Are those mutually exclusive options? The industrial revolution resulted in a lot of reforms in the UK, including the beginnings of the process that ended with universal suffrage. It was also a time when the poor were exploited and oppressed, although this time by the upper middle classes rather than (or, more accurately, as well as) the aristocracy. Given the end results, I think most of us living in the countries that benefitted from this process are glad that it did, as well as being glad that we can look back on it as a transitional step. Given the choice, I'd much rather that we had gone straight to a post-scarcity utopia in 1750, but as far as I know no one has yet come up with a way of making that happen...

You might want to check the history of the industrial revolution a bit more carefully. Worker conditions in Foxconn factories look like paradise in comparison to conditions in England back then.

Hiring good MS admins is hard too. Possibly harder, because, although there are more, the signal-to-noise ratio is lower.

I actually did see someone using one of them last week, but he was a tango teacher who also DJ'd. If you'd asked me the same question a week ago, I'd have said over a year ago. I rarely see people with stand-alone MP3 players now that a cheap smartphone and a decent sized SD card can be had for about the same price as an MP3 player.

Just to clarify: you are talking about mechanical hard disks there, right?

SSDs have been roughly doubling in capacity for the same price every 9 months for the last 15 years. If that continues, then they'll be where hard drives are now in 2-3 years in terms of price per GB. It's important to remember, however, that a more important metric than price per GB is price of the smallest drive bigger than what I need. For a lot of corporate desktops, 40GB hard drives are big enough. They get re-imaged periodically, so a larger hard drive isn't that important, and everything except the OS and a few apps is stored on a file server. The cheapest hard drive I can buy is 1TB at £60. The cheapest SSD I can buy is 32GB at £35. I can also get a 60GB SSD for £40. If I am buying 1,000 machines that are going to need under 40GB of local storage, I save £20,000 by going with the SSD.

Unlike hard drives, it's quite easy to make smaller-and-cheaper flash drives: just put fewer chips in the enclosure.

You are conflating a bus address width with a storage technology. There is no currently-planned 1TB SD card, there is just a plan for the next generation of the standard to support addressing up to 1TB. If you made the same assumption about addressability equalling shipping products, then most current laptops would have 256TB of RAM...

I was at a talk buy some guys from FusionIO a few weeks ago. They said a lot of interesting things, but one of the points that they made was that every generation of flash was slower than the last, as well as less reliable. That's the trade you make for greater capacity, but it's not sustainable in the long term. It's not that flash is worse but getting better, it's that flash is better (but more expensive) and getting worse.

If current trends continue, then in a few years the improvements in capacity will be lost completely to the extra duplication required to achieve reliability. Flash is basically a dead end at this point. It will almost certainly be replaced by PCRAM, MRAM, memristors, or some hybrid, although I wouldn't be surprised if the the result is marketed as flash...

It's a mistake to think of Samsung as a single company. It's more a tightly-cooperating group of businesses. Departments try hard to buy components from other Samsung departments, and to cooperate on mutually relevant projects, but aside from that they're run more or less independently. This is, in part, why Samsung suing Apple while selling them a load of components makes sense: the part suing Apple and the part selling to them are almost separate entities. The CPU and flash manufacturer parts sell to both Apple and the phone-making part of Samsung and has no interest in the lawsuits in either direction except as far as it changes the amount that their customers are willing and able to buy.

Possibly. Given Intel's failure to produce compilers that gave good performance on Itanium, however, it is more likely that we'd have seen a bit more competition in the 64-bit arena. HP had committed to killing Alpha and PA-RISC, but POWER, SPARC and MIPS were both doing quite well until x86-64 squeezed them out. If they'd only been competing against Itanium, they'd have had a much better chance.