Re: [Gnu-arch-users] ANNOUNCE: Spaces in filenames, finished

[Tom Lord]

    > From: Jan Hudec <address@hidden>

    > Actualy, there are two escaping shchemes already widely used:

    >     1) XML Entities: &#<unicode-codepoint-number>;
    >     2) Old C's scheme extended for unicode:
    >        \u<unicode-codepoint-number-in hexa>

    > Why does hackerlab/pika/tla invent another one?
    > (The argument that R5RS only requires \" is clear to me, but does not
    > look sufficient to me).

Regarding C's: how many hex digits go there?  It seems to me that the
only sane ways to escape unicode codepoints in a string all involve a
delimited hex number.  C's syntax is also no good for having
multi-character symbolic names for characters (for the same reason --
delimeter issues).

Regarding XML: it's simply a judgement call on my part about where the
greatest potential for synergies arise.  XML sure gets a lot of
exercise, but the toolsets that have arisen around it don't really
impress me as giving a lot of leverage in this situation (and, not
without reason).  The Scheme-derived syntax, which I'm working with
some sucess on turning into an Official Standard, I think will turn
out to work better with other tools in the medium to longer term.
It's just a judgement call....I can't prove it.  Yes, it's an
"underdog" bet but, that's where I laid my money down.

-t