Mozilla, for the love of god, stop breaking APIs, you morons.
That's actually the entire point of this move. The problem is that the current addon "API", such as it is, is literally every class in the entire freaking browser, which is an untenably huge and perpetually changing surface to maintain. The only way to keep the current API and stop breaking stuff constantly is to freeze all development on Firefox now and forever.
That's not really a viable approach.
The alternative is to come up with a more stable API surface, from the ground up, and provide a transition period for add-on developers to move from the large, unsupportable infrastructure to the stable one that won't be -- as you correctly observe -- constantly breaking.
Rather than developing a new API, the add-ons team decided to leverage the work that Chrome has already done in this space, which has the nice side effect of making life much easier for developers who want to write cross-browser add-ons.
One of the things that's getting lost in the noise here is that the portion of the API based on Chrome's current design is just the start. There will be additional API surface to enable some of the things that had been possible with the legacy wild-west-style Add-On approach. Since reading articles is not particularly trendy, I'll quote the relevant passage here:
A major challenge we face is that many Firefox add-ons cannot possibly be built using either WebExtensions or the SDK as they currently exist. Over the coming year, we will seek feedback from the development community, and will continue to develop and extend the WebExtension API to support as much of the functionality needed by the most popular Firefox extensions as possible.