On Sat, Mar 24, 2018 at 10:28 AM, John Levine <john.levine@standcore.com> wrote:
In article <CAF1dMVE_-YME=0QKormaY8QUkr5ebm+zOJbDKOM6DxGFkyvMFQ@mail.gmail.com> you write:
>- There are several places where 'folder' and 'mailbox' seems used
>interchangeably in the doc and could lead to confusion.  Need to pay more
>attention on subsequent edits.

A mailbox can have multiple folders.  I'll see if I can make that clearer.

>- The doc stated that using A-label in mailto: will shorten the whole
>string, which sounds like an encouragement.  I'm concern on that.

I don't feel strongly about it.  I would expect addresses to work
whether the domain is encoded as U-label or A-label

>- I would love to see more write up regarding spam filtering, particular on
>determining whether certain character is valid or not (even calling it out
>of scope for this doc), and to raise awareness that detection algorithm may
>detect downgraded message as suspicious/malformed/etc.

It shouldn't.  Remember that UTF-8 is already valid throughout message
bodies, and propertly downgraded headers are all ASCII.

Addressing Dr. Ajya's mail here too:

I think I'm not on this in previous mail.
The concern is not UTF-8 or ASCII, it's the occurrence such as empty group may had the message marked spam or killed.  The message is in ASCII with correct syntax, but some detectors may consider it bad regardless.  This doc is great to raise awareness to these potential issues to the vendor/admins beyond the mailbox/domain whitelist & blacklist IMHO.

Best,
Joseph
 


R's,
John