Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.python > #107182 > unrolled thread

Moderation and slight change of (de facto) policy

Back to article view | Back to comp.lang.python

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Moderation and slight change of (de facto) policy Tim Golden <mail@timgolden.me.uk> - 2016-04-17 17:57 +0100
    Re: Moderation and slight change of (de facto) policy Wildman <best_lay@yahoo.com> - 2016-04-17 12:21 -0500
      Re: Moderation and slight change of (de facto) policy Tim Golden <mail@timgolden.me.uk> - 2016-04-17 20:50 +0100
        Re: Moderation and slight change of (de facto) policy Jon Ribbens <jon+usenet@unequivocal.co.uk> - 2016-04-19 09:03 +0000
          Re: Moderation and slight change of (de facto) policy Tim Golden <mail@timgolden.me.uk> - 2016-04-19 10:16 +0100
            Re: Moderation and slight change of (de facto) policy Jon Ribbens <jon+usenet@unequivocal.co.uk> - 2016-04-19 10:43 +0000
              Re: Moderation and slight change of (de facto) policy Steven D'Aprano <steve@pearwood.info> - 2016-04-20 02:05 +1000
                Re: Moderation and slight change of (de facto) policy Random832 <random832@fastmail.com> - 2016-04-19 13:30 -0400
                Re: Moderation and slight change of (de facto) policy Jon Ribbens <jon+usenet@unequivocal.co.uk> - 2016-04-19 17:55 +0000
                  Re: Moderation and slight change of (de facto) policy Random832 <random832@fastmail.com> - 2016-04-19 14:58 -0400
                  Re: Moderation and slight change of (de facto) policy Ethan Furman <ethan@stoneleaf.us> - 2016-04-19 12:01 -0700
                  Re: Moderation and slight change of (de facto) policy Random832 <random832@fastmail.com> - 2016-04-19 15:05 -0400
                  Re: Moderation and slight change of (de facto) policy Ethan Furman <ethan@stoneleaf.us> - 2016-04-19 12:21 -0700
                  Re: Moderation and slight change of (de facto) policy Grant Edwards <grant.b.edwards@gmail.com> - 2016-04-19 23:21 +0000
                Re: Moderation and slight change of (de facto) policy Tim Golden <mail@timgolden.me.uk> - 2016-04-19 20:17 +0100

#107182 — Moderation and slight change of (de facto) policy

There's been a bit of chatter lately about the moderation on the Python 
List (and, indirectly, comp.lang.python). The list moderators have 
suspended a couple of posters for a while and we've been discussing a 
little our policy towards non-subscribed posts.

First, a quick summary of the current settings of the list:

* Any post from someone who's not subscribed to the list is held for 
moderation

* Subscribers are also held for moderation until a moderator clears that 
flag (which we usually do on the first post, unless there's some doubt). 
This is basically to prevent canny spammers from signing up and then 
posting garbage.

* There are a few other rules which will cause posts to be held for 
moderation: unduly large posts, certain odd headers, large-scale 
cross-posting, etc.

* All attachments are stripped

Exactly how held posts are handled is down to each moderator: generally, 
though, it's quite obvious as most spam is very blatant. Occasionally, 
of course, we have a post which borders on (or is clearly) 
objectionable, and we have to decide whether to reject it at source or 
to let it through and let the community deal.

Our approach to non-subscribed posts has been to let them through if 
they're clearly genuine, ie non-spam. However, as has been pointed out 
recently, this can quite easily result in useful advice falling on deaf 
ears. The OP isn't subscribed to the list, may not be reading it via 
gmane/ggroups etc. and may simply expect people to cc them directly. The 
list subscribers have no way of knowing whether someone's subscribed or 
not, so they reply to the List. And, of course, some people object to 
cc-ing individuals as well as the List.

Our new approach (from as soon as we set it up) will be to reject 
unsubscribed posts with a friendly message indicating how to subscribe. 
The only exception we expect to make is if we spot a regular subscriber 
who's come in on a different address for some reason (eg posting from a 
phone).

The main effect, we hope, will be that people asking questions actually 
see the answers. Of course, as we've seen in the past, some people will 
be confounded by the way in which a mailing list works (ie that they see 
all the chatter going through not just the answers to their question). 
But I don't see there's very much we can do about that except to help 
them to understand how it works.

In general, please feel free to feed back to the list owners. Like 
everyone around here, we're all volunteers so we can't guarantee to 
respond in any particular timeframe, but we'll try.

