On 20/01/18 20:05, Mark Sapiro wrote:

On 01/20/2018 10:18 AM, Hal via Mailman-Users wrote:

...

I'm a little confused about the "reply-to" setting as I was pretty sure I had set my list up so that all replies by default go back to the list, but for some reason a reply goes directly to the sender.

recent T'bird if the message has a List-Post: header and T'bird offers a "Reply List" button, "Reply" will ignore Reply-To: if it's the list address and reply to the From:. In more recent T'bird, you can restore the expected behavior be setting mail.override_list_reply_to False in the config editor (see <https://bugzilla.mozilla.org/show_bug.cgi?id=1392371>), but this has to be done by every list member that uses T'bird.

There are other possibilities, but I think the above is the likely issue in your case.

Thanks Mark! Yes, that was the problem here as I'm a Thunderbird user. Changing the above preference now makes it possible to just press "Reply" and it goes to the list.

I did notice a "Reply to list" option ("Message"-"Reply to list" menu) but I like to keep things simple, so a single "Reply" button is much preferred.

I've posted a message to my mailing list about this so that hopefully every Thunderbird user will do the same (only to remember to do a "Reply to all" and remove the list address for reply-postings not meant for the list but rather directly to the sender).

Apart from Thunderbird, are there other email apps which cause this issue?

Hal

On 01/20/2018 12:05 PM, Mark Sapiro wrote:

What actually happens with "reply" depends on a few things. If the mail client involved is Thunderbird, it doesn't behave as expected. See <https://bugzilla.mozilla.org/show_bug.cgi?id=1309486>. In short, in recent T'bird if the message has a List-Post: header and T'bird offers a "Reply List" button, "Reply" will ignore Reply-To: if it's the list address and reply to the From:. In more recent T'bird, you can restore the expected behavior be setting mail.override_list_reply_to False in the config editor (see <https://bugzilla.mozilla.org/show_bug.cgi?id=1392371>), but this has to be done by every list member that uses T'bird.

Oh for $&#* sake. I'm not going to get into the pros / cons of either way.

I tried following the bugs to discern the actual behavior changes that mail.override_list_reply_to changes. But I was unable to make heads or tale of what it does. Further, my tests didn't shed any light on things.

Will someone please enlighten me on how mail.override_list_reply_to behaves when set to true (the default) vs false?

I couldn't keep track of the previous behavior and current behavior through all of the things that I read.

-- Grant. . . . unix || die

On 01/22/2018 10:37 AM, Mark Sapiro wrote:

With the default mail.override_list_reply_to = False, for a message with a List-Post: header and with the list posting address also in a Reply-To: header, T'bird will ignore the Reply-To: header and address a "Reply" to the From: address.

Setting mail.override_list_reply_to = True will restore RFC compliant behavior and address a "Reply" to the Reply-To: header address(es).

Thank you for the concise answer Mark. :-)

What I find interesting is that mail.override_list_reply_to is set to True by default in my copy of Thunderbird, 52.5.0.

-- Grant. . . . unix || die

On 01/22/2018 12:17 PM, Mark Sapiro wrote:

My bad. I was confused. In my answer above, "False" should be "True" and vice versa.

;-)

-- Grant. . . . unix || die

On 01/24/2018 01:50 AM, Stephen J. Turnbull wrote:

I'd appreciate if those who have strong opinions on this would take a look at the analysis below and tell me if I'm missing something. If not, maybe I'll write up a BCP (non-standards-track RFC[1]) so it's on record.

See my comments inline below.

This proposal actually has a history going back to about 2005. I didn't do anything about it because I got a lot of pushback from MUA writers, and writing RFCs is worse than writing PEPs (Pythonistas are either sane or go away soon, not so for IETF mailing lists . But if its still an issue maybe it's worth the effort.

I think there's an obvious algorithm for "smart single reply":

I doubt that's the case.

1. If there is a Reply-To, address the message to Reply-To.

Baring other influences, this is where the author or the message sender (if it's someone other than the author) wants replies to go to.

2. Else if there is a List-Post, address the message to List-Post.

I don't think that it's appropriate to always prefer the List-Post over the From ~> Reply-To.

MUAs (are starting to) have separate functions for reply to From / Reply-To vs reply to list.

I can see a case for a broadcast mailing list that's open for all members to post to where neither From nor Reply-To munging takes place. The author can send from one address (From) and want replies to go elsewhere (Reply-To) while the MLM adds a List-Post header to comply with other standards. I feel like a reply to such a message should go to the Reply-To (as set by the author) and not the List-Post as set by the MLM.

