common logical negation operators in major programming languages. “~” may be familiar to computer programmers who have ever needed to perform bitwise negation, or to those interested in mathematical logic. But they are equally likely to be familiar with the meaning “approximately”, or its use as an equivalence operator: “x ∼ y” has the meaning “x is equivalent to y”.
>
> It’s a mistake, I think, to point at usage in a specific subfield of mathematics in order to justify a somewhat confusing choice of operator, rather than drawing a perfectly reasonable alternative from the field of computer programming, which is far more relevant to our use case.
>
> Regards,
>
> BTW
>
> >
> > On 18 Apr 2022, at 14:02, Steven Pemberton <steven.pemberton@cwi.nl> wrote:
> >
> > Nah!
> > ~ has meant "not" for far longer than ! has.
> >
> > My anecdote about !important was intended to poke fun at people who couldn't cope with things having different meanings in different languages.
> >
> > I think !important in CSS a far better use of ! than in C etc., since at least the ! in CSS is closer to its real-life meaning. In real life ! doesn't mean 'not'.
> >
> > Steven
> >
> > > On Monday 18 April 2022 11:27:38 (+02:00), Norm Tovey-Walsh wrote:
> > >
> > > Hello,
> > >
> > > As long as I’m whittering on about syntax, I might as well run this up
> > > the proverbial flagpole.
> > >
> > > In my experience, “~” is much more commonly associated with
> > > “approximately” than “not”. I understand that it is used for negation in
> > > some mathematical notations, but “in some mathematical notations” covers
> > > a lot of ground. To be honest, I’d expect “￢” rather than “~” in math.
> > > (I suppose “~” may be used in some notations to mean “complement” which
> > > is precisely what we mean in ixml, but even we don’t try to use the
> > > technical term “complement”.)
> > >
> > > Furthermore, I think “!” is *very* commonly associated with “not”. It’s
> > > used that way in a whole range of languages with a C-derived syntax.
> > > Steven even has an anecdote about how the use of “!” to mean important
> > > in CSS is confusing *because* so many people think of it as negation.
> > >
> > > I propose that we use “!” to negate a character set rather than “~”. In
> > > addition to being (IMHO) more obvious to users, it would free up “~” for
> > > some future use. Given our determination to use US ASCII, we need all
> > > the special characters we can get!
> > >
> > > Be seeing you,
> > > norm
> > >
> > > --
> > > Norm Tovey-Walsh
> > > Saxonica
> > >
> > --
> >
>
>