Owners

Jonas Sicking

Status

• Too early for a bug.
• Spec available at [1]
• Start implementing in 2008 and landing initial implementation in 2009.

Overview and Motivations

• More reliable XBL implementation. I.e. following an available spec rather than "don't change anything, we don't know what people depend on"
• Better security model for XBL.
• XBL form controls on webpages.
• Built on a w3c standard so hopefully will get implemented in other browsers.
• Simpler DOM model (separate DOMs rather than 'anonymous children') which should simplify our content implementation and the code that depends on it.

Risks and Assumptions

Our chrome is heavily dependent on the current XBL1 implementation. Ideally we would like to drop this implementation and let the XBL2 implementation replace it, but that would likely require a complete rewrite of our chrome.

Two possible alternatives:

• Keep the XBL1 implementation working well enough that the two can live together.
• Keep the XBL1 parser, but make it build XBL2 bindings.