Reply-List in such a case is a distinctly different operation.

I can also see a case where a message author might choose to (dynamically) set the Reply-To to something like "Reply-To: Please reply to the Mailman-Users mailing list. <mailman-users@python.org>"

3. Else address the message to From. (If there's no From, the message violates the most basic RFCs so all bets are off.)

¯\_(ツ)_/¯

Sounds like the classic case of "undefined behavior" to me.

Assuming that no lists munge Reply-To, I think you'll agree that this is what you want 90% of the time (conservative estimate). There are some issues with this algorithm in practice:

I disagree for a number of reasons. Some of which are outlined above.

1. Some lists should not encourage reply-to-list (eg, for privacy reasons). This can be worked around by omitting List-Post, or solved by additional protocol so that the list sets a header field saying "don't automatically reply here just because there's a List-Post. Given how conservative MUA writers are, I'd say "KISS" for these, and make users cut-n-paste. Most of the time a reply-to-list here is probably thread hijacking anyway.

I see an opportunity for a "List-Reply-To" header that could indicate if /replies/ should go to the list (List-Post) or the author (Reply-To|From). I suppose that it could also be possible to specify an alternate address for replies to go to, i.e. for thread tracking or something like that.

This would still leave us in the situation where MUAs need to differentiate between a generic Reply and a Reply-to-List behavior. Plus the associated action for the reply keyboard sequence.

I feel like this is /mostly/ a user education issue. There may be some room for UI / UX improvement. Ultimately it's up to the MTA to do what the user wans done. Consider the following:

From: Author <author@author.example> To: List <list@list.example> Reply-To: Author <author+list@author.example> List-Post: List <list@list.example>

Where should replies to the author go to? Where should replies to the list go to? Where should the (undefined) "reply" go to?

I don't think that it's likely for the MTA to automagically know what needs to be done.

2. Some users will want to override the algorithm and reply specifically to list or author. MUAs should provide buttons or menu items for these infrequently used options.

I think it is wrong for us to ascribe frequency of use for other users. Just because I do something some way does not mean that others do so with the same frequency, or even the same thing.

I personally use Reply List more than I use Reply (Author [From|Reply-To]).

3. Your favorite list munges Reply-To. Nothing changes here, people are still going to be embarrassed by sending remarks intended to be private to a broad audience, and in some configurations of Mailman the original Reply-To or the From will get dupes. At least you can override with a reply-to-author function.

I feel like this is a user education issue. Sadly, pain of embarrassment is a good teacher.

I don't understand what Thunderbird thought they were doing. <shrug/>

I think they were trying to apply a technological solution to what I believe is fundamentally a user education issue.

-- Grant. . . . unix || die

On 1/28/2018 8:40 PM, Stephen J. Turnbull wrote:

I believe that many users think of mailing lists as fundamentally different from personal email, and they would like their MUAs to distinguish automatically. This algorithm, I believe, would do a pretty good job of that.

This particular user distinguishes between mail to one (human) recipient and mail to multiple recipients, but the difference between two and a thousand is only shades of gray, and whether some are from mailing list expansion is mostly unimportant.  Either I want to reply to the author alone, or to everybody, or (rarely) to some other subset.  The obvious Reply and Reply All behaviors handle the first two, and the last is probably best handled as Reply All followed by editing the address list.

I suspect that there will always be disagreement as to what a single "one button reply" button should do, whether it should reply to the author or reply to everybody.  I doubt that there will ever be a solution, server-side or client-side, that will make everybody happy.  I can only hope that whatever standards develop make both "reply to author" and "reply to all" convenient.

(And that's another of the key items:  the "Reply-To: <list>" configuration makes it *difficult* to reply to the author, and that seems just plain rude.)

Side question:  when you have a message addressed to multiple mailing lists, what does "reply to list" even mean?

...

I want "Reply" to go to the author, and "Reply All" to go to the author, the list, and any other To or CC destinations.  I simply can't understand any other answer.  I don't understand why anybody feels a need for "Reply List".

Your preference is noted, but you are definitely in a minority of those whose opinions I've seen over the decades. Even those who use Reply and Reply All as you do (I do on this list, for example), usually have considered it suboptimal. The preferences of list owners also should be respected, to the extent that replying users don't care. The prevalence of reply-to-munging says that they (or perhaps a majority of their subscribers) want replies to automatically go to the list.

Lists at my company are simply never configured that way; I don't think our e-mail system even has the option.  (And at ~140K users and thousands of mailing lists, that's not a trivial data point.)

