On Mon, Apr 27, 2009 at 04:09:19PM +0100, Brett Parker wrote: > On 27 Apr 15:41, Noah Slater wrote: > > You're arguing that a Reply-To header is "harmful" (not that I am > > convinced) and > > Think of the occasions when you actually do want to do an offlist reply - it's > not that uncommon - having Reply-To set to default to the list causes a lot of > people to "get it wrong" because they're used to sensible mailing lists that > get it right - this happens quite often on one of the mailing lists I'm > subscribed to.
Considering that we're discussing on a mailing list, it's reasonable to assume that the common case is replying to the list. Why optimise for, what is surely by definition, the uncommon case? You're asserting that most people are "used" to "sensible" lists that don't set the Reply-To header. Unfortunately, without a survey to back up that claim I cannot believe it - based upon my own, extensive, experience. I realise that I am making a similar assertion. > > so people should learn to use some additional, uncommonly found, feature of > > their MUAs to work around the technological problem. I don't buy this > > argument > > at all. Technology should adapt to human behaviour, and not the other way > > around. There is something fundamentally wrong when we try to solve a > > technical > > problem with a Code of Conduct. > > It's not a technical problem, it's a social problem. Technical solutions to > social problems are always wrong. How is it a social problem? That clearly doesn't make any sense. There is only one problem here. Debian doesn't want mailing list replies to include the original senders. That isn't a social problem, that's a technical problem. When people are replying to mailing list traffic, they hit the button they've learnt to hit. There's nothing more complex going on. When a Reply-To header is added by the mailing list software, a larger proportion of people's default software interaction works as desired. > > Without a Reply-To header, we should expect people to Reply To Group. It > > doesn't > > matter if we have a Code of Conduct, people will always make mistakes. The > > only > > sensible thing to do in this situation would be to recommend that people who > > care properly configure their Mail-Followup-To and Mail-Reply-To headers. > > I wouldn't expect that. I'd expect that if they usually reply to the list they > would configure their MUA to reply to the list. I'm not sure what you mean by this. Can you clarify? > > If Reply To Group is so harmful that we want to avoid it completely, then I > > think we should consider adding a Reply-To header to the mailing list > > emails - > > like many other mailing lists do for exactly this reason. > > Many many more don't add the Reply-To header as it is harmful. Without evidence to back this up, we're both just making assertions. I just created a Google Group, with what must be the largest provider of hosted mailing lists currently in existence. The default setting appears to be the inclusion of a forced Reply-To header. Given Google's popularity, it seems reasonable to conclude that adding a Reply-To is more common. Additionally, I've not heard a convincing argument why Reply-To is harmful. It would solve the spurious CC replies, but at what specific cost? > > P.S. I had to manually edit the To and CC headers of this email before > > sending > > out because I had forgot to press the L key in mutt, one of the few clients > > that > > actually has such a feature. > > So, user error, not software error... This illustrates my point perfectly! It's not user error, because I'm just doing what I've learnt to do. When software use becomes habitual, usability is increased. This is how usability is defined, instead of some abstract sense. Software that behaves according to a user's mental model is easy to use. Forcing people to adjust their behaviour is a poor substitute for a technical solution. Software should adapt to human behaviour, not the other way around. Best, -- Noah Slater, http://tumbolia.org/nslater -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
