ATTN DIG, Threading still faulty.

J. J. Lodder

unread,

Mar 26, 2015, 4:24:02 AM3/26/15

to talk-o...@moderators.isc.org

The threading still faulty.

What I see is that the References: header contains only the content of
the original Mesasge-ID:
while omitting all following References: items.

Jan

David Iain Greig

unread,

Mar 26, 2015, 12:14:00 PM3/26/15

to talk-o...@moderators.isc.org

J. J. Lodder

unread,

Mar 26, 2015, 3:58:59 PM3/26/15

to talk-o...@moderators.isc.org

David Iain Greig <dgr...@darwin.ediacara.org> wrote:

> On 2015-03-26, J. J. Lodder <nos...@de-ster.demon.nl> wrote:
> > The threading still faulty.
> >
> > What I see is that the References: header contains only the content of
> > the original Mesasge-ID:
> > while omitting all following References: items.
> >
> > Jan
>
> Nope:
> # cd /usr/local/news/spool/articles/talk/origins
> # grep References *
> <various lines omitted>
> 993:References: <md4kuc$arh$1...@dont-email.me> <mda2qk$j9b$1...@dont-email.me> <d71

> 996:References: <md4kuc$arh$1...@dont-email.me> <mda2qk$j9b$1...@dont-email.me> <838

763d9-6606-4e13-...@googlegroups.com>
> 997:References: <b486869b-5c90-4435...@googlegroups.com> <5b2d5
cf9-d538-4dc6-b...@googlegroups.com> <me2nta$9ve$1...@dont-email.me>

> 998:References: <89pmki....@news.alt.net>
>
> The server seems to be preserving References: headers?

Sometimes yes, sometimes no.

Jan

Example: (just one, there are more)
================
From: Dale <da...@dalekelly.org>
Newsgroups: talk.origins
Subject: Re: Casimir effect not valid
Date: Thu, 26 Mar 2015 04:34:00 +0000 (UTC)
Message-ID: <8asnoq....@news.alt.net>
References: <8apekq....@news.alt.net>
==================
is a reply (content is quoted) to
===================
From: nos...@de-ster.demon.nl (J. J. Lodder)
Newsgroups: talk.origins
Subject: Re: Casimir effect not valid
Date: Wed, 25 Mar 2015 09:13:18 +0100
Message-ID: <1m1u0b0.16h56o7w9rd5zN%nos...@de-ster.demon.nl>
References: <8apekq....@news.alt.net>
===========================================

Note the missing Reference: to
Message-ID: <1m1u0b0.16h56o7w9rd5zN%nos...@de-ster.demon.nl>
in the reply.

Hence it appears in the thread display
as if Dale was replying to himself.
Which he wasn't,

Jan

jillery

unread,

Mar 27, 2015, 7:03:56 PM3/27/15

to talk-o...@moderators.isc.org

All messages coming from GG are limited to at most two References.
No messages comeing from other news servers are so limited.
I don't know if this is a GG problem or a Darwin problem.

--
Intelligence is never insulting.

Dale

unread,

Mar 27, 2015, 10:54:27 PM3/27/15

to talk-o...@moderators.isc.org

On 2015-03-27, jillery <69jp...@gmail.com> wrote:
> On Thu, 26 Mar 2015 20:55:23 +0100, nos...@de-ster.demon.nl (J. J.
> Lodder) wrote:
>
>>David Iain Greig <dgr...@darwin.ediacara.org> wrote:
>>
>>> On 2015-03-26, J. J. Lodder <nos...@de-ster.demon.nl> wrote:

I'm not using google groups?

> --
> Intelligence is never insulting.
>

--
(my whereabouts below)
http://www.dalekelly.org

jillery

unread,

Mar 27, 2015, 11:33:56 PM3/27/15

to talk-o...@moderators.isc.org

On Sat, 28 Mar 2015 02:52:32 +0000 (UTC), Dale <da...@dalekelly.org>
wrote:

>I'm not using google groups?

