I’ve just read the Wikipedia page about Unparliamentary Language [1]. I recommend that everyone read it, if only for the amusement value, among other things it links to incidents where elected representatives acted in a way that would be expected of primary school children. The general concept of having rules about Unparliamentary Language is that MPs are permitted to say anything in Parliament without the risk of being sued or prosecuted, but certain things are inappropriate – the most common example is directly accusing another MP of lying. One of the main aims of rules against Unparliamentary Language is to prevent attacks on the honor of another member.
Having just witnessed a mailing list discussion go widely off track when a free software project was denigrated, it seems to me that we could do with some similar guidelines for mailing list discussions. The aim would be not to just prevent excessive attacks on the honor of other members but to also protect the honor of the free software projects. So for example one might recommend not using a particular program because of design decisions which seem dubious or a bad security history, but saying “it’s crap” would be considered to be inappropriate. Not that rejecting a program based on design decisions or a history of security flaws would be uncontroversial, but at least that gives objective issues to discuss so if there is a debate it will educate some of the lurkers.
Note that I’m not claiming to be better than other people in this regard, I’ve described software as crap on more than a few occasions. But I will try to avoid such things in future.
Finally does anyone have a good suggestion for a Free Software equivalent to the term “Unparliamentary Language”? It seems that to a large extent the support of certain ideas depends on having a catchy name and I can’t think of one.
The only problem with decreeing that one may not say “foo is crap” on
the bar mailing list is what happens when someone does so. Presumably
someone else replies with “You may not say that here” (or an
elaboration).
A lurker who then reads the archive sees “foo is crap” followed by
someone swiftly clamping down on such outspokenness! This reinforces
the original poster’s point, which I’m guessing is not the aim of the
policy…
The obvious response to this is that instead of writing “Don’t say
that here”, an author should write a balanced critique of why what the
original poster wrote is incorrect (or at least not sufficiently well
argued). But that’s the case with or without an “unlisty language”
rule.
Netiquette?
Rupert: In Parliament MPs can be ejected from a debate and even banned from debates for some period of time for Unparliamentary Language. That would probably work.
If the original poster’s point was “it’s a conspiracy against me” then being suspended from the list could be taken as supporting evidence. But if the point was “I used this program once, don’t care to give a full review, and just want to say it’s crap” then suspending them might do some good.
malte: Can you cite any Netiquette guide that has similar rules? Generally there are suggestions to avoid flaming people etc, but nothing to do with the case of being disrespectful to a random software product.
Note that Free Software is very different to proprietary software in this regard because in any largish mailing list when a moderately popular package is described there will be someone who has contributed to the package and takes the criticism personally.
etbe: Hmm, but what do you do with the archived post? I agree that
banning the trolling moron from the list is sensible, but you’re still
left with an email saying “foo is crap”. Do you delete it from the
archive? Modify it to add a huge “[We don’t agree with this -ed]”?
Rupert: Comments are NEVER deleted from Hansard, so if we extend the analogy there would be no changes to mailing list archives for that reason alone. On a practical matter, people rarely start new flame-wars based on list archives, it’s all what’s in their inbox.
Also I am not just talking about trolls (really there are much better ways to troll). When asked for advice on which program to use someone might give a well considered comment that “program A is crap”, they may have good reasons for their opinion based on it failing in some repeatable and objective ways in use-cases that are important to them. But merely saying that it’s “crap” doesn’t help other people who might have different use-cases for which the program works, doesn’t help developers fix the bugs, and can just distract everyone.
The term “mudslinging” comes to mind, though it probably doesn’t cover as wide a range of behaviour as is suggested by “unparliamentary language”
When I was helping run the OpenSUSE Linux forums we routinely had trolls and various fights break out amongst members. What worked for us at the time were the following:
* For first time trolls, a moderator would post a simple reminder of community rules.
* For hotbutton issues that were obviously heading down the path of unparliamentary language, a moderator would give a single warning in-thread to settle down (with link to community rules), then monitor each involved member’s contributions across the forum for 1 week. Occasionally hotbutton issues would have their threads closed. It was very common to warn two arguing members in one thread, watch that thread die down or get closed, and then see the offending members fight over a completely different thread within 2 days.
* For any thread involving obvious personal threats or unparliamentary language, an administrator would either warn or close the thread, then send personal messages reminding the involved party members of rules. In serious cases, members would be banned from the forum for 1 week to 1 month. Only two or three times over my four years did we permanently ban members from the community.
We never deleted anything so as to have a clear paper trail as to which member actions lead to which administrative decisions.
I think for mailing lists (which obviously don’t have the “close the thread” option) a clear statement of rules, warning conditions, and tiered bans (1 day, 1 week, 1 month, permanent) would be very acceptable.
Andrew: That’s interesting, thanks for the summary.
It’s worth noting that you CAN close a thread on a mailing list. It’s not difficult to configure a mail server to reject mail that matches a particular subject. It wouldn’t be difficult to have a cgi-bin script that the moderators can run to add new entries to a list and a cron job to remove entries after a year.
I’m not aware of anyone doing this, but there is no technical obstacle.
Frankly, I find the whole idea of a blacklist of words to be criminally shortsighted.
You’re not allowed to say “you’re lying”, but you /are/ allowed to say “you’re using terminological inexactitude”? You’re allowed to insult any person to whatever extent you wish, but you’re not allowed to say “damn”, “drunk”, or “fuddle duddle”? What’s the point?
Bad behaviour is an intent, not a choice of words. To create a blacklist of words in order to prevent bad behaviour is a bit like hoping that a stateless firewall will filter bad traffic on the internet; it will make bad behaviour slightly harder, but is trivially bypassed.
I agree that a word blacklist is a bad idea, this has been addressed in Northern Ireland, here is an extract from Wikipedia:
The Speaker of the Northern Ireland Assembly, William Hay MLA, gave a ruling in the Chamber on 24 November 2009 on unparliamentary language.[13] In essence rather than making judgements on the basis of particular words or phrases that have been ruled to be unparliamentarily in the Assembly or elsewhere the Speaker said that he would judge Members’ remarks against standards of courtesy, good temper and moderation which he considered to be the standards of parliamentary debate. He went on to say that in making his judgement he would consider the nature of Members’ remarks and the context in which they were made.