[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

2nd draft (was: Re: Revising the Code of Conduct)



So, some updates:

On Tue, May 21, 2013 at 11:59:06AM -0400, Brian Gupta wrote:
> I'm wondering if there can be a guidelines/advice section also? For
> example, perhaps we want to capture something along the lines of (The
> wording needs a ton of tweaking): Please remember that Debian is a
> multicultural organization, with members from around the world, who
> speak a variety of native languages. Please think twice about how
> someone might take your words in the context of their own culture. In
> all cases, please remain respectful and professional in your dealings
> with other project members, especially if you do not know them well.
> Conversely, when reading communications from other project members, do
> bear in mind cultural differences.

I'm slightly reluctant to add too much non-normative content to any CoC.
It's supposed a "Code", not "Guide" of conduct; as such, it should only
contain the things that we really really think are necessary.

However, I do note that the version of the draft that I sent was
slightly older than what I'd been writing; I had a bit more language in
the penultimate paragraph about "assume good faith" and similar things.
The draft at the end of this mail contains it.

Does that alleviate your concerns?

On Tue, May 21, 2013 at 07:08:17PM +0900, Charles Plessy wrote:
> Hi Wouter,
> 
> many thanks for this initiative !
> 
> Here are some comments.
> 
>  - I think that we shoud encourage more private replies.  For instance,
>    "If you want to complain to someone who sent you a carbon copy when
>    you did not ask for it, do it privately" (from the current CoC), but also
>    for +1 messages, etc.  To balance this, we may mention that people starting
>    and fuelling a long thread would be very welcome to post a summary at
>    the end.

I disagree with that. First, in our social contract, we encourage
openness, not privacy. In addition to that, sending a reply in private
has several issues:
- People don't see the replies, which tends to result in having the same
  argument be repeated. Several times, if all of them reply in private.
  This is an issue not only for technical arguments, but also for
  "please behave" style of messages. In fact, in the latter case it can
  be more problematic, since receiving a high number of such messages
  may amount to a mobbing and have the opposite effect from what was
  intended.
- When replying with technical advice in private, you deprive the larger
  community of that advice.
- When replying with erroneous technical advice in private (as opposed
  to on a mailinglist), the chance of the error being spotted reduces
  tremendously.

>  - I think that in typical threads where the number of messages is expected to
>    be large (perhaps this one for instance), people should really do their best
>    to limit the number of messages they post.  If others agree, I recommend
>    to add this to the CoC.
> 
>  - How about recommending to let a discussion start before answering to an
>    email ?  Here is one interesting extract from another CoC:
> 
>     When responding to a very simple question, use the following algorithm:
> 
>      - compose your response
>      - type 4*runif(1) at the R prompt, and wait this many hours
>      - check for new posts to R-help; if no similar suggestion, post your response
> 
>       (This is partly in jest, but if you know immediately why it is suggested, you
>       probably should use it! Also, it's a nice idea to replace 4 by the number of
>       years you have been using R or S-plus.)
> 
>       http://www.r-project.org/posting-guide.html

I think this is going a bit too much in detail, but I suppose it's fair
to summarize these into a generic "don't overdo it on the posting"
paragraph? I've added such a paragraph to this draft.

>  - How about separating the technical and social aspects ?  I feel that comments
>    about Cc headers, length of lines or presence of HTML tags tend to inflate
>    tensions, rather than helping others to optimise their communication.

That makes sense, I suppose; I've removed the "form" bits from this
draft.

>    For instance I find some recommendations against to posting
>    borderline insulting.

Er, I've gone over that sentence several times now, and I'm not exactly
sure what it's supposed to mean. Can you clarify?

On Tue, May 21, 2013 at 03:30:08PM +0100, Ben Hutchings wrote:
> On Tue, 2013-05-21 at 10:32 +0200, Wouter Verhelst wrote:
> [...]
> > The Debian mailinglists exist to foster the development and use of
> > Debian. This Code of Conduct exists to help towards that goal.
> > 
> > In particular, the following rules should be adhered to by participants
> > to discussion on Debian mailinglists:
> > 
> > 1. Do not flame, use foul language, or in general be abusive or
> 
> 'flame' is slang and I suspect it is not that widely understood among
> those who are unused to mailing lists.  Try to find a standard English
> term instead.