Are you asking or saying? Don't you know how you post?

jillery

unread,

Mar 28, 2015, 5:48:55 AM3/28/15

to talk-o...@moderators.isc.org

On Fri, 27 Mar 2015 23:29:10 -0400, jillery <69jp...@gmail.com>
wrote:

>On Sat, 28 Mar 2015 02:52:32 +0000 (UTC), Dale <da...@dalekelly.org> wrote:
>

>>I'm not using google groups?
>>
>>

>>> --
>>> Intelligence is never insulting.
>>>
>
>>I'm not using google groups?
>
>
>Are you asking or saying? Don't you know how you post?

I understand your question now, and you're correct to have asked it.
Yes, in the example J.J. gave, your post's header is missing a
reference, but you do *not* use GG, which makes my hypothesis
inaccurate at best. GG is *not* the only source of the problem. Thank
your for pointing out my error.

Öö Tiib

unread,

Mar 28, 2015, 8:43:56 AM3/28/15

to talk-o...@moderators.isc.org

AFAIK Outlook Express (was part of Windows XP, handy little mail and
news client) had ability to cause similar problems. It had differences
in headers and so the servers tended to thread the messages
wrongly. Also something about it was "viral" so responding to Outlook
Express message with some other client sometimes caused the response
to have oddities in headers too so these did thread wrongly too.

David Canzi

unread,

Mar 28, 2015, 2:03:54 PM3/28/15

to talk-o...@moderators.isc.org

David Iain Greig <dgr...@darwin.ediacara.org> wrote:
>On 2015-03-26, J. J. Lodder <nos...@de-ster.demon.nl> wrote:

>Nope:
># cd /usr/local/news/spool/articles/talk/origins
># grep References *
><various lines omitted>
>993:References: <md4kuc$arh$1...@dont-email.me>
><mda2qk$j9b$1...@dont-email.me>

[snip]

><me2nta$9ve$1...@dont-email.me>
>998:References: <89pmki....@news.alt.net>
>
>The server seems to be preserving References: headers?
>

>--D.

I'm testing a hunch... If I'm right this article will be
threaded incorrectly.

--
David Canzi | "Sir, saying it louder and relentlessly is not going
| to make it true." - Sam Harris, Nightline Face-Off,
| Does God Have a Future?

Chris Thompson

unread,

Mar 28, 2015, 2:18:55 PM3/28/15

to talk-o...@moderators.isc.org

On 3/28/2015 2:00 PM, David Canzi wrote:
> David Iain Greig <dgr...@darwin.ediacara.org> wrote:
>> On 2015-03-26, J. J. Lodder <nos...@de-ster.demon.nl> wrote:
>>> The threading still faulty.
>>>
>>> What I see is that the References: header contains only the content of
>>> the original Mesasge-ID:
>>> while omitting all following References: items.
>>>
>>> Jan
>>
>> Nope:
>> # cd /usr/local/news/spool/articles/talk/origins
>> # grep References *
>> <various lines omitted>
>> 993:References: <md4kuc$arh$1...@dont-email.me>
>> <mda2qk$j9b$1...@dont-email.me>
> [snip]
>> <me2nta$9ve$1...@dont-email.me>
>> 998:References: <89pmki....@news.alt.net>
>>
>> The server seems to be preserving References: headers?
>>
>> --D.
>
> I'm testing a hunch... If I'm right this article will be
> threaded incorrectly.
>

I believe it's showing up incorrectly using Mozilla Thunderbird.

