The answer is disappointing: the new uses were mostly of a single new category -- the Casino. The entire remainder could be grouped into a mere two additional categories -- Oracles and Names.
The Blockchain has been referred to as "The Trust Machine". However, a Casino involves a different type of trust than a Bank. Casinos
[1] consume disposable income to provide a brief experience, [2] they exist in a competitive environment, and [3] they have custody of a small amount of funds for a small amount of time. A Bank (comparable to Bitcoin) is almost the exact opposite -- [1] stores up funds for further use, [2] has monopoly power over their customer's funds (if you fail to withdraw from a frozen Barclay's account, you cannot "take your withdrawal business elsewhere" to Bank of America), and [3] holds a large amount of funds for a large amount of time.
Money is sent to Casino-types ("value-accepters") in many ways, but it is sent to Bank-types ("value-storers") in very few ways.
In fact, the structure of a blockchain -- everyone downloads and verifies everything -- implies that they will only be efficient where "what other people do" affects you. With Money this is true (as no one wants to end up holding a counterfeit bearer asset), but with Contracts it is false (I couldn't care less about my neighbor's car payments).
This argument is applied *< nasal honk >* to Slock.it, which *literally* does not require a blockchain. Moreover, the features of the blockchain -- censorship resistance, regulatory arbitrage, anonymity, open access, etc. -- contradict the concept of visiting a physical place at a known time.
These conditions favor solutions which are *specific*, and not *general*.
Of course, all views are my own and do not necessarily represent Bloq's or anyone's, and so forth ( http://www.truthcoin.info/#disclaimer ).
Essentially, Sidechains will be second-rate to the blockchain that secures it.