I gave this some thought, but can't come up with a better term. I
suspect it is a domain-specific term rather than slang.

The English wiktionary lists it as:

(Internet) To post a destructively critical or abusive message,
especially to provoke dissent or controversy

(http://en.wiktionary.org/wiki/flame#Verb)

Since the proposed paragraph already contains alternatives such as
"abusive" and "disrespectful", it should be clear enough from context
what is meant here. As such, unless a better wording can be found, I
will leave this part as is.

(modulo editorial changes suggested by MJ, see below)

> >    disrespectful towards other people on the mailinglists or elsewhere
> >    in Debian. That type of behaviour is not constructive and can quickly
> >    lead to a degradation of the quality of a discussion.
> [...]
> > 4. [...] You should preferably also use a
> >    mailer which respects the Mail-Followup-To: header, or make a
> >    best-effort attempt at respecting it manually if you don't.
> 
> I think we should give up on M-F-T; it has never been standardised and
> is not widely supported.

(MJ Ray expressed a similar concern, see below)

> The most annoying reply behaviour I see is people replying to one list
> rather than the multiple lists I sent the original message to.  We
> should encourage use of Reply-to-all instead, as erring on the side of
> inclusion is safer than erring on the side of exclusion.

If switching to mail-followup-to would be controversial, switching to
reply-to-all would certainly be; it is a full 180 on our current
recommendation. While I agree it's probably the better option, I'd like
to see more consensus on this point first.

> [...]
> > 6. You should avoid sending attachments; this generates a lot of
> >    unnecessary bandwidth on our listservers. Instead, put the file you
> >    would like to attach online somewhere and post a link.
> 
> It may be worth clarifying that this applies only to the mailing lists,
> not the BTS.

The document would be called the "mailinglist code of conduct", and
would be posted on lists.debian.org; I think that should be clear
enough. Do you disagree?

> [...]
> > Thoughts?
> 
> I think it should incorporate the appropriate parts of the Debian
> Community Guidelines.

I've added a "further reading" section that contains a link to the dcg;
however, I am reluctant to turn guidelines into rules, especially over
that document's author's explicit objections
(<[🔎] 20130521121958.GA8347@enricozini.org>)

On Tue, May 21, 2013 at 04:57:23PM +0100, MJ Ray wrote:
> Wouter Verhelst <w@uter.be> proposed:
> > The Debian mailinglists exist to foster the development and use of
> > Debian. This Code of Conduct exists to help towards that goal.
> > 
> > In particular, the following rules should be adhered to by participants
> > to discussion on Debian mailinglists:
> 
> That second paragraph looks like it should get complaints from
> debian-l10n-english, so let's rephrase:
> 
> Participants in discussions on Debian mailing lists should follow these
> rules:

Sensible, thanks.

> > 1. Do not flame, use foul language, or in general be abusive or
> >    disrespectful towards other people on the mailinglists or elsewhere
> >    in Debian. That type of behaviour is not constructive and can quickly
> >    lead to a degradation of the quality of a discussion.
> 
> I feel it would be nicer to open with a positive point rather than a
> big "DO NOT".  I'd rephrase even "do not flame" as a positive
> instruction, so I'd reorder/rewrite the first three items like so:
> 
>   1. You're welcome to use our mailing lists to ask questions, but
>      please use the most appropriate list you can see.  If you are
>      unsure, use debian-user for support-related questions, or
>      debian-mentors for development-related questions.  Be prepared to
>      ask your question on a different list if told to do so, and
>      mention that it is a resent question.
> 
>   2. Avoid flaming, cursing and other abusive or disrespectful
>      behaviour as much as you can.  That usually distracts from the
>      real discussion and is not constructive.
> 
>   3. Use the correct language when sending mails to our lists. This is
>      usually English, unless otherwise noted in the description of the
>      mailing list in question.
> 
> I think "mailing list" is still usually two words, so I'd change that
> throughout.
> 
> Oh and I added mentioning that a question has been resent, as it can
> be annoying to find half a discussion months later.