Chris

David Canzi

unread,

Mar 28, 2015, 2:33:54 PM3/28/15

to talk-o...@moderators.isc.org

David Canzi <dmc...@uwaterloo.ca> wrote:
>David Iain Greig <dgr...@darwin.ediacara.org> wrote:
>>On 2015-03-26, J. J. Lodder <nos...@de-ster.demon.nl> wrote:
>>> The threading still faulty.
>>>
>>> What I see is that the References: header contains only the content of
>>> the original Mesasge-ID:
>>> while omitting all following References: items.
>>>
>>> Jan
>>
>>Nope:
>># cd /usr/local/news/spool/articles/talk/origins
>># grep References *
>><various lines omitted>
>>993:References: <md4kuc$arh$1...@dont-email.me>
>><mda2qk$j9b$1...@dont-email.me>
>[snip]
>><me2nta$9ve$1...@dont-email.me>
>>998:References: <89pmki....@news.alt.net>
>>
>>The server seems to be preserving References: headers?
>>
>>--D.
>
>I'm testing a hunch... If I'm right this article will be
>threaded incorrectly.

My hunch is confirmed. trn generated a single-line References
header with two message ids on it. I split it into two lines,
like this:

References: <1m1vwbh.hqhyqg1evbxfzN%nos...@de-ster.demon.nl>
<dgreig-slrnmh8...@darwin.ediacara.org>

In message headers, a line beginning with a blank or tab is a
continuation of the previous line. Your moderation software
is assuming that the References header is all on one line, so
only the first line got into the posted article, and my
article ended up threaded incorrectly.

Google Groups is splitting long references headers into a first
line and one or more continuation lines. I was able to find out
that this was happening by examining the headers of articles in
soc.culture.jewish.moderated.

Dale

unread,

Mar 28, 2015, 6:53:53 PM3/28/15

to talk-o...@moderators.isc.org

On 2015-03-27, jillery <69jp...@gmail.com> wrote:
> On Thu, 26 Mar 2015 20:55:23 +0100, nos...@de-ster.demon.nl (J. J.

> Lodder) wrote:
>
>>David Iain Greig <dgr...@darwin.ediacara.org> wrote:
>>
>>> On 2015-03-26, J. J. Lodder <nos...@de-ster.demon.nl> wrote:

>>> Nope:
>>> # cd /usr/local/news/spool/articles/talk/origins
>>> # grep References *
>>> <various lines omitted>

>>cf9-d538-4dc6-b...@googlegroups.com> <me2nta$9ve$1...@dont-email.me>

>>> 998:References: <89pmki....@news.alt.net>
>>>
>>> The server seems to be preserving References: headers?
>>

> --
> Intelligence is never insulting.
>

I'm not using Google Groups ...

jillery

unread,

Mar 28, 2015, 9:48:54 PM3/28/15

to talk-o...@moderators.isc.org

On Sat, 28 Mar 2015 18:33:31 +0000 (UTC), "David Canzi"

Congratulations, but I do have a question. In looking at headers from
users who post with GG, I noticed the following:

The last confirmable "good" GG message I can find is:

<84033235-218f-4d95...@googlegroups.com>

dated On Fri, 6 Mar 2015 04:32:50 -0800 (PST)

The first confirmable "bad" GG message I can find is:

<c7b5d513-0dbd-49f6...@googlegroups.com>

dated On Fri, 6 Mar 2015 11:56:43 -0800 (PST)

DIG posted his update messasge On Fri, 6 Mar 2015 16:58:17 +0000
(UTC), which is inbetween the two GG messages.

To the best of my knowledge, no confirmable "good" GG messages exist
after DIG's post, and no confirmable "bad" GG messages exist before
DIG's post.

When you say "Google Groups is splitting long references", you seem to
imply the problem is with GG. However, it would be remarkable if GG
started doing anything different at the same time DIG installed new
software.

Also, as Dale pointed out, the problem affects his posts but he
doesn't use GG.

Am I missing something obvious?

David Canzi

unread,

Mar 29, 2015, 1:23:55 AM3/29/15

to talk-o...@moderators.isc.org

jillery <69jp...@gmail.com> wrote:
>On Sat, 28 Mar 2015 18:33:31 +0000 (UTC), "David Canzi"
><dmc...@uwaterloo.ca> wrote:
>
>>David Canzi <dmc...@uwaterloo.ca> wrote:

[snip]

It's not obvious, but I was describing the observed phenomena in
non-judgemental language. Since I have so often griped about
the latest stupid thing Google Groups was doing, it may not be
possible any more for me to mention Google Groups without it
sounding like a denunciation.

