Frank Slootweg writes:
Richard Kettlewell wrote:
Frank Slootweg writes:
Richard Kettlewell wrote:
"s|b" writes:
On Sat, 4 Jul 2015 20:53:05 -0700, Norm X wrote:
This message is X-posted to alt.comp.hardware & comp.mobile.android.

Not sure, but I don't think it made it to comp.mobile.android. There's a
space in your Newsreader header.

Looks fine to from here. (The syntax is quite liberal about
whitespace)

No the syntax is *not* quite liberal. It's only that many News servers
or/and newsreaders are 'liberal' about accepting/processing crap from
non-compliant newsreaders. But to be fair, we can't expect them to
already get it after only three-plus decades, can we!? :-(

What did you have in mind?

In the older standards/RFCs a space *between newsgropups* in the
Newsgroups: header was not allowed. So the OP's header

Newsgroups: alt.comp.hardware, comp.mobile.android

was not allowed.

850 and 1036 are rather vague but they do incorporate 822 by reference,
and 822 allows arbitrary whitespace between tokens. So whitespace
allowed in Newsgroups: is at least a plausible interpretation, though in
the absence of a proper grammar you can’t say anything for sure.

You can look at the early implementations though, and it seems B News
accepted spaces in the Newsgroups: header. So in your view of the world
I guess the rot had already set in by the mid 1980s. Personally I’m
skeptical about this idea that everyone including early server
implementors has been getting it wrong for 30 years.

The newer/current standards/RFCs are more lenient, because they allow
for FWS (Folding White Space), but as the name implies, that's intended
for *folding* (too) long lines, not for use within a short line.

Because there is always a lot of old(er) software around, the mantra
is "Be strict in what you send and lenient in what you accept.". So the
OP's newsreader SHOULD NOT have put the space in the list of newsgroups,
but other software - News servers and newsreaders alike - SHOULD accept
the 'broken' Newsgroups: header, and the latter is exactly what
happened.

Well, the statement was about what the syntax is, not about what it may
or may not have been in 1981, and 5536 is the current spec, so I’m
content that it stands.

--
http://www.greenend.org.uk/rjk/