With the new world of rapid browser releases I am trying to figure out how to define & implement our browser compatibility process.
Previously we would always support the latest major version of a browser plus one or two older major versions depending on the browser. When I say support, I mean we would proactively test against these browsers for each of our releases. We had a wider range or browsers that we had "reactive" support for where we would fix a problem if we received a ticket for it.
However, with the new rapid release schedule and the sanctity of major version numbers being violated we need to come up with a new solution. One of the ideas I have was to manually look at the release notes for each browser and make a decision based on those if this was really a major version release or not. For instance, Firefox 5 wouldn't require official testing as the rendering engine was the same as 4 for everything that mattered. Looks like 6 and 7 will be the same way.
Maybe Firefox 9 will have some major bug fixes for some of the standard CSS3 features that we use and we will need to explicitly test 9 for compatibility.
Example Release Schedule Before Firefox 9 IE 8+ Chrome 10+ Firefox 3.6 Firefox 4+
Example Release Schedule After Firefox 9 IE 8+ Chrome 10+ Firefox 4-8 Firefox 9
This obviously has it's drawbacks and I am wondering how other people are handling it in this new world.
EDIT: Are we finally at a point where we can truly trust that the modern browsers are going to handle standards the same way and we don't need to worry about it other than verifying standards compliance?
Browsers aren't going to cheat and try to fix broken code anymore, etc?