Future Planning: Thunderbird as a Web App
kent at caspia.com
Fri Sep 18 15:52:04 UTC 2015
On 9/18/2015 1:46 AM, Volker Birk wrote:
> 2) Moving Thunderbird to a web app will shut down Thunderbird project
> a) There is already gaia. I cannot see why a second web app will help
> here. It would be way more clever to port gaia to Firefox than to
> make Thunderbird a web app.
> b) Web apps may be an idea for mobile, but on the desktop for the user
> an email web app is indistinguishable from a web mailer. This way
> Thunderbird users will probably be converted to web mailer users, and
> Thunderbird will fade out until it's gone.
Mark Banner suggested that "web app" was probably not the best term to
describe the proposed direction, because of confusion of what that means.
I've often tried to describe to friends why Thunderbird is not "the web"
in Mozilla speak. "The Web" as I understand it, and as Mozilla promotes
it, is really a programming stack based on JS and HTML.
Thunderbird, when it uses standardized TCP/IP-based protocols for
communication, is not "The Web" in Mozilla-speak, it is a traditional
desktop client. Mozilla-the-organization is currently dedicated to
making such applications obsolete, and replaced by "The Web". Hence the
strategic misfit of Thunderbird within Mozilla.
If I were to program a game, running entirely locally and offline, but
using the JS/HTML programming stack with Firefox as the underlying
operating system, that IS "The Web". If that program is entirely
indistinguishable from its previous C++/native version, then Mozilla can
claim success in its mission.
On mobile, the equivalent battle is to replace the native Objective-C
iOS or Java Android apps with a JS/HTML equivalent. Really the closest
analogy was the dream of Java to be a write-once, run-anywhere
Undoubtedly many at Mozilla will object to this gross simplification of
their Mission. But I have developed this analogy mostly to present as
benign a view as possible of why Mozilla seems to reject Thunderbird.
There could be another perspective, and I think it is the perspective
taken by most Thunderers-as-Mozillians, that the INTERNET is what is
important, and some of the biggest battles in the near-future to the
INTERNET concern communications privacy and siloing, and Thunderbird is
really the strongest product in the Mozilla family to be the vanguard of
that battle. But we are constantly told that Thunderbird is not
strategic, because we are not "The Web".
In my original proposal, I made one comment that this proposal is not
turning Thunderbird into a "cloud-based" app. But there has been some
confusion here, and Mark Banner is correct that we need to use a
different term than "Web App" which is confusing.
So from the perspective of end-to-end encryption, which as I've
mentioned before is something that Thunderbird agrees is important and
needs to promote, I don't think this discussion affects that directly.
The issues more concern performance and practicality of replacing C++
components and XUL elements with their equivalent in JS and HTML. But
that is an issue that the whole of Mozilla is trying to solve, and has
made enormous progress. I'm just saying we should embrace that direction
- which incidentally leads us down a path of being less dependent on
Mozilla when accomplished.
More information about the tb-planning