Re: [Jmap] Submission

["Chris Newman" <[e-mail]>]

On 21 Apr 2017, at 19:16, Ted Lemon wrote:
> On Apr 21, 2017, at 9:48 PM, Chris Newman <[e-mail]> 
> wrote:
>> My product's existing JMAP-like proxy is a JSON/HTTP to 
>> submission/IMAP proxy. One of the design errors that was made (back 
>> in the 1990s) was to not support submission extensions. This is one 
>> of the largest remaining design errors in our proxy and has caused 
>> significant problems. I do not want the JMAP standard to repeat the 
>> mistake we made.
>
> Can you give an example of a failure mode that resulted from this 
> omission?  (Not doubting you, just curious what the specific problem 
> was).

A rigid proxy design means complexity-increasing changes have to be made 
to three pieces of software instead of just two pieces of software. That 
means the entire system is more complex and less agile. With a proxy 
design that's well-aligned to the submission protocol, most submission 
extensions can be supported with just a configuration change (to pass 
through the new extension). So an extension increases system complexity 
in only two components with the better design.

For our product, we have different developers on each component 
(Submission vs. HTTP proxy vs. Client UI) and some of the developers 
have a significant timezone shift. I leave the problems that causes to 
the imagination of the reader.

		- Chris