Note also that the MailMan UI says "Where are replies to list messages directed? Poster is /strongly /recommended for most mailing lists." so it's not just me.

Interesting.  I was going to say that none of the FOSS lists that I participate in use this configuration, but it seems that a couple do and Thunderbird's mail.override_list_reply_to is silently saving me from their misbehavior.  Yay, T-bird!  Though, while I appreciate the fact that the default is the way I want it, I have to reluctantly say that it's wrong.  It should respect the Reply-To by default, no matter how wrong it is.  But note also:  the fact that the T-bird authors chose this behavior by default suggests that they are not members of the "Reply-To: <list>" community.

(I'm not sure whether T-bird can save me from a DMARC-munged list that uses "Reply-To: <list>".  That combination just makes my head hurt.)

On 1/29/2018 9:56 PM, Stephen J. Turnbull wrote:

...

(And that's another of the key items:  the "Reply-To: <list>" configuration makes it *difficult* to reply to the author, and that seems just plain rude.)

Why? Nobody is talking about taking away anybody's Reply-To-Author function, and nobody says you personally have to bind "smart reply" to anything in your MUA.

If you have "smart reply" as a separate function, yes.  If you have the typical "Reply" and "Reply All", and the mailing list software sets "Reply-To: <list>", then replying to the author is awkward and error-prone.  RFC-compliant MUAs are unlikely to have a simple operation that replies to the sender.

...

Side question:  when you have a message addressed to multiple mailing lists, what does "reply to list" even mean?

Long answer: click here -> https://tools.ietf.org/html/rfc2369 Short answer: List-Post may occur at most once. It goes there.

So for the general case where you might have gotten a message directly, and through list A, and through list B, the result is random unless you pay careful attention to how you got this particular copy of the message.

...

Note also that the MailMan UI says "Where are replies to list messages directed? Poster is /strongly /recommended for most mailing lists." so it's not just me.

Opposing "Reply-To munging" is nowhere near advocating restricting reply UI to "Reply-to-Author" and "Reply-to-All", no more, no less. In fact, my opposition to Reply-To munging is a good part of *why* I think "smart reply" would be a useful addition to AOL's MUA, inter alia.

OK, so maybe we aren't so far off alignment.  We might choose different options, but that's OK.

It sounds like neither of us want the list to set "Reply-To: <list>".

You want a "smart reply" button that sends to Reply-To, List-Post, or From, in that order.  (Right?) I want plain "reply" that sends to Reply-To or From, in that order.  (I don't mind if it's renamed to "Reply to Author".)

I wouldn't use your "smart reply" button, because I think it does the wrong thing for mailing lists, but if you want to do the wrong thing with your replies, I guess that's up to you.

My only fear is that in the ongoing simplification (dumbing-down?) of this stuff, "smart reply" will become the only option.  And, actually, if that happens then I *have* lost the "reply to author" function.

Jordan Brown writes:

If you have "smart reply" as a separate function, yes.  If you have the typical "Reply" and "Reply All", and the mailing list software sets "Reply-To: <list>", then replying to the author is awkward and error-prone.

Sure, but in this thread we all agree that Reply-To munging is problematic. I believe that the widespread availability of "smart reply" would greatly reduce the incentives for lists to munge. It doesn't eliminate the need for reply-author and reply-all, but in my experience it reduces them dramatically, and basically eliminates the need for reply-list.

The question I asked, which you misinterpreted completely IMO, and Grant partially agreed with is "Does an algorithm which 1. gives overriding precedence to Reply-To, 2. otherwise if List-Post is present directs it there, and 3. finally falls back to From, seem likely to DTRT most of the time?" I know from analyzing my own posts that for me the answer is yes, and from experience I know that I'm reasonably good at flexibly using different reply functions to meet the needs of the moment even though I spend a lot of time using just one function.

I also think that the evidence is pretty strong that the Reply-To mungers would like this functionality. To them, it would seem like their MUA DTRTs even for non-munging lists.

 > RFC-compliant MUAs

... don't exist. RFCs about mail do not apply to MUAs, they apply to the interpretation of the messages. An MUA can do anything it wants to. It's up to the user to decide whether they like the results or not. But the user is ultimately responsible for conformance, not the MUA.

It's true that if a message composed by a borked MUA (eg Thunderbird in some of its recent incarnations, or whatever the MUA is that encourages people to add unneeded Reply-Tos) is interpreted according to RFC 5322 et al, the result is often surprising to the user. But you shouldn't assume that proposed features are going to be added in a borked way. (Statistically, of course, some *will* be borked on introduction. That just means we need to be ready to fix them!)

