These decisions that may seem random or ill thought out to you have data behind them. A surprisingly large number of people would be happy with a back button, a bookmark menu and a way to bookmark a site. Now these might not be the same users that Firefox has but there are a lot of people who only use those functions.
http://blog.mozilla.com/metrics/2010/04/14/menu-item-usage-study-an-update-to-the-initial-analysis/
Re:re: #1 Bookmark keywords have been a part of Firefox since pre 1.0 as it was a Netscape feature.
Re: #2 since Firefox 4 address bar search now does a normal Google search again as it did pre 1.5.
This is not about making more money. Mozilla is not driven by that economic model. Mozilla sees the walled gardens of the current crop of smartphone and similar devices operating systems as a threat to personal freedom and the heath of the internet. This project is an exploration partly to see what technical gaps exist that web application needs access to function in a similar manner to a desktop application.
There are already companies and working groups trying to accomplish this. Rather than have the specs written without any exploratory work. Mozilla is proposing to build out B2G to evaluate where those problems and shortfalls are. To me this seems like a better idea than having device manufacturers who want to expose the world. Is the user on 4g, wifi, what is the access point name, what is the user paying per kb, etc.? Or having working groups building out a spec absent of actual web developers.
One trick pony that produces code that other applications use as a foundation. NSS, Spidermonkey, XULrunner, NSPR, Bugzilla...
Funds open source groups developing code for accessibility, teaching web standards, creating open standards...
Variables don't; constants aren't.