Proposal: MailExtensions API to allow UI overlays, but no script injection

Philipp Kewisch kewisch at thunderbird.net
Mon Oct 14 07:14:22 UTC 2019



> On 14. Oct 2019, at 12:28 AM, Eyal Rozenberg <eyalroz at technion.ac.il> wrote:
> 
> 
> 
>> On 13/10/2019 22:48, Philipp Kewisch wrote:
>> If your strategy is that Thunderbird would accept that add-ons could break in any release and it would be up to the developer to stay up to date, I feel this would be a fair amount of work for developers to keep up with all the UI changes, as we have it now.
> 
> This was always the case until now - except that recently the changes are very frequent and more profound. Do you expect continuing profound changes?
> 
> If so, consider slowing down the release rate to allow us to keep up. And if the changes kind of settle down, then it will be tolerable again, like it was until v52.
I expect the changes to increase. Not only will Firefox continue to do large projects like get rid of xul, I expect Thunderbird to work on their own big projects that will affect the internals that Extensions are currently using.

It will eventually be less major, but I'm guessing that will be a few years out.

Doing less releases will just mean that there will be more for extension developers to adapt to.


> 
> 
>> You see in other threads here requests for documenting exactly what changed. If we change an id, some developers will expect us to tell them, bec. So it would be work for the Thunderbird team to communicate and document this.
> 
> Haven't seen requests for special documentation for every id change... more like, when it seems every other thing has changed, we might be grasping in all directions for something to hold on to.
I didn't say people were requesting each id change, but there is a general request for documentation on things that changed, which I think is a good request.

If we only have APIs to deprecate, we can much easier determine which add-ons need to update, and we can be more proactive and reach out to affected developers instead of relying that they read the hg log.

Philipp 


More information about the tb-planning mailing list