[kde-community] [Kde-pim] A new home for Mozilla Thunderbird at KDE?

[Jos van den Oever]

On Wednesday 27 April 2016 21:42:12 Eike Hein wrote:
> On 04/27/2016 06:36 PM, Daniel Vrátil wrote:
> > I like the idea of having Thunderbird in KDE. It shows that we are an open
> > community and welcoming towards "outside" projects and of course it would
> > be also a good PR for both sides.
> 
> No, it wouldn't. The message wouldn't be "KDE community is open to the
> outside", it would be "KDE offers shelter to legacy project, hoping to
> salvage some attention from it".
> 
> Make no mistake, Thunderbird is a dead project. It's built on a toolkit
> that's EOL, and hardly has enough of a development community to sustain
> the app, much less the stack beneath it. That it has users (like me)
> that still use it despite the mounting bitrot and deteriorating
> performance doesn't change that outlook. Many people who use Thunderbird
> want to switch away from Thunderbird.
> 
> KDEPIM does face some similar challenges, but is actually much further
> along on componentizing its codebase to where e.g. moving from QWidget
> tovother toolkits is feasible, and QtCore is far from dead. As a
> developer, if I wanted to work on email stuff, I'd rather go there than
> invest my hours into Thunderbird. And that's part of the problem, too.
> 
> If we were to incubate Thunderbird, it would need to supply really
> really strong answers for how it's going to pull its own weight to
> offset the resource and PR cost.

Years ago, LibreOffice split off from OpenOffice. Apache OpenOffice is now barely 
alive. They hardly manage to release security fixes. And yet, still more people 
know about OpenOffice than about LibreOffice. Most of these people are on Windows. 
LibreOffice is working hard to change this but it takes very long.

Thunderbird is a very familiar program to many. It is a strong brand. If 
Thunderbird deteriorates, it will leave many to give in and go to webmail 
hosted by an advertising company. That way the number of people using real 
mail clients might be halved.

If the Thunderbird team were to decide to update their codebase and perhaps 
move to use Qt components, they might retain their userbase. Subsurface and 
Gcompris went this way too, to technical success. Any such decisions should be 
made by the Thunderbird developers and there are quite a few of those.

Looking at the commit logs of Thunderbird, the programs certainly does not 
seem dead at all. Last month there were on average two commits per day by 18 
authors. [1] Sure they might have technical debt, but so did OpenOffice. Moving 
away from the link to the Firefox release schedule, might even give breathing 
room for more fundamental work.

Cheers,
Jos

[1] https://github.com/mozilla/releases-comm-central/pulse/monthly
     This is a git mirror of the mercurial repository of Thunderbird