Proposal to start a new implementation of Thunderbird based on web technologies

Paul D. Fernhout pdfernhout at kurtz-fernhout.com
Thu Apr 6 03:51:04 UTC 2017


On 2017-04-04 7:25 AM, Disaster Master wrote:
> On Tue Apr 04 2017 00:02:21 GMT-0400 (Eastern Standard Time), Paul D.
> Fernhout <pdfernhout at kurtz-fernhout.com> wrote:
>> Your point on risk management makes a lot of sense. That's why I
>> suggested a possibility of having a new Thunderbird that could run
>> alongside the existing one and share the same files while offering a
>> different UI.
> 
> First, such a possibility already exists - if you are using IMAP.

Yeah, but I use POP. :-)

> But connecting some brand new client, in beta, alpha, or even pre alpha
> stages of development? This presents a very large risk to your precious
> data, one that I would never be willing to accept, and if anyone did,
> the PR backlash could be devastating when their X years worth of
> precious emails were nuked by some bad code.

You're absolutely right of course.

But, one of the toughest things in a battle is to get soldiers to
abandon their dug-in foxholes to retreat in the face of advancing enemy
fire. The soldiers indeed are safer in the foxholes with all the random
shells flying around -- until the battle line advances to their foxhole
and they are overrun. And the closer the battle line comes, the more
dangerous it is for soldiers to leave their foxholes.

The battle line of entropy for Firefox has been advancing for the past
couple of years as the Firefox browser codebase that Thunderbird depends
on is deprecated and then abandoned. And there is also Mozilla
abandoning financial support of Thunderbird -- it's one area of solid
user growth -- to redirect money to various quixotic activities that
have little to do with either maintaining a popular web browser or
supporting distributed secure communications and web identity management
via email.

At some point, Thunderbird-as-is may fail in a big way given it will be
in practice unmaintainable and so at risk of security issues like a
maliciously-crafted email exploit or perhaps some bad interaction with
an updated OS. Thunderbird also has a known failure mode of 4GB
mailboxes getting corrupted that many long-time Thunderbird users may be
reaching.

So, we're going to have to take some risks sooner or later -- one way or
another. Unfortunately, as time grows smaller and smaller until that day
of failure, the risks grow bigger and bigger because there is less time
to experiment and test. It feels safer to keep using Thunderbird-as-is
and it is safer -- until something bad happens from lack of maintenance.

I've been using Thunderbird almost since it began, migrating from
Mozilla Application Suite. It would be sad to have to abandon Thunderbird.

But I know based on looking at the Thunderbird code and thinking about
the risks that it seems very unlikely Thunderbird-as-it-is will be
maintainable either technically or socially much longer. Dealing with
complex legacy C/C++ code that embeds an abandoned browser in a
convoluted architecture is going to be a high-stress no-fun job. It will
have a high learning curve. Even if some team does support the codebase
for pay, it is going to be hard to do it well for the long haul because
of the complexity of relying on complex embedded browser mixed with the
unforgiving nature of C/C++.

So, we need to do something. It is prudent either to start making new
alternatives or to migrate to existing well-tested proven alternatives.

The biggest thing Thunderbird has going for it is a community of loyal
users. So the best solutions would leverage that somehow -- either by:
* an incremental move such as changing Thunderbird to serve JMAP to web
clients,
* creating a parallel companion to Thunderbird with the risk you mention
(but without the risk of changing much about Thunderbird itself),
* a complete from-scratch-build that gets the Council's blessing
assuming regular Thunderbird users take notice -- or perhaps even
* a collective decision to move to another existing platform (like, say,
Nylas N1 that Ben mentioned) and put energy into enhancing it to have
all of Thunderbird's features, or,
* making something just so compellingly better than Thunderbird that we
all want to switch to it (like a social semantic desktop that just
happens to also handle email well as well as chat, whiteboards,
federated wikis, and so on -- as I tried with Twirlip).

Most likely, Thunderbird users being a large and independent-minded
group (otherwise we'd all be using gmail's web client instead), all
these paths will be tried if/when something big breaks (or perhaps before).

Anyway, with just $1 per Thunderbird user, we could fund the core of
something amazing.

I might forward my job application to the Mozilla Foundation from 2011
to work on Thunderbird to perhaps spark more conversation on bigger
picture alternatives like a semantic desktop.

--Paul Fernhout (pdfernhout.net)
"The biggest challenge of the 21st century is the irony of technologies
of abundance in the hands of those still thinking in terms of scarcity."


More information about the tb-planning mailing list