Future Planning: Thunderbird as a Web App

Eric Moore emoore at fastmail.fm
Fri Sep 18 03:23:06 UTC 2015


> From: Kent James<kent at caspia.com>

> tl;dr Thunderbird over the next 3 years needs to convert to being a web
> app that can run on any browser that supports ES6 Javascript and HTML5.
> (web app does not imply cloud-based, only that the underlying platform
> is js/html).

I understand the need for Thunderbird to become a web app. However, its 
not clear to me why that web app needs to be restricted to just 
Javascript and HTML5. That seems to be a serious handicap, and likely to 
cause a lot of problems with widgets. Think of the classic arguments why 
somebody might prefer Thunderbird over a good webmail implementation 
that supports accounts with multiple email providers. It seems to boil 
down to more customization, expandability (via add-ons), better widgets 
and being able to support a platform specific look and feel. I'm 
concerned we'd lose most of that.

Mailpile uses Python, JS, HTML5 and will let you "Host your install of 
mailpile on your laptop, desktop, Raspberry PI, server in the cloud, or 
put it on a USB stick and carry it in your pocket." They plan on 
eventually supporting apps for Android and iPhone/iPad. "The Plugin 
architecture has being expanded to allow self-contained plugins which 
can contain a mixture of python, HTML (jinja2) and javascript code, but 
this work is incomplete."

Please explain the tradeoffs between MailPile's definition of a web app 
and one that only supports js/html. I'm not trying to argue that we 
should mimic MailPile, they're just a convenient example of a different 
approach.

One benefit of limiting Thunderbird to just js/html appears to be that 
we could share "maintenance work of "low-level" protocol code with Gaia 
email or other components". However, that is a two edged sword as in 
practice it could also mean that we'd have to constantly adapt to 
changes in the Mozilla platform.

Another would be that Thunderbird would be available on many more 
platforms, including ones that we don't even know about. I agree that we 
need Mobile support, but there are diminishing returns at some point in 
trying to support everything.

How does supporting only js/HTML effect other possible long term goals 
such as end-to-end encryption?





More information about the tb-planning mailing list