Right now and for the many months it has been in gestation, Australis has been mostly confined to the UX branch, which we neither support nor build from. Although some platform prerequisites have been implemented in the main browser, and coded around if needed in our case, the browser still looks and acts pretty much the same as it did going all the way back to Firefox 4. But that's about to change: Mozilla is announcing Australis will land probably in the Fx25 timeframe, and if not fully activated then, it will be shortly afterwards.
Let me say this now: I am not impressed with Australis. I think it is an inefficient use of screen real estate, particularly since it wastes space on screens like ours which may not be especially wide with its swooping rounded tabs, and the borders seem abnormally fat. It has some known performance concerns, though in fairness these are being worked on, and the whole thing smacks of "out-Chroming Chrome" because Australis looks an awful lot like a Chrome knockoff. That said, I will still try to support it regardless because we do want to maintain source parity as long as possible, and if it can be made to work technically, it will also maintain our add-on and theme compatibility (and anyway you can bet that someone will create a theme to restore the old tab appearance as soon as this hits the release channel).
Australis is not going to land until after the ESR24 branch, so I'm not going to worry about it now; we're going to deal with it for the unstable releases after that. If Australis cannot be made to work on 10.4/10.5, then we will transplant the old skin over and try to get by with the previous interface for as long as possible, but this will impair add-on compatibility and may make the browser interface buggy. I am, however, cautiously optimistic that based on the underlying system work I've seen land thus far, for better or worse Australis can be ported to 10.4. I think that supporting it is the best long term approach from a technical and maintenance perspective, even though I, for one, do not welcome our new rounded tab overlords.
Other stuff. A while back I mentioned that 43% of Mozilla's user base is on 10.6, which is throwing a wrench in their usual OS deprecation plans. Security company Intego, in their article discussing the end of Camino (and a shout out to us), also points out that by their numbers 35.1% of Mac users in general still use Snow Leopard, more than any other version of OS X. On the eve of 10.9, presumably to be announced at this year's WWDC, Apple may have its own "Windows XP" in the making.
The IonMonkey work continues; BaselineCompiler is about 50% done, and I am filling in the gaps in the trampoline and the new assembler as I discover them. I am still not able to test anything, but at least it still compiles.