[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Scheme-reports] inconsistent use of \| escape

Alex Shinn scripsit:

> Unfortunately, when the formal syntax differs from the prose,
> the formal syntax takes precedence.  Whatever the WG intentions
> may have been, the community ratified a standard in which \|
> is not a predefined escape sequence in strings.  Referring to old
> ballots was fine during the process, but now can only serve for
> historical interests and to dig up rationales.

Unsurprisingly, I don't agree.  The prose has always taken precedence
in my mind, and in fact I tend to forget about the formal syntax, which
is undoubtedly why I forgot to update it.

> Looking for additional discussion it looks like the most recent public
> post was during the formal comments period:
> http://lists.scheme-reports.org/pipermail/scheme-reports/2013-January/003223.html

That post and its followups obviously didn't notice the contradiction
either: they were discussing the formal syntax in its local context,
not in the context of the whole report.  (Nobody's fault, of course.)

> Attention was specifically brought to this, and a fix was made retaining
> the formal syntax, so I think it's difficult to go back on this now.

If the point had been raised that "\|" contradicted the prose, then I
agree that this post would matter; as things are, I again cannot agree
that this post is definitive or even relevant.

I think the only thing we can do in the errata is to point out the
contradiction and say that implementers will have to decide on their
own whether to support \| in strings or not.

Overhead, without any fuss, the stars were going out.
        --Arthur C. Clarke, "The Nine Billion Names of God"
                John Cowan <cowan@x>

Scheme-reports mailing list