Re: glibc's getaddrinfo() sort order

["Giacomo A. Catenazzi" <cate@debian.org>]

Ian Jackson wrote:
> Furthermore, the broken behaviour is NOT THE STANDARD.
>
> LET ME SAY THAT AGAIN: THE BROKEN BEHAVIOUR IS *NOT THE STANDARD*
> LET ME SAY THAT AGAIN: THE BROKEN BEHAVIOUR IS *NOT THE STANDARD*
> LET ME SAY THAT AGAIN: THE BROKEN BEHAVIOUR IS *NOT THE STANDARD*
> LET ME SAY THAT AGAIN: THE BROKEN BEHAVIOUR IS *NOT THE STANDARD*
> LET ME SAY THAT AGAIN: THE BROKEN BEHAVIOUR IS *NOT THE STANDARD*

Anyway it is not the point: do we have a 100% C++ conforming compiler,
do we have a 100% conforming POSIX system? [1]

So I don't think we should follow 100% the STD (and I think nobody
do because some old broken (or unused) behavior: see SMTP (privacy,
relaying, routing mail), TCP (IIRC there was some buggy wording that
cause DoS on TCP connection start), STD5-RFC950 "Internet Standard
Subnetting Procedure", ...)

So also if it was (or if it became) "standard", it doesn't mean that
we should apply literally.

ciao
	cate

[1] We are becoming more posixly, also because POSIX adapts
to Linux usage, so not following a insane standard is the
quick way to have correction to the standard and be more
standardized.