TJG

[toc] | [next] | [standalone]

#107187

On Sun, 17 Apr 2016 17:57:51 +0100, Tim Golden wrote:

> There's been a bit of chatter lately about the moderation on the Python 
> List (and, indirectly, comp.lang.python). The list moderators have 
> suspended a couple of posters for a while and we've been discussing a 
> little our policy towards non-subscribed posts.
> 
> First, a quick summary of the current settings of the list:
> 
> * Any post from someone who's not subscribed to the list is held for 
> moderation
> 
> * Subscribers are also held for moderation until a moderator clears that 
> flag (which we usually do on the first post, unless there's some doubt). 
> This is basically to prevent canny spammers from signing up and then 
> posting garbage.
> 
> * There are a few other rules which will cause posts to be held for 
> moderation: unduly large posts, certain odd headers, large-scale 
> cross-posting, etc.
> 
> * All attachments are stripped
> 
> Exactly how held posts are handled is down to each moderator: generally, 
> though, it's quite obvious as most spam is very blatant. Occasionally, 
> of course, we have a post which borders on (or is clearly) 
> objectionable, and we have to decide whether to reject it at source or 
> to let it through and let the community deal.
> 
> Our approach to non-subscribed posts has been to let them through if 
> they're clearly genuine, ie non-spam. However, as has been pointed out 
> recently, this can quite easily result in useful advice falling on deaf 
> ears. The OP isn't subscribed to the list, may not be reading it via 
> gmane/ggroups etc. and may simply expect people to cc them directly. The 
> list subscribers have no way of knowing whether someone's subscribed or 
> not, so they reply to the List. And, of course, some people object to 
> cc-ing individuals as well as the List.
> 
> Our new approach (from as soon as we set it up) will be to reject 
> unsubscribed posts with a friendly message indicating how to subscribe. 
> The only exception we expect to make is if we spot a regular subscriber 
> who's come in on a different address for some reason (eg posting from a 
> phone).
> 
> The main effect, we hope, will be that people asking questions actually 
> see the answers. Of course, as we've seen in the past, some people will 
> be confounded by the way in which a mailing list works (ie that they see 
> all the chatter going through not just the answers to their question). 
> But I don't see there's very much we can do about that except to help 
> them to understand how it works.
> 
> In general, please feel free to feed back to the list owners. Like 
> everyone around here, we're all volunteers so we can't guarantee to 
> respond in any particular timeframe, but we'll try.
> 
> TJG

How will this change affect posts to comp.lang.python?

-- 
<Wildman> GNU/Linux user #557453

[toc] | [prev] | [next] | [standalone]

#107193

On 17/04/2016 18:21, Wildman via Python-list wrote:
> On Sun, 17 Apr 2016 17:57:51 +0100, Tim Golden wrote:

[... snip my explanation of new moderation for non-subscribers ...]

> How will this change affect posts to comp.lang.python?
>

Not at all, in the sense that the moderation doesn't apply to 
comp.lang.python, except insofar as posts which are sent to the list and 
which we reject obviously won't make it to comp.lang.python either.

TJG

[toc] | [prev] | [next] | [standalone]

#107308

On 2016-04-17, Tim Golden <mail@timgolden.me.uk> wrote:
> On 17/04/2016 18:21, Wildman via Python-list wrote:
>> On Sun, 17 Apr 2016 17:57:51 +0100, Tim Golden wrote:
>
> [... snip my explanation of new moderation for non-subscribers ...]
>
>> How will this change affect posts to comp.lang.python?
>
> Not at all, in the sense that the moderation doesn't apply to 
> comp.lang.python, except insofar as posts which are sent to the list and 
> which we reject obviously won't make it to comp.lang.python either.

But from what you described, it sounds like all Usenet posts will
now be prevented from reaching the list. If so you should probably
consider rmgrouping comp.lang.python or at least abandoning the
idea of it being linked to the mailing list.

(I've no idea of course if you will even see this message!)

[toc] | [prev] | [next] | [standalone]

#107309

On 19/04/2016 10:03, Jon Ribbens wrote:
> On 2016-04-17, Tim Golden <mail@timgolden.me.uk> wrote:
>> On 17/04/2016 18:21, Wildman via Python-list wrote:
>>> On Sun, 17 Apr 2016 17:57:51 +0100, Tim Golden wrote:
>>
>> [... snip my explanation of new moderation for non-subscribers ...]
>>
>>> How will this change affect posts to comp.lang.python?
>>
>> Not at all, in the sense that the moderation doesn't apply to
>> comp.lang.python, except insofar as posts which are sent to the list and
>> which we reject obviously won't make it to comp.lang.python either.
>
> But from what you described, it sounds like all Usenet posts will
> now be prevented from reaching the list. If so you should probably
> consider rmgrouping comp.lang.python or at least abandoning the
> idea of it being linked to the mailing list.
>
> (I've no idea of course if you will even see this message!)
>

Well something in my explanation obviously wasn't clear, because that's 
certainly not the case. I'm not even sure which bit gave you that 
impression (please feel free to quote it back at me!).

