[alpine] multipart quoting of big5 [Was: making posts unreadable...] - Other Technologies

On Fri, 28 Sep 2007, Peter J Ross wrote:

> In news.software.readers on Wed, 26 Sep 2007 20:38:54 -0700, Eduardo
> Chappa <chappa@u.washington.edu> wrote:
>
>> On Wed, 26 Sep 2007, Peter J Ross wrote:
>>
>> If the PINE people can provide examples of the problem they're trying
>> to avoid, I'll be more sympathetic to what they do.
>>
>> I will show you one. This one came from a mailing list and I modified it
>> to obfuscate e-mail addresses, but you can probably track it back to the
>> list because there is enough information in it.

....

> Anyway, it's a mailing list message, not a Usenet message.


For me, the difference is marginal. I hate mailing lists and am reading
everything via news.gmane.org. So perhaps we could drop that argument.
After all NNTP is just as transport protocol (Network News *Transfer*
Protocol), the format of the message itself is the same as in mails. NNTP
RFCs refer to mail format RFCs for the format of the transfered
information.


>> Anyway, the message is at
>>
>> http://staff.washington.edu/chappa/pine/wronghex

Bye,
Vitus
--
Vitus Jensen, Hannover, Germany, Earth, Universe (current)

Peter J Ross wrote:
> In news.software.readers on 28 Sep 2007 20:58:24 GMT, Blinky the Shark
><no.spam@box.invalid> wrote:
>
>> ["Followup-To:" header set to news.software.readers.]
>
> Mine isn't, but I've reached the point of being bored with all these
> attempts to defend the indefensible.
>
> Dear pine-users, feel free to mix and match funny MIME-types, funny
> quote-markers and funny .sig delimiters. Just don't get upset if your
> posts in groups other than CMP are ignored or flamed.
>
> Bye-bye.

I just engaged my twit filter, as well, Peter.

--
Blinky RLU 297263
Killing all posts from Google Groups
The Usenet Improvement Project - http://improve-usenet.org

In news.software.readers on 28 Sep 2007 22:21:53 GMT, Blinky the Shark
<no.spam@box.invalid> wrote:

> Peter J Ross wrote:
>> In news.software.readers on 28 Sep 2007 20:58:24 GMT, Blinky the Shark
>><no.spam@box.invalid> wrote:
>>
>>> ["Followup-To:" header set to news.software.readers.]
>>
>> Mine isn't, but I've reached the point of being bored with all these
>> attempts to defend the indefensible.
>>
>> Dear pine-users, feel free to mix and match funny MIME-types, funny
>> quote-markers and funny .sig delimiters. Just don't get upset if your
>> posts in groups other than CMP are ignored or flamed.
>>
>> Bye-bye.
>
> I just engaged my twit filter, as well, Peter.

Xref?



--
PJR :-)

["Followup-To:" header set to news.software.readers.]
Peter J Ross wrote:
> In news.software.readers on 28 Sep 2007 22:21:53 GMT, Blinky the Shark
><no.spam@box.invalid> wrote:
>
>> Peter J Ross wrote:
>>> In news.software.readers on 28 Sep 2007 20:58:24 GMT, Blinky the Shark
>>><no.spam@box.invalid> wrote:
>>>
>>>> ["Followup-To:" header set to news.software.readers.]
>>>
>>> Mine isn't, but I've reached the point of being bored with all these
>>> attempts to defend the indefensible.
>>>
>>> Dear pine-users, feel free to mix and match funny MIME-types, funny
>>> quote-markers and funny .sig delimiters. Just don't get upset if your
>>> posts in groups other than CMP are ignored or flamed.
>>>
>>> Bye-bye.
>>
>> I just engaged my twit filter, as well, Peter.
>
> Xref?

From

--
Blinky RLU 297263
Killing all posts from Google Groups
The Usenet Improvement Project - http://improve-usenet.org

On Fri, 28 Sep 2007, Peter J Ross wrote:
> The point that you're missing is that no example has yet been provided
> for why multipart/mixed should be used *on Usenet* instead of a simple
> character set declaration. I want to see some examples of broken
> Usenet posts that Pine has fixed.

The fact that you are unaware of something does not translate into that
something not existing.

Why should I go digging through years of archives? "Because some Peter J
Ross won't do his own research" does not consitute a sufficient reason.

>> I wonder if DMR gets lectures from self-appointed C experts... ;-)
> I wonder if he's pig-headedly defensive if he does something annoying.

Why are you annoyed? If you have a proper MIME-compliant client you
wouldn't even see it.

Sheesh. Get a life.

>> I am also fascinated by your bizarre claim that the problem doesn't exist.
> How is it bizarre? Where are the examples of the problem on Usenet?

This has already been answered. Next time, take your fingers in your
ears and stop yelling "I can't hear you" when people answer.

> I'm beginning to think that Pine is yet another example of software
> that tries to do both mail and news and does at least one of the two
> imperfectly.

