Re: Annoying Reply-To - Mailing list pgsql-general

From Mikkel Høgh
Subject Re: Annoying Reply-To
Date
Msg-id 74826FB8-0529-486A-A6A5-4C6DE161D3C9@hoegh.org
Whole thread Raw
In response to Re: Annoying Reply-To  (Bill Moran <wmoran@collaborativefusion.com>)
Responses Re: Annoying Reply-To  (Bill Moran <wmoran@collaborativefusion.com>)
List pgsql-general
On 17/10/2008, at 13.20, Bill Moran wrote:

> In response to "Mikkel Høgh" <mikkel@hoegh.org>:
>
>>
>> On 17/10/2008, at 12.24, Tomasz Ostrowski wrote:
>>
>> But again, how often do you want to give a personal reply only? That
>> is a valid use-case, but I'd say amongst the hundreds of mailing-list
>> replies I've written over the years, only two or three were not sent
>> back to the mailing list.
>
> You're forgetting the cost of a mistake in that case.
>
> As it stands, if you hit reply when you meant reply-to, oops, resend.
>
> If it's changed and you hit reply when you want to send a private
> message
> to the poster, you just broadcast your private message to the world.

And again, how often does this happen? How often do people write
really sensitive e-mails based on messages on pgsql-general.

Because if we wanted to be really safe, we should not even send the
mailing-list address along, so even if someone used the reply-all
button, he could not accidentally post his private e-mail on the web.

In true McDonalds-style, we could change the mailing-list-address to
be pgsql-general-if-you-send-to-this-your-private-information-will-be-posted-on-the-internet@postgresql.org

How far are you willing to go to protect people against themselves?

>> Yay, even more manual labour instead of having the computers doing
>> the
>> work for us. What's your next suggestion, go back to pen and paper?
>
> Don't be an asshole.  There's no need for that kind of cynicism.

In my opinion, asking for sane defaults is neither cynicism or being
an asshole.
I may have put it on an edge, but having to manually add a Reply-To
header to each message I send to pgsql-general is not my idea of fun.

>
>
>> Well, my point is that Reply-To: is only dangerous if you're not
>> careful. Not so with the other examples you mention :)
>
> But as it is now, it's not dangerous at all.

No, just annoying and a waste of time, energy, bandwidth and
ultimately, money.

>
>
>> If you're writing something important, private and/or confidential,
>> don't you always check before you send? You'd better, because a small
>> typo when you selected the recipient might mean that you're sending
>> love-letters to your boss or something like that :)
>
> I'd rather know that the computer had my back in the case of an error,
> instead of it helping me mindlessly even when I'm doing the wrong
> thing.
> To me, that's also the difference between MySQL and PostgreSQL.


Well, in the above case, the computer doesn't have your back. If you
told it to send the e-mail to Marty Boss instead of Maggie Blond,
that's exactly what it'll do.

Currently, when I tell my computer to reply to a message on the pgsql
mailing list, it'll do something else, because who ever set it up
decided to cater to the 0.1% edge-case instead of just having the
default action be what it should be 99.5% of the time.

You may not care about usability or user experience, but remember that
what seems to be correct from a technical perpective is not always the
"right" thing to do.
Attachment

pgsql-general by date:

Previous
From: Ivan Sergio Borgonovo
Date:
Subject: searching through a catalogue with tsearch2 (examples, references, HOWTO)
Next
From: Bill Moran
Date:
Subject: Re: Annoying Reply-To