The Usenet gateway operates distinctly from the Mailman moderation -- in 
fact we have to jump through a hoop or two if we *do* want to moderate 
someone who usually posts via Usenet.

TJG

[toc] | [prev] | [next] | [standalone]

#107310

On 2016-04-19, Tim Golden <mail@timgolden.me.uk> wrote:
> On 19/04/2016 10:03, Jon Ribbens wrote:
>> But from what you described, it sounds like all Usenet posts will
>> now be prevented from reaching the list. If so you should probably
>> consider rmgrouping comp.lang.python or at least abandoning the
>> idea of it being linked to the mailing list.
>>
>> (I've no idea of course if you will even see this message!)
>
> Well something in my explanation obviously wasn't clear, because that's 
> certainly not the case. I'm not even sure which bit gave you that 
> impression (please feel free to quote it back at me!).

You said "Our new approach ... will be to reject unsubscribed posts".
Usenet posters are not subscribed to the mailing list, hence their
posts will be be rejected from the list, according to what you said.

> The Usenet gateway operates distinctly from the Mailman moderation -- in 
> fact we have to jump through a hoop or two if we *do* want to moderate 
> someone who usually posts via Usenet.

Are you saying that Usenet posts skip the moderation entirely?

This whole thing seems a bit of a mess. What you really need to be
doing is changing the Usenet group to be moderated, otherwise you're
going to end up with two different views supposedly of the same thing
which in fact may be completely different.

[toc] | [prev] | [next] | [standalone]

#107342

On Tue, 19 Apr 2016 08:43 pm, Jon Ribbens wrote:

> Are you saying that Usenet posts skip the moderation entirely?
> 
> This whole thing seems a bit of a mess. What you really need to be
> doing is changing the Usenet group to be moderated, otherwise you're
> going to end up with two different views supposedly of the same thing
> which in fact may be completely different.

It's worse than that. There are many other places that mirror this group,
such as gmane, Activestate, bytes.com, gossamer-threads.com, and of course
Google Groups. They all have their own moderation and filtering policies,
which may differ from those of the others. Some of them may do only
mirroring, while others (including GG and gmane) may allow posting.


And yet, we manage to muddle on. If the plethora of mirrors with
inconsistent views is the worst that happens to us, I for one will be
amazingly happy.


-- 
Steven

[toc] | [prev] | [next] | [standalone]

#107350

On Tue, Apr 19, 2016, at 12:05, Steven D'Aprano wrote:
> It's worse than that. There are many other places that mirror this group,
> such as gmane, Activestate, bytes.com, gossamer-threads.com, and of
> course
> Google Groups.

Google groups goes through Usenet, and would be affected equally by any
solution to the usenet issue.

Gmane passes posts through normal email submission. It'd be mildly
irritating to have to be subscribed to be able to post through gmane,
but it wouldn't be the first group to have that requirement.

Gossamer threads and Activestate both appear to use the mailing list
archives and not allow posting.

Bytes.com appears to mirror *stack overflow*, not this list, unless I'm
missing something.

[toc] | [prev] | [next] | [standalone]

#107352

On 2016-04-19, Steven D'Aprano <steve@pearwood.info> wrote:
> On Tue, 19 Apr 2016 08:43 pm, Jon Ribbens wrote:
>> Are you saying that Usenet posts skip the moderation entirely?
>> 
>> This whole thing seems a bit of a mess. What you really need to be
>> doing is changing the Usenet group to be moderated, otherwise you're
>> going to end up with two different views supposedly of the same thing
>> which in fact may be completely different.
>
> It's worse than that. There are many other places that mirror this group,
> such as gmane, Activestate, bytes.com, gossamer-threads.com, and of course
> Google Groups. They all have their own moderation and filtering policies,
> which may differ from those of the others. Some of them may do only
> mirroring, while others (including GG and gmane) may allow posting.
>
> And yet, we manage to muddle on.

We've muddled on so far, but apparently we're just about to have
a significant change in moderation policy which sounds like it may
very well add to the confusion.

> If the plethora of mirrors with inconsistent views is the worst that
> happens to us, I for one will be amazingly happy.

