I have to agree here. Not being able to override the reply-to: fields on lists is a *MAJOr* problem. I've also had this problem with my lists. And I was hoping for a 'fix' but it looks like it's intended behaviour?.

I think if thats going to be the case there should be another option. 'Force list replyto' .

Regards Simon

On Tue, Nov 14, 2000 at 12:43:23AM -0500, Phil Barnett wrote:

On 14 Nov 2000, at 0:32, Barry A. Warsaw wrote:

...

...

...

...

...

> "PB" == Phil Barnett midnight@the-oasis.net writes:

PB> Am I likely to encounter problems going from 2.0rc2 to 2.0b6?

Do you mean downgrading? Don't do that! If you really mean from 2.0b6 to 2.0rc2 then you should be fine. Any data formats that need updating will get updated automatically.

I specifically mean downgrading. I have a hornets nest on my hands and I'm about to lose my lists back to egroups.

This lack of reply to the list 'updated feature' has them HEATED.

Perhaps I'm not clear. If I lose them to egroups, there is no point to me running Mailman anymore.

This 'upgraded feature' is suicidal for Mailman on my machine.

-- Phil Barnett mailto:midnight@the-oasis.net WWW http://www.the-oasis.net/ FTP Site ftp://ftp.the-oasis.net

---

Mailman-Developers mailing list Mailman-Developers@python.org http://www.python.org/mailman/listinfo/mailman-developers

At 5:50 PM -0800 11/14/00, Marc MERLIN wrote:

2. If you don't want to take that advise, you certainly shouldn't overwrite a users' reply to with the list's

I'll disagree. If you feel you must coerce reply-to, coerce it unconditionally. You've already decided the list's needs outweigh the individual's needs, or you woudln't be coercing in the first place Once you make that decision, whether or not the user ALSO uses a reply-to is meaningless, and trying to take that into consideration only confuses things, because then the list has its reply-to coerced -- sometimes, and you have to know how to read mail headers ot figure out when. Not good for the typical user, who may never figure out why something isn't working randomly.

At 11:42 PM -0500 11/14/00, Tanner Lovelace wrote:

There are arguments on both sides for this issue. I personally think it depends a lot on what the people on your list prefer.

But you need to be wary of making the mistake of "who's making noise" being "what the people on the list prefer". Becaues I've found, if I survey my lists, that the ones making the noise *don't* represent the overall preferences of the list, but are instead a noisy minority. Just because someone's complaining doesn't mean they're right....

On 14 Nov 2000, at 1:07, Barry A. Warsaw wrote:

I'm surprised. I would think that munging reply-to is only useful for lists of people who don't know how to drive their MUAs. Is the problem just that they got used to one way of doing things, and now there's a different way? That I can sympathize with. Is it that suddenly, some people are getting two copies of messages because people don't know how to trim their headers? Yeah, that sucks.

Or is it something else?

I use Pegasus. My replyto is global. If I unset my replyto, it's removed from ALL of my mail.

I don't really want to be going to my configuration page and blanking it for some mail and turning it back on for other mail. First, it's a hassle. Second, I'll undoubtedly do it wrong at some point. I want a replyto in my mail.

Every mail list that I have participated in overrides replyto when replyto is set to the list. To my fellow list users who have been using lists that do this for the last three years, not having it work that way is disconcerting to say the least.

egroups and opensource.org are two examples of list providers that override replyto.

But, I certainly understand the desire to have it work the way you have it set now. I just think the behaviour should be selectable. It doesn't have to be all or none.

Thanks for the patch!

Now, if I could create a new list, I'd be back on track...

At 10:25 AM +0000 11/14/00, John Block wrote:

...

I'm surprised. I would think that munging reply-to is only useful for lists of people who don't know how to drive their MUAs.

MUA?

Mail User Agent (aka client)

Other jargon:

MLM (mail list manager) MTA (mail transfer agent)

In general I'm against coercing Reply-To. Strongly.

But there are times when it's the right/necessary thing to do, so you want that option. In general yous houldn't use it, but sometimes you need it. One classic case is a mailing list under NDA discussing a beta software release, where all e-mail has to be logged and evaluated, and the *wrong* thing to do is reply privately to a person, because the beta team needs a copy. In that case, reply-to gets coerced to the list, and that reply-to needs to be dominant (i.e., an existing reply-to from a user *can't* take precedent).

At 11:40 AM -0500 11/14/00, Barry A. Warsaw wrote:

That's the key thing that I don't like about Reply-To munging: it makes it much more difficult to do private replies. Maybe it's just the lists I manage, but my users are just as heated about /not/ doing reply-to munging as apparently other users on other lists are about doing it.