Here my main question is whether for *many* users "smart reply" would "DTRT" enough to streamline their UI and reduce mistaken addressing. It should be an additional option, not replacing any of the now- traditional features (reply to author, reply to all, reply to list). (However, in my experience it completely replaces reply to list.) If it's not close enough to what a large number of users want to be the recommended binding for the "obvious" gestures for "generic" reply, I don't think the additional variety (complexity) in UI configuration is worth it.

So for the general case where you might have gotten a message directly, and through list A, and through list B, the result is random unless you pay careful attention to how you got this particular copy of the message.

Yes and no (I partly disagree with Mark here). It's definitely deterministic, and *not* random, but to users it may seem arbitrary. This can be mitigated in many cases by list owner coordination and subscriber setup. So, for example, a subscriber who sorts different lists' traffic into different folders is likely to be aware which list it will go to (the List-Id and List-Post fields will be consistent throughout that folder). If it's a group of related lists with different topics (such as mailman3-users, mailman-users, and mailman-developers) there's likely to be strong social strictures against cross-posting. And with hierarchical structure, the owners can set things up to DTRT. Examples:

Announce list: does not set List-Post Users list: does set List-Post (cross-posting discouraged) Dev list: does set List-Post (cross-posting discouraged)

The announce list being gated to users and devs respectively "obviously" should only get posts from approved sources. This DTRTs.

Advisees: does not set List-Post Undergrads: does set List-Post (cross-posting discouraged) Grads: does set List-Post (cross-posting discouraged) Graduates 2018: does not set List-Post Graduates 2019: does not set List-Post Graduates 2020: does not set List-Post

The "all my advisees" list functions as an announce list (when I'm out of town, all-hands meetings and university deadlines announcements, etc). The undergrads and grads are socially distinct and occasionally have discussions among themselves. Rarely the 1st, 2d, 3d year grad students have discussions specific to that year, but mostly discussion relates to seminar presentations and general research methods, so it's not that inconvenient to have no List-Post for individual classes. It's not perfect DWIM, but it's pretty close. I don't often see this as being a problem, and lists that set Reply-To would surely make it worse.

So overall, I don't see this as likely to be a big problem in practice.

It sounds like neither of us want the list to set "Reply-To: <list>".

I think that's pretty general feeling among list management software developers. Users and list owners, on the other hand, frequently disagree. I believe that is due to a deficiency of MUAs, to wit, not offering a "smart reply" function and not encouraging its use by less sophisticated users.

You want a "smart reply" button that sends to Reply-To, List-Post, or From, in that order.  (Right?)

I would (and do, where available) use it, but I really want it as an option, even default, for the crowd of users who expect Reply-To to be set to the list with current MUAs.

I wouldn't use your "smart reply" button, because I think it does the wrong thing for mailing lists,

I don't understand why you think that. So far you have consistently responded to this thread on-list AFAICS, and everybody in this thread got here by reading it on the mailing list (all first responded to a mailing list post, not to one where they were personally addressed).

I think that is by far the majority case. It is *quite* unusual for non-subscribers to be explictly addressed in my experience (see below for an exception), and (for spam control reasons) it is very common for lists to make it difficult for non-subscribers to post. It is not unreasonable in this day and age to assume that all posters are subscribers, and that all explicitly named addressees got there because of "reply all" rather than because somebody added a non-subscriber. On this list, and many like it, I personally would rarely ever need anything other than "smart reply", and it also works fine for most of my personal mail, which almost always should get a reply only to author (or Reply-To, which smart reply does).

But I'm interested to hear about use cases where that is *not* the case.

For example, I participate in one list at work where "smart reply" would often do the wrong thing, because the subscribers to that list are fellow faculty and staff who deal with applicants to the grad program. I end up splitting between smart-reply (which goes to list for discussion of committee policy and the like) and reply-all (which goes to the list so others know I dealt with that case, and to the non-subscriber to whom I provide information). However, this decision *must* be made case-by-case, depending on the nature of the reply: relying on reply-all (as I do on most list traffic when my MUA doesn't have "smart reply") would quite frequently do the wrong thing and require non-list addresses to be edited out. Even when reply-all is appropriate, a non-subscriber frequently needs to be added by hand to coordinate a student-adviser pairing (mostly by the staff, not often by faculty members). Reply-author gets used, but quite rarely because of the nature of committee work. So even though I can't rely on smart reply to DWIM, it is nevertheless useful, and fails safe (we rarely care all that much if any given applicant falls through the cracks, and most cases where the mail goes to the list but not the applicant gets caught same day, but internal committee discussion should *not* go to random non-university people!)

