On 04/07/2016 4:53 PM, Henry Jones wrote:
> This is a courtesy message to TOSEM, and as such, can be ignored by
> everyone else.
>
> I know you to be a class A troll,
Oh, he's a lot more than a mere troll; like L. Ron Hubbard he bought his
own publicity eventually and became a full-blown kOok.
> but since you're the only one who responded, and, since your response
> does not appear to be a troll, in and of itself, I will courteously
> respond accurately and completely to your question (since I am entering
> you into my killfile as soon as I send this out).
Futile, especially if you tell him you're doing it. He'll morph and/or
spawn a fresh sockpuppet.
> You, TOSEM, are never going to be able to answer the question, simply
> because you do not know the answer to the question. It's as simple as
> that. I'm far more likely to find out the answer by experimentation, than
> I will ever get an ounce of useful information from you.
SPNAK!
*snicker*
> You will spend infinite time accusing me of being everyone you know, so, I
> will save you the time by admitting I'm everyone you think I am. I'm
> D*rbysh*r*, and I have sex with sheep and I'm Paul, and whatever you want
> me to be.
*snicker*
> You don't have the skills required to answer the question.
> The only skill you have is the ability to type and concoct meaningless
> conspiracy theories, which you are welcome to fester in your polluted mind.
*snicker*
> *To the newsserver admins*:
> Q: How do the date headers get into the header line.
> Specifically, how does the "±0000 (UTC)" get there if the news agent
> doesn't add it? (or does it?)
Date: is a required header field, per RFC1036, which also recommends GMT
(but should probably be UTC these days) for that field.
RFC5537 clearly implies (in §3.4) that a proto-article (i.e., what you
compose in a newsreader and click "send" on) may lack the Date: field,
but states that this is a mandatory field in an injected article, which
means the server must add it if it is missing. The server must also add
the Injection-Date: header. Typically if the server adds both they'll be
identical, though I don't think that's guaranteed.
If you want to know whether your newsreader is setting the Date: field
or the server is adding it, the surest way to know is by empirical
testing. Use a server that doesn't mandate encryption (i.e., not Mixmin;
aioe should do nicely) and turn off encryption, then send an article
however you normally do while Wireshark is set up and running and
configured to capture NNTP traffic. Then you can inspect the actual
network packets your newsreader sent, and see whether or not it
generated a Date: header of left it up to the server to do so.
My guess, but I emphasize that it's a guess and only examining your
network traffic (or the source code, if available, for your newsreader)
can answer the question definitively, is that your particular newsreader
is not generating a date header, and that the time zone setting in it is
solely for deciding how to display dates *to* you in *downloaded* articles.
Per RFC1036 §2.1.2 "It is recommended that times in message headers be
transmitted in GMT and displayed in the local time zone," i.e. clients
may use a local time zone setting to display article dates in the local
time zone regardless of the original header value's time zone, for
instance by displaying 16:00 EDT as 13:00 PDT for a user on the West
Coast. So the existence of a time zone setting in your newsreader need
not mean that this determines the TZ of the Date: on an article you
*post* with it, or even whether the proto-article *has* a Date:.
I hope this answers your question as best as possible. Further answers
will, again, require examining your newsreader's source code and/or the
network traffic it generates, which means it's unlikely a server admin
here will be able to tell you anything new, except in the unlikely event
that one of them volunteers to watch the *inbound* traffic while you
post an article to tell you whether your client generated a
proto-article with or without a Date: header.
As for -0000 vs. +0000, the documentation I've read says to use -0000 to
indicate an unknown time zone, and +0000 to indicate a known offset of 0
from UTC. That means if an article's date is given in UTC the offset
should be +0000, not -0000, whether or not that's the author's local
time zone. Most likely -0000 is used by servers if a client sends a
Date: without specifying either an offset from UTC or a recognized time
zone name such as GYT.
HTH.
--
"To explain the unknown by the known is a logical procedure; to explain
the known by the unknown is a form of theological lunacy." ~David Brooks
"I get fooled all the time by the constant hosiery parade
in here." ~Checkmate