they do seem pretty serious.
I don't think anyone's arguing that they're not serious. They don't appear to be claiming anything impossible, they simply claim to be the first ones to get a practical, small-scale fuel cell array suitable for home use. The question is not whether that's possible, it's whether this particular system is actually as practical and cost-effective as claimed.
Even assuming everything said about it is true, there are still two big problems. Firstly, fossil fuels. There may be efficiency gains, but this still uses methane, and therefore has all the same problems with supply, pollution and so on that any other methane powered generator has. The whole point of fuel cells is that they can run on hydrogen, which can be produced in any number of sustainable ways. Making them run on fossil fuels may be easier in the short term, but completely defeats the object of having the in the first place. The really sad thing is that they seem to be using the fact that it runs on fossil fuel as one of the major selling points, rather than a serious hurdle that will have to be overcome in the future.
Secondly, the grid. The national grid, or whatever your country's equivalent, is not something you can simply plug stuff into at will. There's a lot (
really a lot) of technical management that has to go on to keep it all working at the right current, voltage, frequency, phase, and so on. This is not too hard when you have relatively few centralised power stations, but the more spread out generation is, the more difficult it becomes. One of the big problems with things like wind power is that most grids won't actually be able to cope with a large amount without significant upgrades. It's something that's not often mentioned in the PR, but the UK's plans to have 20% of our power generated by wind power are simply not possible without some major expenditure getting the grid to cope with it, on top of the cost of actually building the wind farms.
And wind turbines are pretty big things in the MW range, built in carefully chosen locations and managed by power companies. Now imagine that instead of a few hundred of those you have a few hundred thousand much smaller generators scattered randomly all over the place in various states of repair.
This is the problem with any distributed generation, whether it's fuel cells, solar, wind, or whatever. As soon as for than a few people start using it, you need to start making massive investments into power infrastructure and active management. It's certainly possible, and if something like the Bloom Box could make residential power generation practical and economic then it may well be worth making the effort. But we can't just suddenly all go out and buy one and expect everything to carry on working the way it does now.