No, the problem is that you have falsely concluded that the way your news
reader works is the one and only true way, and that any other way is
heretical and must be suppressed.

>> As one of the designers of MIME, I said it before, and I'll say it again:
>> BASE64 and QUOTED-PRINTABLE content should ALWAYS be guarded inside a
>> multipart.
> Except when there's no reason to send QP or B64 in the first place.

Some messages transit both email and USENET. Some messages transit one,
or the other. Some messages start out transiting one, and later one
travel over the other. For this reason, it is a lost cause to keep BASE64
or QUOTED-PRINTABLE out of USENET.

Some messages are PGP or S/MIME signed. For this reason, it is a lost
cause to keep multiparts out of USENET.

Once you accept that this is a lost cause, and use a proper MIME-compliant
client, the entire point of arguing about this issue vanishes.

> NNTP is 8-bit clean, and has been for a long time. I don't think such
> contortions are needed for Usenet.

Talk is cheap. What surety do you provide to guarantee compensation those
users whose messages are subsequently damaged if you get your way?

>> I know that (almost) all USENET transport is 8-bit.
> NNTP is. I'm less sure about UUCP, if anybody's still using it.

I know of at least one NNTP implementation that is 7-bit. It is nearly
extinct, but it exists. As far as I know, it is the only NNTP
implementation that can run on that particular platform.

> I'm reading Usenet. Why use workarounds that are needed only in
> another environment?

Your environment may be restricted to USENET, but not everybody has the
same restricted environment as you.

> Why blur the distiction between news and mail by
> reducing both to the same common denominator?

That battle was lost 15 years ago. It's somewhat cute to see this zombie
pop up from the dead; like that stupid whack-a-mole game at the carnival.

> You understand wrongly. I'm trying to defend text newsgroups in
> general against an influx of posts that are difficult for existing
> software to display and might cause Pine users to be plonked by both
> news users and news providers.

Who died and appointed you net.cop?

No news providers are plonking Pine users. They are plonking USENET
instead by shutting down their NNTP service.

USENET is dying. Most newsgroups are sewers, and the few that remain have
a small handful of users. There's only a few dozen people at UW who still
read USENET out of a 6-digit user community.

> No, I want slrn to support multipart/signed, and I don't mind if
> support for multipart/mixed is added too, as long as I can switch it
> off.

If you demand that 15-years obsolete (due to the advent of MIME) slrn be
accomodated, then I demand that you accomodate a system I have at hand
which became obsolete 20 years ago and can only do 7-bit ASCII and ISO
2022.

If slrn is updated to accomodate MIME, then you should never see, nor
care, about the MIME artifacts in question.

>> But you are not the only person who has to deal with
>> legacy issues. Sometimes, one has to do unpleasant things to get data
>> through paths that they have NO control over.
> You admit that the NNTP transport is happy with 8-bit articles, yet
> you think a workaround for a 7-bit context is needed.

There is no way to predict with 100% certainty, nor control, when 8-bit
content hits a 7-bit pipe. This is the case EVEN IF the content is posted
to NNTP.

Some people prefer not to risk their content being mangled. A Pine user
has the choice.

You are demanding that the choice be taken away, and for what purpose?
So that your obsolete slrn client won't show MIME artifacts?

Sheesh. Get a life.

> Does Pine have an option to use a conventional .sig delimiter?

If, by this, you mean sigdashes, the answer is yes.

If you wish to make sigdashes mandatory, then you need to produce a
standard to do so and go through the process of getting that standard
approved. Otherwise, who died and appointed you net.cop?

-- Mark --

http://staff.washington.edu/mrc
Science does not emerge from voting, party politics, or public debate.
Si vis pacem, para bellum.

On Fri, 28 Sep 2007, Jeffrey Goldberg <nobody@goldmark.org> wrote:
> In <p9ovs4-k86.ln1@wiegner.user.individual.de>, Thomas Wiegner wrote:
>
>> With corrupted articles I'm talking about postings having undeclared
>> 8bit in their header, having iso-8859-1 or utf-8 encoded characters in
>> the body which is declared as us-ascii or without any charset
>> declaration. I've also seen postings declaring a charset which does not
>> exist.
>
> We were talking about different things.

Yes, indeed.

> I fully agree that what you
> describe are corrupted. Has pine been a source of such corrupted
> messages? If alpine is, I'm sure the developers would like to know about
> it.

The sources for such articles are mostly OE (surprise, surprise), Google
Groups and Forte Agent. I haven't seen such articles from (al)pine, but
I don't see too much (al)pine postings in the usenet anyway. Or I
oversee them, because it doesn't provide a user-agent header and I
don't display the message ID.

Cheers,

Thomas

On Fri, 28 Sep 2007 20:27:09 +0000, Blinky the Shark wrote:


> s/now/not
>
That's not funny. Stop it NOW~@

Top