Of course that's just *one* use case. That's why I ask for others.

but if you want to do the wrong thing with your replies, I guess that's up to you.

It's not about *me*. I have this feature in my own MUA (or will, as soon as I re-add it to this new upgraded version). And it does the *right* thing for me more than 95% of the time. I still don't understand what makes you think it would do the wrong thing for *you*, let alone the wrong thing for *me*, except that you object to it being labelled "Reply" because you associate that very strongly with "Reply to Author".

My only fear is that in the ongoing simplification (dumbing-down?) of this stuff, "smart reply" will become the only option.  And, actually, if that happens then I *have* lost the "reply to author" function.

I don't think that level of paranoia is justified. Sure, some dev organizations will make that kind of mistake, as we've seen with Thunderbird. But all of the MUAs I know that do have specialized reply-to-list (mutt, Gnus) have very flexible interfaces for binding UI gestures to functions, and far more available functions than "one-click" or "one-key" gestures. If yours doesn't, then yes, you're at risk that a whim of the developers you could lose essential functionality. But that's a problem with your MUA and its dev team, not with the suggested new functionality.

Steve

Jordan Brown writes:

On 2/5/2018 12:29 AM, Stephen J. Turnbull wrote: [various stuff, citation line preserved to make a point below]

You don't mention what your "smart reply" does with To and CC addresses.  Discards them, I assume?

Yes. It's intended to do what a certain large group of naive users expect "Reply" to do. "Smart Reply" is not intended to do what "Reply All" does, period. It is intended to provide a variant on what "Reply" does that automatically handles the common case (on the Internet at large) of discussion lists where all participants are subscribers. It always picks an unique reply address (except in some rare cases where From or Reply-To contains multiple addresses).

But:  in my work contexts, it is quite common for a discussion to span two teams.  Again, a "reply" that goes to the List-Post address (versus All) won't do the right thing.  Normal "Reply All" does the right thing.

OK. I'm assuming that each team has its own list, only one List-Post is present, so you need Reply-All even for list posts, right? I almost never need to reply to such posts, and when I do, it's invariably only to the list I received it from. If reply-both is a frequent use case on your lists, I can see how smart reply would almost never DTRT for you.

But:  Even in a mailing list context, I think that "To: <author> CC: <list>" conveys useful context; I'm replying to what *you* said, and including everybody else in the audience.  Reply All does the right thing.  (Yes, it's suboptimal in that the To/CC list tends to accumulate people over time [...].)

I find the suboptimality aspect more important, and determine who was replied to more from the citation line than the addressee list. YMMV, of course.

As for "knowing who's on the list", almost all of the lists I'm on require membership to post, including "private" lists (address books do get pwned), and deliberate addition of 3rd parties is *extremely* rare, except for certain announcements. But in that announcement use-case, for me Reply-Author is the only mode I ever use.

I do *not* want my "Er, did you really mean to say <stupid mistake>" note to go to the entire audience.

This turns out not to be a problem for me. Smart reply has never tricked me into addressing a private reply to the list (let alone actually sending one). I suspect it's unlikely to catch my main audience for the feature very often -- and much less often than Reply-To munging does. But to find out I need to get it into a mass MUA as default. :-/

 > Normal "Reply" does the right thing (assuming non-munged > Reply-To).

Reply-To munging is precisely the issue this is intended to address. Munged lists *have* caught me (although actually sending a message misaddressed to list is extremely rare). I think the difference is that when I use "smart reply" I have implicitly requested that it go to the list. If I really want to reply to author (which is not that rare), I do use Reply-Author, and find it natural. (I'm not saying you would.)

than Reply or Reply All, as appropriate.  (If you're interested, I'll see if I can do an analysis of my message traffic to see how often it would do something that I would consider to be clearly wrong and how often it would be an improvement.)

I would be interested in that. I expect that you'll find a pretty high ratio of wrong to right. But if it came out anywhere near even, it would be a pretty strong indication in favor of writing an RFC. I don't expect that to be enough to interest you in changing (there would be muscle memory costs, etc).

I would appreciate it if you would *not* count "omitting the author of a list post from the reply" as "wrong" for this purpose because I don't think my target audience for "smart reply" would count it as wrong.

