Most likely uses would be remote installations for sensor platforms and such. It's not a huge amount of power for human usage but you could run low power electronics for all sorts of things. Or it could slowly and constantly charge more power hungry equipment if it didn't need to be used constantly. Aerial drones that could remain charging most of the time and had multiple copies with similar power stations across a wide area could cover remote territory, performing tasks like search and rescue and even relaying small payloads like emergency provisions to people in the wilderness until rescuers can get there.
It is important to remember that a "useful" amount of power can be different depending on who or what is using it and how practical other forms of power are in the context.

250 watts for the price of a compact car sounds like a rip off to me.

The price of a compact car yet will be completely taken out by the first tree branch that is on that river...That's an expensive risk.

Most rivers/streams around here are not pristine but loaded with twigs, and other floating and suspended debris. These would easily jam the impeller. Even an upstream screen would clog pretty quickly. What about feasibility in colder climates, where icing would be an issue?

No worries about fish? No screen-how will it stop debris from clogging it? Way too expensive right now. Good idea though-work out the kinks. Adding a jacuzzi type shell might propel the water even faster for more power. Maybe horizontal blades that flip would hurt fish less.

Talk about reinventing the wheel! Water power was optimised in the 19th century with overshot water wheels reaching >90% efficiency. They dont clog, are almost fit&forget reliability and are simple and cheap to make. Hi-power fixed magnet generators which operate at low RPM that can turn this motive power into electricity.
This little thing is way too expensive, will clog easily and because it has gears turning 90 degrees in it will wear out.
Another solution to a problem which does not exist.

I like the design and could see these powering a house. Unlike a high pressure design, it could be used on a property in series. In other words you could put a bunch of them on the same property. You would just have to put enough space between them to allow the water to gain speed again. Pricing seems to be a problem in this article. They should have put out a number, the whole "compact car" thing lets a very large number appear in my head..... The assembly holding the turbine in the video is a bad design. They should have dock floats attached so the turbine will stay at the same height as the river as it rides and falls. With feet so it cannot bottom out. The whole thing tied off to post besides the river with cable.

overengineered and overpriced is no way to advance remote energy generation ...
if this is the best the green energy movement can come up with then we need a new green energy movement ...
these folks offer a 1300 watt unit for $1475 ...
http://www.aurorapower.net/products/categoryid/4/list/1/level/a/productid/209.aspx?gclid=CNnawqjWl7QCFe1xOgodzBIATA
and "speeding" up the water flow doesn't "add" power ... the energy contained in flowing water can't be "increased" it can only be extracted ...

Agree with above comments re debris and price- a 'small car' price is vastly excessive for something that is technologically simpler than the average modern turbocharger.
In the UK the best small hydro plants seem to be the ones that use weirs, with fish bypass, and which take the form of a worm drive. As the water isn't guided into a casing there is far less potential for debris to get stuck in it.

It would not be too difficult to keep debris out by building a screen several meters upstream but the only advantage it provides over a undershot waterwheel is that it is easier to conceal.