(Top Post)
Exactly. Could not have said it better myself.
There is one additional issue here. JMAP (or some hypothetical
xMAP) can either aspire to be a cleaner, more modern, less
LISP-like interface to (more or less, see below) the same
functionality as IMAP _or_ it can aspire to be a better idea,
with a different underlying architecture and/or reference model,
functionality, etc.
Can't have it both ways, at least without moving from
discussions about proxies and overlays and into discussions
about translators that are a lot more like gateways, gateways
that, like most of their predecessors, mostly work and mostly
preserve functionality and semantics.
My confusion about this effort -- and it has been considerable--
is that I've heard different statements claiming goals in both
categories. That isn't acceptable, if one because a WG proposal
for one approach should be evaluated differently from one
proposing the there one. The choice really must be made now
and, whatever it is, should be absolutely crystal-clear in the
charter... in addition to the things Dave describes below, which
should also be spelled out.
john
--On Monday, February 13, 2017 09:30 -0800 Dave Crocker
<dhc(_at_)dcrocker(_dot_)net> wrote:
On 2/13/2017 9:17 AM, Dave Cridland wrote:
I don't think this is true.
...
JMAP, on the other hand, can cope with both gmail style
labels and IMAP-style mailboxes, by stated design.
The constraints/flexibility of an effort depend on the goals
it asserts and the support for those goals. I think it's
dandy to target an access protocol that is substantially
different from the features/style/whatever of IMAP. The
requirements are to be very clear about the
features/style/whatever of the new effort, very clear about
the expected benefits, and very clear about the support for
that.
Given the long history of IMAP, the proposed JMAP effort
should offer clear assertions of the relative differences and
benefits of the new effort, to aid in assessing likely appeal
of the effort.
So, for example, the above comment about JMAP's handling of
different reference models is substantive and appealing.
However the various notes that have been posted about JMAP
leave me thinking that there is less clarity about the effort
than one would like, among JMAP's supporters.
The embodiment of the clarity needs to be the charter text and
I don't think it's made its case well enough. For example, I
don't see a reference to the ability to support different
referencing models.
d/