On what might be a side note, I think there might be a key difference in attitude between different camps.  One side wants to keep discussion on the mailing list when possible;

This feature is definitely aimed at that camp. I'm not interested in discussing whether encouraging them is a bad thing in this thread. If you want to talk about that (it does matter to me, it's just a separate discussion), let's start a new thread or we can go offline.

It is of course completely deterministic.  But note that I said "unless you pay careful attention to how you got this particular copy of the message".

I disagree with that. In your case, where you apparently get a lot of cross-posts, yes, you'd need to pay attention. But I think you'd figure it out quickly because AIUI you'd normally be on your team's list but not the other. Your operational problem wouldn't be figuring out which list delivered it, but simply that you frequently need to respond to both, and the list you're not on would be omitted unless you Reply-All. Am I missing something? "Smart reply" is simply not designed to be useful in such cases.

However, in many cases, such as the two I described, there is not a big problem unless I'm on both lists that provide List-Post, which is relatively rare in my experience. Even then, I don't recall wanting to cross-post a reply more than once or twice. This does depend on the fact that I sort each list into a different folder on List-ID or List-Post, so folder context means my expectations are correct. Not sure how this would play out in my target audience.

You don't know about the private conversations :-)

No, of course not. I do know that you've been active on the thread, and if you've been sending private replies, you had to take a different action to do that from the one you use to reply on-thread. So it's not obvious *why* that couldn't be Reply-Author vs Smart-Reply instead of Reply-Author vs. Reply-All until we start breaking down the various use cases.

You also suppose that this style of mailing list dominates my mailing list usage... it doesn't.

I don't suppose any such thing. I simply don't know about it, which is why I ask. I'm not sure your use cases are relevant to my target audience, but they could be. (Eg, I hadn't thought about the Boy Scouts type of case, although I'm not sure it's really different from my advisees case.)

One might say that different behaviors are appropriate for different fora, and that wouldn't be totally wrong, but remembering that different fora will behave differently requires effort, and since Reply/Reply-All do the right thing in *every* fora, why would I want to spend that effort (and take the risk of mixing it up)?

Well, I did it because I'm (intermittently) on a crusade to eliminate Reply-To munging. (Just so you know there is *some* method to this madness.) I realize that's a very specialized motivation. ;-)

I also disagree that Reply-All does the right thing in the subscribe- to-post discussion lists I participate in. Sure, I can go back and edit out all but the person I'm replying to, but even I don't always do that, and most people *never* do. YMMV, of course.

I do 99%+ of my e-mail with T-bird on a Windows system, but there's still that <1% that's done with the Mail app on my iPad, which is the opposite end of the flexibility spectrum.  That's the end that concerns me.

I don't see them adopting this in lieu of Reply-Author, though I could be wrong.

 > And even T-bird is not immune to the "remove features to simplify > things" disease.

To be frank, I was a little shocked by the T-bird story. "What *were* they thinking?!"

Steve

On 01/28/2018 09:40 PM, Stephen J. Turnbull wrote:

OK. But I'm not saying "always." I'm saying that this would DTRT for me a very large proportion of the time, and for AOLers, about 100% of the time to 6 sigmas.

I think that's a question of corpus. DTRT for you is different from DTRT for me which also likely differs from other subscribers to this list.

Others have used the word "right". From the point of view of the Internet, there's no "right" off the Internet, and MUAs are off the Internet.

Just because an MUA isn't on the Internet, does not mean that it shouldn't play by the same or very similar rules. Further, where an MUA is run, be it a fat local client like Thunderbird, or a think web client like Gmail, shouldn't change what the MUA does.

The question is desired behavior, and whether that desired behavior can be achieved efficiently (little information to remember, few keystrokes, etc) and mnemonically for a given set of users who desire that behavior.

Agreed.

In the end it's an empirical question. Unfortunately it's hard to get information about the target population (it's not Mutt users!) without getting the algorithm into one of the big MUAs.

I would go so far as to say that this is likely something that should be a user definable configuration. Which means that MUAs should understand multiple operations and let the end user decide what they want to do.

Mutt and Gnus have had that for as long as I can remember. But there's always a huge constituency for a one-button do-what-I-mean function. "It's obvious what I want, why doesn't this stupid software get it?" I think this algorithm provides that function.

The more we discuss this and the longer that this thread goes on, makes me think that this should be a user configurable action that the MUA prompts the user for what they want to reply to in the ambiguous case. Likely with some tuning and parameters to reduce the number of pop ups.

Where is List-Post a conformance issue? You add it if you want to inform people and MUAs where to post, and you don't if you don't.