Well indeed, the main reason I posted at all was because it sounded
like the new policy hadn't thought about Usenet at all and was about
to block Usenet posts completely. If that's not true then I'm more or
less happy ;-)

[toc] | [prev] | [next] | [standalone]

#107358

On Tue, Apr 19, 2016, at 13:55, Jon Ribbens wrote:
> On 2016-04-19, Steven D'Aprano <steve@pearwood.info> wrote:
> > It's worse than that. There are many other places that mirror this group,
> > such as gmane, Activestate, bytes.com, gossamer-threads.com, and of course
> > Google Groups. They all have their own moderation and filtering policies,
> > which may differ from those of the others. Some of them may do only
> > mirroring, while others (including GG and gmane) may allow posting.
> >
> > And yet, we manage to muddle on.
> 
> We've muddled on so far, but apparently we're just about to have
> a significant change in moderation policy which sounds like it may
> very well add to the confusion.

It does look like some attention may need to be given to gmane (either
to allow posts to continue to be submitted, or to notify gmane that the
list should be put in "Non-public (posting through Gmane allowed for
list members)" mode.

[toc] | [prev] | [next] | [standalone]

#107359

On 04/19/2016 10:55 AM, Jon Ribbens wrote:
> On 2016-04-19, Steven D'Aprano wrote:

>> And yet, we manage to muddle on.
>
> We've muddled on so far, but apparently we're just about to have
> a significant change in moderation policy which sounds like it may
> very well add to the confusion.

The only thing changing is what happens if someone posts /directly/ to 
the mailing list (not through gmane, etc).

And the change is: Instead of passing the message through, with the risk 
that the poster will never see the responses), we are instead rejecting 
the post with a short explanation of how to either sign-up to the list, 
or use a mirror, etc, to interact with the list/newsgroup/mirror/whatever.

> If that's not true then I'm more or less happy ;-)

Excellent.

--
~Ethan~

[toc] | [prev] | [next] | [standalone]

#107360

On Tue, Apr 19, 2016, at 15:01, Ethan Furman wrote:
> The only thing changing is what happens if someone posts /directly/ to 
> the mailing list (not through gmane, etc).

Is the mailing list software able to make that distinction? I thought
that was on gmane's end.

[toc] | [prev] | [next] | [standalone]

#107363

On 04/19/2016 12:05 PM, Random832 wrote:
> On Tue, Apr 19, 2016, at 15:01, Ethan Furman wrote:
>> The only thing changing is what happens if someone posts /directly/ to
>> the mailing list (not through gmane, etc).
>
> Is the mailing list software able to make that distinction? I thought
> that was on gmane's end.

As far as I understand it, mirrors/usenet are either simply allowed 
through (with spam checks), or the service has it's own user account.

--
~Ethan~

[toc] | [prev] | [next] | [standalone]

#107376

On 2016-04-19, Random832 <random832@fastmail.com> wrote:

> It does look like some attention may need to be given to gmane
> (either to allow posts to continue to be submitted, or to notify
> gmane that the list should be put in "Non-public (posting through
> Gmane allowed for list members)" mode.

While there is such a mode, many lists that don't accepts posts from
non-memebers never inform gmane of it.  In my experience they usually
just discard non-member posts and send an email to the poster saying
that they only accept posts from membersk.

When that happens, you join the list, turn off delivery, and then go
back to gmane and post again.

--
Grant

[toc] | [prev] | [next] | [standalone]

#107362

On 19/04/2016 18:30, Random832 wrote:
>
>
> On Tue, Apr 19, 2016, at 12:05, Steven D'Aprano wrote:
>> It's worse than that. There are many other places that mirror this group,
>> such as gmane, Activestate, bytes.com, gossamer-threads.com, and of
>> course
>> Google Groups.
>
> Google groups goes through Usenet, and would be affected equally by any
> solution to the usenet issue.
>
> Gmane passes posts through normal email submission. It'd be mildly
> irritating to have to be subscribed to be able to post through gmane,
> but it wouldn't be the first group to have that requirement.
>
> Gossamer threads and Activestate both appear to use the mailing list
> archives and not allow posting.
>
> Bytes.com appears to mirror *stack overflow*, not this list, unless I'm
> missing something.

Just to be clear: GMane does some kind of header-munging to post on its 
users' behalf; GGroups comes in through usenet.

Very little is actually going to change. Really. (I'm almost sorry now I 
gave the explanation...). We're not disconnecting from Usenet, or GMane 
or anything else.

TJG

[toc] | [prev] | [standalone]

Back to top | Article view | comp.lang.python

csiph-web