What Google Groups is doing is actually right, this time.

This problem has happened before in talk.origins. Long ago the
moderation software on darwin didn't handle multi-line References
headers correctly. It was modified to fix the problem.

I think, when darwin was rebuilt on Mar 6th, that the re-installed
moderation software was an older version without the modification.

J. J. Lodder

unread,

Mar 29, 2015, 9:13:52 AM3/29/15

to talk-o...@moderators.isc.org

jillery <69jp...@gmail.com> wrote:

> On Thu, 26 Mar 2015 20:55:23 +0100, nos...@de-ster.demon.nl (J. J.
> Lodder) wrote:
>
> >David Iain Greig <dgr...@darwin.ediacara.org> wrote:
> >
> >> On 2015-03-26, J. J. Lodder <nos...@de-ster.demon.nl> wrote:
> >> > The threading still faulty.
> >> >
> >> > What I see is that the References: header contains only the content of
> >> > the original Mesasge-ID:
> >> > while omitting all following References: items.
> >> >
> >> > Jan
> >>
> >> Nope:
> >> # cd /usr/local/news/spool/articles/talk/origins
> >> # grep References *
> >> <various lines omitted>
> >> 993:References: <md4kuc$arh$1...@dont-email.me> <mda2qk$j9b$1...@dont-email.me> <
d71

> >ee378-a613-4ed8-...@googlegroups.com> <mdqn1q$71e$1@dont-email.

> >5a5cf-de8b-4e73-...@googlegroups.com> <me2gm0$khi$1@dont-email.
w

Don't think it was like that in the past. So Darwin.
If true, it is a good reason to ban google altogether.

Jan

jillery

unread,

Mar 29, 2015, 4:40:07 PM3/29/15

to talk-o...@moderators.isc.org

On Sun, 29 Mar 2015 05:19:01 +0000 (UTC), "David Canzi"

That's a probable inference. Anyway, congrats again. I hope DIG
finds your analysis helpful.

RSNorman

unread,

Mar 29, 2015, 4:58:50 PM3/29/15

to talk-o...@moderators.isc.org

On Sun, 29 Mar 2015 16:37:36 -0400, jillery <69jp...@gmail.com>
wrote:

My recollection is that DIG greatly prefers to be reached by email
rather than by posting comments here. Unfortunately I have no idea
what his email might be. Certainly others here know and can provide
it.

J. J. Lodder

unread,

Mar 30, 2015, 3:43:50 AM3/30/15

to talk-o...@moderators.isc.org

The FAQ says: talk-origins-request at ediacara dot org

No idea if that still works, for ediacara.org is down,

Jan

Öö Tiib

unread,

Mar 30, 2015, 10:18:48 AM3/30/15

to talk-o...@moderators.isc.org

On Sunday, 29 March 2015 16:13:52 UTC+3, J. J. Lodder wrote:

> jillery <69jp...@gmail.com> wrote:
>
> > All messages coming from GG are limited to at most two References.
> > No messages comeing from other news servers are so limited.
> > I don't know if this is a GG problem or a Darwin problem.
>
> Don't think it was like that in the past. So Darwin.
> If true, it is a good reason to ban google altogether.

If we haven't tracked down a defect but are just unsure then it is
a FUD (fear, uncertainty & doubt). FUD is *rather* *dumb* reason
to ban anything. FUD is good reason to improve our awareness.

If we have tracked down a Google's defect then it is good reason
to send feedback / report about issue somewhere. Maybe there:
https://productforums.google.com/forum/#!categories/apps/google-groups-for-business

In my experience it is lot easier to get issues in Google's software
fixed just by reporting those compared with for example issues
in Microsoft's software.