It's true that in this moment it is hard to come up with practical benefits of semantic tags/structured data for end-users. Still, the structured graph they are part of holds a great potential and the Web of data is evolving fast.. I guess the time will show.
In general, I recommend not spending a huge amount of effort (equivalent to a big bet) without taking some effort to vet the concept in a low-cost way.
I understand your point. For now, I compromise between 'why' and 'how'. Explaining 'the why' of semantic tags (and the Semantic web in general) is still a great challenge, but I feel that moment is very close. Thanks for sharing your thoughts :)
