Invitation for technical discussion on next-generation Thunderbird (Semantic Desktop)

Ben Bucksch ben.bucksch at
Fri Apr 28 14:09:48 UTC 2017

Gervase Markham wrote on 28.04.2017 15:49:
> I would certainly be interested in the comments of those thinking about
> TB++ as to why it would not be a quicker route to our goal to
> collaborate on or build on top of Nylas. Is it just our pride that makes
> us want to build a new Thunderbird from the ground up?

Personally, if I was to embark this endeavor, I'd certainly look at 
Nylas N1/mail and Gaia/GELAM. That would be one of the first steps of 
the project, after the initial prototype phase.

For me, the decisive question, code quality.

* How easy is it to work with this code, to adapt, massage and expand 
it? Is it a good base to start?

* How much does it help us? How much is smart code with good solutions 
and good APIs? How much is trivial to replicate, in a much better way?

TB:NG should be a solid base for a large feature set (replicating 
current TB Gecko) and, more importantly, for the next 20 years, so I'd 
want to ensure that the foundation is sound.

If that's the case, I'd be in favor of using Nylas N1 as base.


Also, Nylas N1 is GPL3. GPL3 is incompatible even to GPL2, and maybe to 
MPL2 (you probably know better), so that's at least a question to 
consider. But as you said yourself, we should *not* get into a license 
discussion rathole now. I'd first evaluate the code base, then if I 
determine that it's a great base and there's a large amount of good code 
that we could use and that would take years to replicate, then we can 
talk about whether the license is acceptable. I don't want to discuss it 
now. I just want to point out that the problem exists.


More information about the tb-planning mailing list