<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<p><br>
</p>
<br>
<div class="moz-cite-prefix">Am 14.02.2018 um 12:39 schrieb Henri
Sivonen:<br>
</div>
<blockquote type="cite"
cite="mid:CAJQvAudPUvuVREjjQE-eqKjOQ7OyhJUic+P=4=VzgAsHB_VPqw@mail.gmail.com">
<div dir="ltr"><br>
<div class="gmail_quote">
<div>I think the ideas of a greenfield rewrite of everything
at once under the Thunderbird brand are a recipe for the
death of the Thunderbird that has the users.</div>
</div>
</div>
</blockquote>
<br>
A superficial look at things can be decieving.<br>
<br>
What I propose is the same that Firefox has done with Servo.<br>
<br>
<blockquote type="cite"
cite="mid:CAJQvAudPUvuVREjjQE-eqKjOQ7OyhJUic+P=4=VzgAsHB_VPqw@mail.gmail.com">
<div dir="ltr">
<div class="gmail_extra">
<div class="gmail_quote">
<div>I think Thunderbird would be better served by realistic
plans for a series of piece-wise rewrites that keep the
product shippable at all times from the repo of record.<br>
</div>
</div>
</div>
</div>
</blockquote>
<br>
The core problem with a "gradual rewrite" is that MVP is impossible.
Additionally, the scope of Thunderbird features is huge, and the
interdependencies are so close and the architecture so arcane.<br>
<br>
We have made several attempts at replacing components of Thunderbird
in the shipping product, e.g. JSMime and JSAccount. All of them with
an extreme amount of effort by some of our most skilled developers,
and none have succeeded in actually replacing the old component. But
they still took several times more efforts (in case of JSAccount,
probably several orders of magnitude more effort) than a clean
rewrite.<br>
<br>
<br>
<blockquote type="cite"
cite="mid:CAJQvAudPUvuVREjjQE-eqKjOQ7OyhJUic+P=4=VzgAsHB_VPqw@mail.gmail.com">
<div dir="ltr">
<div class="gmail_extra">
<div class="gmail_quote">
<div> </div>
<blockquote class="gmail_quote" style="margin:0 0 0
.8ex;border-left:1px #ccc solid;padding-left:1ex">
<div text="#000000" bgcolor="#FFFFFF"> There doesn't seem
to be much competition in the desktop email market, so I
think Thunderbird has had it "easy" - we've gotten away
with the existing performance etc without loosing the
existing core of our users. That might not continue
forever, or it might, be we'll also be forever
struggling with what we want the architecture of
Thunderbird to be. Not a good place to be in.<br>
</div>
</blockquote>
</div>
<br>
</div>
<div class="gmail_extra">That there isn't much competition in
the desktop email app space and practically no competition in
the space of cross-platform desktop apps make it all the more
important to keep Thunderbird going and not let it be killed
by rewriting everything at once or chasing mobile at this
time.<br clear="all">
</div>
</div>
</blockquote>
<br>
<br>
FWIW, I am proposing to keep maintaining Thunderbird as we do, and
keeping it working, until the replacement is ready. The rewrite
would be a parallel project with a different team, like Servo was. I
expect it to take 3 years for 10 developers, to be usable for 80% of
the users. For comparison, Mailpile took 5 man years (less than 2
years for 2.5 developers), with a smaller scope, but still a fair
product.<br>
<br>
Ben<br>
</body>
</html>