I don't think me adding the List-Post header to a message going into a mailing list will work out very well. - I expect that the MLM would munge it (if configured to add the List-Post header itself) or remove it.

I'm saying we can exploit a high correlation between "availability" of posting to the list (the RFC semantics of List-Post) and a desire to direct discussion (ie, replies) to the list.

I think that it would be nice to express such a desire. However I don't think the List-Post header is for that purpose.

It has been a user education issue for 40 years in my experience, though. At some point we need to accept that users are ineducable.

Agreed.

I still believe that user are the root cause of much angst.

I'm suggesting that there should be four functions (reply to author, reply to list, reply to all, and "smart" reply). I suspect that for a lot of users, "smart" reply will be all they ever use.

Fair.

There are a number of people eating Tide pods too. I can't help them and I'm getting tired of Darwin taking too long to help them.

Taken seriously that would mean you believe that UI/UX design is impossible. You actually deny you believe that, and I can't go down that road. Most users are not willing to design their own UI.

I think you misunderstood what I was saying.

I was saying that I think it's wrong for us to make assumptions about what other people do, and to further turn those assumptions into belief that they will do what we think.

UI/UX design can help with some, if not many, things. But the users have to have a fundamental understanding of what they are doing.

Without said fundamental understanding, the very best UI / UX will still fail.

Users may not be willing to design their own UI, but many do choose the UI that they use. Thus, there is choice involved.

If I thought this was just me, I wouldn't have posted. I've been observing the concerns of mailing list owners for two decades, and I believe that if this algorithm were used in all major MUAs, there would be no demand for Reply-To munging.

Maybe, maybe not.

-- Grant. . . . unix || die

On 01/30/2018 03:04 PM, Jordan Brown wrote:

Even getting agreement on what constitutes an ambiguous case might be tough.

Agreement between people may be problematic.

I think it will be quite simple to get people to define what they like and dislike. Which will likely differ from what other people say.

It is absolutely, 100%, clear to me what I want to happen on Reply and Reply All. But it seems that that is not what you want to happen...

We are all entitled to our own opinions. ;-)

-- Grant. . . . unix || die

Grant Taylor via Mailman-Users writes:

Just because an MUA isn't on the Internet, does not mean that it shouldn't play by the same or very similar rules.

If it doesn't DWIM, I don't use it. But that's not the same as talking about conformance of clients. The question is whether the MUA *as used by a particular user* usually produces messages that when interpreted according to the RFCs do what that user expects. For example, you could label the reply-author function "Launch Missiles" and the reply-all function "Buy anti-Trump ad on Breitbart", which hardly matches the idea of "conformant MUA", but once muscle memory kicks in it would produce conformant messages that DTRT. :-)

I would go so far as to say that this is likely something that should be a user definable configuration. Which means that MUAs should understand multiple operations and let the end user decide what they want to do.

Sure, WFM. But in practice the problematic users (remember, this is inspired as an anti-Reply-To-munging proposal!) do not make decisions and reconfigure their software; they bitch and moan and expect everybody else to change.

...

Where is List-Post a conformance issue? You add it if you want to inform people and MUAs where to post, and you don't if you don't.

I don't think me adding the List-Post header to a message going into a mailing list will work out very well.

That would be non-conformant to the RFC. List-* headers should only be added by list managers (typically software).

• I expect that the MLM would munge it (if configured to add the List-Post header itself)

That's what a conformant MLM would do, yes.

or remove it.

That doesn't work for me. I have a couple of cases where I have an umbrella list with dependent lists that are convenient for membership management (students who get moved by class to alumni lists, see reply to Jordan for details). In those cases I want the umbrella list's List-Post passed through to the dependent lists' distribution.

I still believe that user are the root cause of much angst.

Reply-To munging being the salient case.

I was saying that I think it's wrong for us to make assumptions about what other people do, and to further turn those assumptions into belief that they will do what we think.

I don't agree with that as a categorical statement. Of course we shouldn't just "assume", but rather base it on data of various kinds. I think there are many cases where we can be sure enough what people will do that it's worth betting the default setting on it.

UI/UX design can help with some, if not many, things. But the users have to have a fundamental understanding of what they are doing.

Again, I disgree with your wording, at least. UI/UX design helps if and only if the designers have a fundamental understanding of what the users think they are doing. I think the Thunderbird mistake was to think that what the users think they're doing is what the developers think they should be doing. And that's not the case. We All Hate Reply-To Munging, but some users like it. The idea behind this proposal is that they don't like munging for its own sake, but because it allows them to delegate the decision about where to address a reply to the MUA in a natural way ("natural" according to them).

