I think the comment made about "open source" winning... was not about the users choosing open source versus closed systems. Most typical users don't recognize these differences and, to some degree, don't care.
However, "open source" does win the hearts and minds and R&D machines of manufacturers so, as a result of these varying options, it ultimately helps open source systems to win over closed systems.
Think Microsoft SQL, for instance. It's powerful. It's very capable. However, it costs money. No person in the world is going to avoid a website that runs MSSQL or is going to only visit websites that run mySQL. If anything, most people don't recognize these differences. However, the website OWNERS certainly see the difference, when it comes to cost... ability to customize... etc, etc... So, ultimately, these cheaper more customizable features win over developers and manufacturers. This, in turn, results in more consumers using these services by proxy.
So, with phones, not everyone wants to buy the same piece of hardware everyone else is using. Not everyone wants to drive the exact same car as everyone else. Not everyone wants to wear exactly the same clothes as everyone else. So, no matter how "perfect" a particular car is... or a particular set of clothing is... or a piece of hardware is... no single car, set of clothes, or piece of hardware is going to dominate a market 100%.
Given this, an operating system that can run on multiple diverse pieces of hardware is simply going to win in numbers.
If Apple allowed iOS to run on other pieces of hardware with little or no fees, it would likely win a much much larger share than it has. It may even dominate. However, Apple isn't in the business to help other hardware manufacturers... they're in it to help themselves. Google, on the other hand, doesn't make the hardware... so they're in the business to help hardware manufacturers. This helps all other hardware manufacturers trust and follow Google's lead, except for those who are still sticking to their own proprietary OSes.
This general fact helps cement Android as the most dominant OS in smart phones for now and the future. The same will eventually be the case for tablet computing devices... televisions... car computer systems... DVRs... etc, etc, etc... It may take 5 years or 10 years or 15 years for this to be eventually realized, but it will happen.
The only thing that could change this course is a new open source OS that can run on diverse hardware for little or no cost which innovates faster than Android.
If that came to fruition, we'd all (as consumers) still win.