Yes, that looks much better; thank you.

> >    [...] You should preferably also use a
> >    mailer which respects the Mail-Followup-To: header, or make a
> >    best-effort attempt at respecting it manually if you don't.
> 
> Yeah, I hope there's significant opposition to this change!  I think
> it's a great shame that Mail-Followup-To is still stumbling around, 15
> years after its fatal wounding in IETF DRUMS.  It still doesn't work
> and last I knew, mutt implemented it a different way to djb's spec and
> some other clients, which complicates its use.  For one typical
> discussion, see http://www.imc.org/ietf-822/mail-archive/msg05692.html
> 
> In short, I'd prefer the code of conduct to encourage people to take
> control rather than recommend rejected / divergently-implemented headers:
> 
>      You should check whether to reply to the List-Post address only,
>      or whether the original author would like to be a Cc recipient.
>      This may be indicated in the non-standard Mail-Followup-To header.

This does look better, indeed.

I updated the language in the next paragraph, too.

> > Repeated offenders may be temporarily of permanently banned from posting
> 
> s/ of / or /

That's a dutch-ism ;-)

Fixed, thanks.

New draft:

---
# Debian mailing list Code of Conduct

The Debian mailing lists exist to foster the development and use of
Debian. This Code of Conduct exists to help towards that goal.

Participants in discussions on Debian mailing lists should follow
these rules:

1. You're welcome to use our mailing lists to ask questions, but
   please use the most appropriate list you can see.  If you are
   unsure, use debian-user for support-related questions, or
   debian-mentors for development-related questions.  Be prepared to
   ask your question on a different list if told to do so, and
   mention that it is a resent question.
2. Avoid flaming, cursing and other abusive or disrespectful
   behaviour as much as you can.  That usually distracts from the
   real discussion and is not constructive.
3. Use the correct language when sending mails to our lists. This is
   usually English, unless otherwise noted in the description of the
   mailing list in question.
4. You should check whether to reply to the List-Post address only, or
   whether the original author would like to be a Cc recipient. This may
   be indicated in the non-standard Mail-Followup-To header.
5. If you wish to be part of a discussion, you should preferably
   subscribe to the relevant mailing list, even if only temporarily. If
   you choose not to, you should remember that you may lose out on part
   of the discussion, even if you explicitly asked to be copied on
   replies.
6. You should avoid sending attachments; this generates a lot of
   unnecessary bandwidth on our listservers. Instead, put the file you
   would like to attach online somewhere and post a link.
7. Please ensure that your mail system never sends automatic replies to
   the list. If you do, listmasters may remove you from the list with
   immediate effect to avoid flooding or annoying participants. You may
   resubscribe when the automatic messages have been disabled.
8. Replies to a post on a mailing list should, in general, go to the same
   mailing list. Do not send private replies, unless posting something
   sensitive. Do not change the mailing list, unless you are posting
   something that is no longer relevant to the original discussion and
   clearly off-topic for the mailing list where it is being discussed.
9. Try to go for quality, not quantity. While it may be tempting to send
   a large amount of email with repeated arguments to a thread about a
   subject that is dear to you, this usually doesn't help the
   discussion.

While these rules should be adhered to by participants, we recognize
that sometimes people may have a bad day, or be unaware of some of the
rules in this code of conduct. When that happens, you may reply to them
and point out this code of conduct. Such messages may be on the list or
off the list, whatever is most appropriate. However, regardless of
whether the message is on- or off-list, it should still adhere to the
relevant parts of this code of conduct; in particular, it should not be
abusive or disrespectful. Assume good faith; it is more likely that
participants are unaware of their bad behaviour than that they
intentionally try to degrade the quality of the discussion.

Repeated offenders may be temporarily or permanently banned from posting
to our mailing lists at the Debian listmasters' prerogative.

# Further reading

<link to dcg>
<link to docmuentation on what to do in case of technical problems>
(possibly more links?)
---

-- 
This end should point toward the ground if you want to go to space.

If it starts pointing toward space you are having a bad problem and you
will not go to space today.

  -- http://xkcd.com/1133/


Reply to: