Re: [Imap-protocol] Registration of keyword which enables fetching external images

On 5/20/2015 2:29 PM, Brandon Long wrote:
>
>
> On Wed, May 20, 2015 at 12:07 PM, Andrew Sutherland <[email protected] 
> <mailto:[email protected]>> wrote:
>
>     On Wed, May 20, 2015 at 8:47 AM, Jan Kundrát <[email protected]
>     <mailto:[email protected]>> wrote:
>
>         Are some other client developers interested in this?
>
>
>     It seems reasonable to me to standardize a flag for this purpose,
>     even if many clients prefer to keep the setting local-only. 
>     Thunderbird uses a per-message "remoteContentPolicy" property that
>     is persisted locally in the msf files but never persisted to IMAP
>     servers.  Once set, it allows external images for the specific
>     messages to be fetched going forward.  (Some information leakage
>     will be suppressed on repeated viewings because of the network
>     cache, but the images/etc. are likely to be rotated out.)
>
>     Note: I no longer actively contribute to Thunderbird, so if anyone
>     is going to change things to use the flag it won't be me.  I work
>     on the Firefox OS email app, and we currently have no plans to
>     persist "show extenal images" on a per message basis, but if we
>     did, we would use the proposed flag since we would synchronize the
>     flag to the server when possible.
>
>
> I think we persist per sender, not per message... and we know proxy 
> most of that stuff, so I think a lot of users just show all 
> messages... or we have some complicated logic about when it's ok to 
> show and when it's not which is calculated on the fly and not exposed.

Thunderbird also allows you to persist a show-remote-images per-sender 
as well (the UI says something like "Show content for this message" and 
"Remember for this sender" as options).

If an IMAP keyword were standardized, I'd probably recommend we'd use 
it, but I wouldn't prioritize implementing it. I believe that when we 
use the per-sender option, we don't bother to set the flag on the 
message itself, so I'm not sure that a keyword would be totally 
reflective of the current state in Thunderbird.

Ultimately, I'm convinced of neither the utility nor the inutility of 
this feature. The biggest benefit appears to be "synchronization of 
settings," but it's not exactly a sufficient synchronization due to the 
existence and use of coarser-grained (e.g., per-sender) policies, so I'm 
hesitant to support such a standard. On the other hand, I don't really 
have a coherent reason to object to the standard.

-- 
Beware of bugs in the above code; I have only proved it correct, not tried it. -- Donald E. Knuth

_______________________________________________
Imap-protocol mailing list
[email protected]
http://mailman13.u.washington.edu/mailman/listinfo/imap-protocol