Users may not be willing to design their own UI, but many do choose the UI that they use. Thus, there is choice involved.

Sure, but in many cases that choice is not informed. They use what's there, without understanding, and if the results aren't what they expect, they want somebody else to fix it. They don't realize that they have alternatives that would work better for them, and they typically are quite unwilling to change to a better UI when they're told about one (unwilling for good reasons as well as bad ones).

...

I believe that if this algorithm were used in all major MUAs, there would be no demand for Reply-To munging.

Maybe, maybe not.

That's not helpful. I've explained why I think this would work out well: users who like Reply-To munging like it because they have a one-button solution to most of their reply-addressing needs. They want replies to personal mail to go to author, and replies to list posts to go to the list. My proposed algorithm does that (if the list doesn't munge but does supply List-Post). It also provides familiar behavior with Reply-To munging, unlike the Thunderbird mod. For this group of users, I believe it could be substituted for the reply-author function, and they would think it "works better" (because it works as *they* expect for non-munging lists). (For you and Jordan, feel free to change that UI gesture back to reply-author.)

Under what circumstances would this fail for those people? Do you have evidence that they actually are rare, and that the demand for Reply-To munging is based on a different psychology? Are you really sure "smart reply" wouldn't work for you in most cases where you aren't confronted with a "which reply function to use" decision every few posts on some list?

On 2018-01-24 02:50, Stephen J. Turnbull wrote:

I'd appreciate if those who have strong opinions on this would take a look at the analysis below and tell me if I'm missing something.

While I don't have a strong opinion, getting two copes of the same message (usual "reply all") behaviour is suboptimal. I've the "please CC me when replying to list" .sig for lists where people do it a lot -- not that I've been on many lately.

The BBS/webforum convention is "private message" that may or may not come with "PM sent" note in the discussion thread. I think that's the reasonable way of doing it and that a MUA, when replying to a ML post, should give the user 2 options: "reply to list" and "reply off-list".

Since the list should rewrite the From header for DKIMARC, these would correspond to list-post/from and reply-to resp.

Dima

On 01/22/2018 11:20 AM, Grant Taylor via Mailman-Users wrote:

So, I think Thunderbird's new default is going to cause messages to go back to the author, ignoring the Reply-To.

That's correct.

I can see how this could be annoying as a message author who wants messages to be directed to the mailing list, particularly if I set the Reply-To to be the mailing list.  *sigh*

The T'bird developers view is that in these cases, you are offered a "Reply List" button and therefore, if you use "Reply" instead of "Reply List" you must want the reply to go somewhere other than the list. This of course ignores all those people who will use "Reply" or control-R out of long habit and not because they want something other than "Reply List".

The real problem is the default for mail.override_list_reply_to is True and it should be False, but at least in T'bird 52.x we have an option which we didn't have when the new behavior was first introduced in T'bird 50.0.

-- Mark Sapiro <mark@msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan

On 01/22/2018 12:10 PM, Grant Taylor via Mailman-Users wrote:

Wow.  How could the Thunderbird developers even fathom to introduce an RFC compliant dictated behavior /without/ giving an option to revert.

There is a long history behind this and I agree that T'bird has not always made good decisions on this, but if you want a better understanding and have a few hours to kill, start with <https://bugzilla.mozilla.org/show_bug.cgi?id=77304> (created 17 years ago). Read the entire comment thread there and then those in <https://bugzilla.mozilla.org/show_bug.cgi?id=1309486> and in <https://bugzilla.mozilla.org/show_bug.cgi?id=1392371>.

It's like a game of "telephone" to see what the original request morphed into.

-- Mark Sapiro <mark@msapiro.net> The highway is for gamblers, San Francisco Bay Area, California better use your sense - B. Dylan

On 01/22/2018 03:55 PM, Steven D'Aprano wrote:

Its worse than that: what about people who intentionally set the Reply To header on non-mailing list emails?

I believe the new behavior is only triggered when the Reply-To: and List-Post: headers match.

I guess that might be a problem if the mailing list manager alters the Reply-To: header. But I think that would be the case despite of Thunderbird's recent change.

-- Grant. . . . unix || die

On 22/01/18 18:24, Grant Taylor via Mailman-Users wrote:

Will someone please enlighten me on how mail.override_list_reply_to behaves when set to true (the default) vs false?

If set to TRUE (the default value) replies will go directly to the sender of the message.

If set to FALSE, replies will go to the mailing list. That's my experience anyway (I use Thunderbird 52.5.2 Mac).

Hal