Future Planning: Thunderbird as a Web App

Kent James kent at caspia.com
Fri Sep 18 18:41:26 UTC 2015

On 9/18/2015 1:03 AM, Mark Banner wrote:
> How about calling it something like "Thunderbird post-XUL" or 
> "Thunderbird on web tech"? The issue with "Web App" is that this is 
> going to pre-load people's thoughts about what you mean with something 
> completely different.

Right - as responses to this thread have already established. I'll avoid 
"Web App" in the future!

> I would recommend talking to people from the Android/iOS teams. For 
> example, they switched a lot of the Android front-end implementation 
> from XUL to native android a few years back because of performance 
> issues. Obviously all the backend is still the existing gecko & I 
> believe some js base, but just switching isn't necessarily going to 
> give you this for free.

It seems to me that we are operating at a level above the Fennec user 
interface. But we'll need all of the advice we can get.

> Going with the mobile theme for a bit, I think consideration needs to 
> be given about what a mobile Thunderbird would look like. Something 
> that's as fully featured as desktop is likely to be very difficult to 
> do a good UX for and is likely to be a lot slower.

Agreed that a phone app is a bit of a stretch, and likely would require 
some radical rethinking. But an Android tablet is quite feasible, and I 
think should be a goal. Maybe I'm just more tablet focused than most, 
but I spend hours per day on my tablets.

> Hence if you're looking at rewriting a lot of the back-end, then it 
> would be worth giving a thought to how it would operate/be constructed 
> in a mobile world where you don't necessarily want all the features. 
> For example, lets assume filters aren't required on mobile (I don't 
> know of a mobile app that'd do that) - it be good to have that part of 
> the back-end "plug-in" when required, potentially with hooks like 
> today's add-ons can use, rather than for it to be built-in permanently 
> even though there's no filters included.

I'm all in favor of addons as a way to manage complexity. It's great for 
me to hear that Hello wants to ship as an addon, as to me that means 
that Firefox will continue to think carefully of issues associated with 
using complex addons. But it is a good time to remind ourselves of that 
strategy. There are many pieces of Thunderbird that I would prefer to 
break out as shipped addons (such as junk mail processing, or LDAP, or 
account migration) only because they are major pieces of code only used 
by a minority that can complicate the UI and code for the majority.


More information about the tb-planning mailing list