I'll just play devils advocate here. As a consumer, I really like buying from the Mac App store solely for the user experience of, when I eventually new a new Macbook Pro, I can easily sync all of my purchases, rather than having to retrieve the licenses and go download the latest version of each piece of software I've purchased. As a developer, I've heard countless horror stories and totally understand.
The sad thing is, they didn't even need an App Store to provide this type of service, and it would have been more useful without a store. Apple already has iCloud; they only needed a variation of the current sync mechanism (something a bit smarter that knows how to find stuff in Applications and related files like Library->Preferences). If they're comfortable giving away free gigabytes of storage to people, they shouldn't care about uploading all of a user's apps — from a store or not — and syncing them anywhere.
As a user/consumer I consider the search or discovery of new apps on the MAS quite horrible, the features you mention are nice but IMHO they come "after you found the app".
IMHO an app dev is better off implementing the payment system and update than using MAS, for ex in my case there is no way I would wait 7 days to have an update out of the door (eg. I don't automate everything to get blocked on the last step).
Brew is not consumer friendly at all. My parents (in their 60's) own a mac that I bought for them. It was a great decision for me as I barely have to support them for technical problems. I only get questions on how to do something. If I was to try and explain brew to them it would be a disaster. Explaining the Mac App Store is far easier.
Brew is not even tech-person friendly. I'm comfortable with the command line and I had a ton of problems getting cask to work. I LOVE the idea and I will use it every time I setup a computer, but there's so many issues to work out. Mainly, some apps not updating correctly, the storage of app versions in cask, settings, licensing. It's a mess.
I really hope it gets better. I'd contribute if I was knowledgeable enough to do so.
Be aware that Homebrew Cask [1], which your comment is referring to, is a separate project being built on top of Homebrew ("brew") [2] and it currently has a number of shortcomings that are independent of anything to do with Homebrew itself.
