Social networking sites/apps are, by definition, completely different beasts. They're shallow (I don't necessarily mean this in a negative sense BTW), they don't get out of your way, and users generally don't want them to. That's why sites like Facebook don't just have a list of friends, but are crammed full of games (many of which are "social" as well), videos, pictures, ads, polls, questions, suggestions, etc.
Building a system like this is diametrically opposed to what Google has been doing. So it would make sense that they aren't good at it... it's just not what they do. But I can understand that they might feel they need to produce something like this, otherwise they're going to miss out on the Next Big Thing. If people are spending all of their time on Facebook or wherever, and Facebook decides to add a decent search engine, well... =/
The very fact that you can go back in history and replay, have multiple people talk together and the way that it's built up is not what I would consider cutting out the noise.
But that might just be me.
A busy wave is noisy. But I think it can be quite an improvement over plain email.
For instance their spam filter in gmail is almost flawless at least from my experience.
Anything that's about algorithms being part of the design they are great.
But they fail completely when it comes to allowing people to interact with each other. They simply don't have the design principles for that. It's not part of their DNA.
What Google doesn't do successfully, is frame those interactions to make their own network "sticky". It's those second-order interactions ("likes", "reblogging", "fans", "mayor", etc) that Google doesn't seem to get traction with.
But that's a very different thing than enabling interaction.
But the actual interaction part could be improved quite a lot.
Real people hated it because it was distractive, annoying and embarassing but they never tought about that because it showed their big technical achievement: the wavelet syncronization in real-time. REAL-TIME!!