[ih] mailing list magic, was FTP RIP

[Jack Haverty]

That "very smart/clever" behavior explains a lot more. 

It reveals why sometimes when you send a message to a list, and to
yourself either intentionally or as a result of the way your "reply"
behaves, you mistakenly conclude that the message never made it to the
list, since it never made it back to you (from the list).  So you send
it again.  And again.  And then ask if the list is working.

My mail receiving program (Thunderbird, FYI) is set up to automatically
categorize all incoming messages that arrive via a mailing list into a
separate folder, which then serves as a long-term archive.  But of
course it won't archive messages it never receives because of something
"clever" being done out there in the wilds of the Internet.   That's
probably why my own archive of this mailing list has few messages that I
sent.

This kind of clever behavior makes it much more difficult to figure out
what's wrong when something goes wrong, or even that it is wrong.   Mail
today often seems to get "lost", and it's not easy to figure out why
with all of the "cleverness" in the system.

I wouldn't use the term "standard" with such hacks.  They may look
clever, but there are unintended consequences.

Back in the 70s when we were experimenting with mailing lists and office
automation, I recall discussions about the need to have mechanisms
enabling mail servers/clients to provide such functionality in a
standard and predictable way.   For example, perhaps a mechanism and
protocols for managing lists, which would include such things as a way
for one server to ask another about the contents of a mailing list.  Or
a network service that would reliably archive important material (the
DataComputer then, now a NAS).   Or a way to reliably identify
addressees.   And ways to balance security and privacy.

There were a lot of questions and technical challenges of course.  That
effort never got very far, and was placed on the back burner so that the
interim "simple" email mechanism could be put in place.

That was about 45 years ago.  There's been a lot of technical progress
since then, e.g., in "distributed databases". 

Perhaps it's time for a rework of the "simple" mail architecture.   One
for your list, John Gilmore....

/Jack Haverty



On 10/2/20 10:24 AM, John Levine via Internet-history wrote:
> In article <CAEf-zri_tsxYViqsx0kxk=8g+qTtEHgCHDZ3yJ=W2ATSKe27Lw at mail.gmail.com> you write:
>> as it seems that the IH mailing list SW thinger at ISOC is Very
>> Smart/Clever in this regard and seems to "know" (looking and seeing of To:
>> geoff at iconia.com in the header).
>>
>> ERGO, yours truly only got ONE copy of your reply (directly from you and
>> none from/via the list), viz.:
> Yeah, that's a standard mailman feature.
>
> I find it annoying and turn it off because my list mail and my direct
> mail go to different places.
>
> R's,
> John