It's a religious issue to some level, but I've done any number of studies (at least half a dozen) on it with my users over the years. I've never found *any* list or user population where the majority wants reply-to coerced. What you really end up with, if you survey the *entire* user base, is a noisy minority that wants it, and that minority is under 20-25% of those that have opinions.

So what ends up happening is that because they're the ones making noise, the list gets set that way. But if you survey the entire user base and get a non-biased sample, they're the minority viewpoint (the same is true of Subject line tweaks, like the [Mailman-users] flag. Only with those, I've never found a list where even 5% of the users wanted it, and there's usually a solid majority that hate the damn things.... But again, you tend to run into noisy minorites that push their preferences, and unless you take the time to go out and run formal list surveys (which are time consuming), it's hard to tell whether it is a squeaky wheel or whether it's a list consensus...

The only grumblings occur when people don't trim their CC headers (like I've done here) and folks start getting duplicates.

Another vocal minority. Less than 1% of a typical subscriber base cares about this. Those that do tend to be very sensitive to it and vocal, but whacking the server for that group isn't a smart idea (IMHO), since there are client ways of doing it, like message-ID trapping.

That can be handled in other ways (e.g. the list /could/ suppress deliveries to list members that it sees explicitly in the recipients list, although as we've discussed before, that has some potential for abuse).

True, but it's what I'd like to see happen down the road, at least as a configuration option for the server admin. If you know they're getting it direct, why send them a second copy? or maybe tag this to a users "metoo" flag? If they've told the server not to send them copies of their own posts, assume they also don't want duplicates? That might be the easiest way, and leave it by default off, but then users can set it if they care.

I've slept on this one and I'm prepared to change things for 2.0 final so that if Reply-To munging is turned on, it'll override any existing Reply-To field in the original message. The deciding factor for me was realizing how difficult it was to send private replies with munging turned on, so it might as well be equally difficult for every poster.

I think that's the right thing to do, and it sets you in line with how other MLMs work, so it better matches user expectations, I think...

It's a touch issue, because there are personal preferences, because Reply-to is used for different things (not always correctly), and because it just plain old isn't all that well defined as a header, despite it's long-standing usage...

At 5:39 PM -0500 11/15/00, Gerald Oskoboiny wrote:

Another solution to this is the Mail-Followup-To header: (not widely supported yet)

http://cr.yp.to/proto/replyto.html

No, it's actually stillborn... dan may still be pushing it, but I don't think anyone else is.

Since the new RFC supports list-post, we really don't NEED Mail-Followup-To. Instead, an MUA can look for the existance of List-Post and use it to enable reply-to-list.

...

...

...

...

"CVR" == Chuq Von Rospach chuqui@plaidworks.com writes:

CVR> Since the new RFC supports list-post, we really don't NEED
CVR> Mail-Followup-To. Instead, an MUA can look for the existance
CVR> of List-Post and use it to enable reply-to-list.

Chuq, are you referring to draft-ietf-drums-msg-fmt-09.txt of 11-Sep-2000? There's no mention of List-Post in there that I can find.

-Barry

...

...

...

...

"CVR" == Chuq Von Rospach chuqui@plaidworks.com writes:

CVR> Since the new RFC supports list-post, we really don't NEED
CVR> Mail-Followup-To. Instead, an MUA can look for the existance
CVR> of List-Post and use it to enable reply-to-list.

Duh. List-Post: is in RFC 2369 and of course Mailman already inserts this header in messages. Sad what happens to short term memory when you get older. :)

BTW, I think this is the right bit of information that an MUA would need to support a "post-to-list" option. Seems to me all the information is there without Reply-To munging for the user agent to support whatever followup/reply operation it wants. Reply-To munging should be considered a hack that's (questionably) necessary until the MUAs catch on.

-Barry

At 1:07 -0500 11/14/00, Barry A. Warsaw wrote:

I'm surprised. I would think that munging reply-to is only useful for lists of people who don't know how to drive their MUAs. Is the problem just that they got used to one way of doing things, and now there's a different way? That I can sympathize with. Is it that suddenly, some people are getting two copies of messages because people don't know how to trim their headers? Yeah, that sucks.

I'm not on such a list (we're expecting to switch to Mailman when it settles down). But I would be perturbed if on a single list sometimes the Reply button sets up a reply to the list and sometimes it sets up a private reply. [I can--and do--deal with both kinds of list, and both are a nuisance some of the time.]

The idea that Reply's result on a single list depends on who* sent the message seems like a faulty solution to the unsolvable problem that neither reply-goes-to-the-list nor reply-goes-to-the-sender is right. [Right would take mindreading: reply goes where I expect *this* reply to go.]

*Or, even worse, where a given person was were when the message was sent.

--John

John Baxter jwblist@olympus.net Port Ludlow, WA, USA