This is the mail archive of the mailing list for the glibc project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: Consensus on MT-, AS- and AC-Safety docs.

On Tue, 2013-11-26 at 17:32 -0200, Alexandre Oliva wrote:
> On Nov 25, 2013, Torvald Riegel <> wrote:
> > On Mon, 2013-11-25 at 17:40 -0200, Alexandre Oliva wrote:
> >> On Nov 25, 2013, Torvald Riegel <> wrote:
> >> 
> >> > On Sat, 2013-11-23 at 11:02 -0200, Alexandre Oliva wrote:
> >> >> On Nov 22, 2013, Torvald Riegel <> wrote:
> >> >> 
> >> >> > Thus, it will be in the context of full English sentences
> >> >> 
> >> >> Actually, no; the context of the keywords is supposed to be a one-line
> >> >> âtableâ already containing such abbreviated English words as MT-Safe and
> >> >> AC-Unsafe.  It's not running text in any way you look at it.
> >> 
> >> > I was referring to the documentation that even the one-line "table"
> >> > would be part of.  Which is mostly full English sentences, right?
> >> 
> >> I don't think so.  At least the following line, like every line that
> >> would precede the table, doesn't seem like English, let alone English
> >> sentences, to me ;-)
> >> 
> >> void *realloc (void *addr, size_t size)
> > But we document what realloc does in full English sentences, don't we?
> Yeah, but why should a table without running sentences that's
> conceptually and physically closer to the non-English prototype

If you want the properties to be code or a formal notation, you better
get a proper, formal definition first.  Right now, they are informal
annotations, so keywords consisting of non-made-up words seem to be the
right thing for me.

> be
> forced into standards that don't even apply to the running text, where
> non-English words and contractions often appear in the middle of
> sentences?

I'm not arguing that you should stick to any rules, nor that you are not
paying attention to tradition, or something like that.  I'm saying that
I think that cryptic keywords aren't the right thing to do in this
particular case, and why.  Even if we have other cases of cryptic
keywords or such, this doesn't mean that's the golden standard to

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]