But if that caused a severe performance penalty, that wouldn't be a step forward. The success metric was to get the product shipping fast so that it could be sold over the Christmas period, not to make something that would be wonderfully maintainable for all time.

Sucks to be them if that matters to them. If they'd wanted good internet, they'd have not gone out in the boonies, but would have picked some nice small town that has just enough population to support good networking without the trouble of larger places. Instead, they trade that for lots more space; it's a valid option, even if not one that I'd ever pick.

But if it's that old, it's probably also massively underdocumented (if at all) and so if something unexpected happens, your ass is still hanging out the window. Producing the documentation of what was actually done is at least as valuable a part of a replacement project as the change to the new system, as it should allow someone to start looking at which parts are required, which parts are emulating interfaces (from both sides, usually) that could be de-layered for improved performance and capabilities with no down-side, and which parts are just dumb holdovers from a few systems ago that nobody needs any more at all.

Just because something is painful doesn't mean you can get away without doing it.

Those who hope for a reduction of government meddling in their affairs will see it as a sign of true hope: the sky didn't fall in, despite the fact that the politicians couldn't agree on the most basic thing of all. Throwing them all out of office and only then starting work on the replacement would in fact be just fine...

Maybe, but the effects are less severe because state lines are enormously more difficult to change for short-term political advantage. State-level gerrymandering requires sustained visible policies that affect migration and/or birth rates over decades.

It's not too hard.

It's the long-lived toxic nucleotides that are the real problem. Keeping something safe for 50 years isn't too hard (particularly if everything is vitrified and kept as small pellets so you can use passive cooling) but keeping it safe for 5000 years is a massive headache. So how should we deal with them? Bombard with more neutrons. Like that, they transmute into something hotter which will decay away much more rapidly.

What's the likelihood of that happening, versus the likelihood of something happening to an above-ground cable? Note that you should be thinking about putting the cable well down so that you're unlikely to hit it by accident, just like with water and sewage infrastructure (though even more like gas, if you're in an area with it piped in). Heat dissipation isn't a big deal with domestic supply; you use reasonably thick cabling and aren't really carrying that much current in the first place in normal service.

Hmmmm... Interesting.
here's an Indian (entrepreneural) perspective.

Let me start with a few stories:
The founder of Snapdeal http://www.snapdeal.com/ (an India focused ecommerce company) wanted to do a startup in the Valley but didn't get a visa. So he decided to stay back and do his startup in Bangalore. Snapdeal is now valued at ~$10B and is challenging both Amazon & Flipkart here. Their YoY growth is 6x.

Zomato http://www.zomato.com/ started of by consolidating its presence in India and has now gone multi-national (15 countries and expanding). They haven't gone to the US yet.

There are quite a few startups happening here that are focused on India, the rest of AsiaPac & EU regions and a lot of these have just begun to scale. These startups have started taking the cream of the Engg folks who would have otherwise gone to Infosys, Wipro and from there on to the US.

In fact, I know of a couple of headhunters who place US engineers with Indian startups in India. It is a trickle now and I think it would be good to cherry pick the better ones from out there. There are lots of seriously good engineers who we can use.

As an entrepreneur/co-founder myself (of an early stage in the enterprise space), it makes a lot of sense for us to be India and AsiaPac focused - We have a large market that we would convert first. First of all it is so much easier for us to sell in my backyard and then I honestly don't have the time to wade through all the Visa & other issues that the US would throw at us.

The only reason I would consider the US is the size of the market which will be important to me once I've consolidated and have become profitable. The market that is right in front of me (India and then China) is large enough for me to grow to be a fairly large entity.

I think there's probably less of a problem with voltage drop in electric power circuits built to European standards (and even more so with the massively over-specified UK standards) as the higher voltages mean there's less current in the wire and so less of an impact due to the resistance of the wire itself. That eliminates a lot of the local problems (e.g., due to having kitchen appliances) or reduces them to the level where the switch-mode power supply can usually compensate easily.

I don't know whether the quality of the power delivered by the utility is better. It is for me, but I'm not that far from a major hospital and a major Grid node (and the wires between are underground). I don't pretend that that extends to anyone else other than my immediate neighbours.

Would you accept "inherently impossible to predict any significant length of time ahead"? It's all very well to pick on the reason for the unpredictability (be it quantum uncertainty or extreme sensitivity to initial conditions because of non-linearity) but at a functional level, the outcome is similar: some stuff just can't be predicted in detail long term, and will continue to be like this whatever we do.

Google has a much higher share of search in the EU than in the US, so the EU is far more interested in keeping a close eye on what they're doing. That said, the story was about something that the EP is pushing for and they're not exactly the most influential set of people ever (they've got budgetary powers, the ability to vet the Commissioners, and the ability to revise proposed legislation). What's more, it would probably be possible to split the search side from the rest of Google enough to satisfy at least some of the EP without breaking up Google entirely. (Satisfying the whole EP? Might as well ask for something that satisfies the whole of Congress.)