I could be imagining things, but I think there's actually a slight decrease in ability to carry on these kinds of discussions than there was a half-generation ago. You could assume that most computer scientists in the '80s and '90s had basic ability to use a listserv and carry on a conversation suited to the medium. I think that is less true now: many computer scientists in 2013 have absolutely no idea how to carry on a productive text-based discussion.

If you're trying to recreate a physical meeting, I agree. But it's quite possible to have productive virtual meetings if people adapt to working in a manner suited to the medium. I have a regular group of collaborators who I sometimes meet with in person, and sometimes meet with on IRC. The two kinds of meetings are both productive, real meetings, but with different strengths and weaknesses. However it works because we're all familiar with IRC and how to use it productively, rather than trying to shoehorn some other communication style into it.

Generally true, though the tricky part is that in disagreements between divorced parents, courts sometimes try to take the kids' wishes into account in ways that wouldn't have legal standing outside the divorce context. That's because the court is supposed to do many things in the "interest of the children", and when divorced parents disagree over what that is, they might try to discern from the children what that is (with varying degrees of success).

Here the court seems to have taken the children's wishes into account, but ultimately decided that, when mom and dad disagreed over what was in the interests of the children, medical science was a better tie-breaker.

I haven't worked anywhere that locks the stairs. Not only do I regularly take the stairs, but there are five stairwells (one in each corner and one in the middle), so I can take whichever one's closest.

Man, you people must work in some weird-ass buildings. When I want to see someone above or below me, I just take the nearest stairs up. Sure, there is only one central elevator bank, but there are 5 flights of stairs (one in each corner of the building and one in the middle, near the elevators). Takes not even two minutes.

I've worked in skyscrapers and haven't really found that to be true. Especially within your bank of floors, e.g. if your department is floors 15-20, people treat it as pretty local.

Architects in the early 20th century came up with an interesting solution to this: use a third dimension, and install elevators. Now you can walk horizontally in two dimensions, and travel up/down, bringing a large company's employees all within relatively short distances of each other.

It's also a completely different problem from information retrieval in a messy domain like "all documents on the internet". Watson is built mainly out of more structured data: dictionaries, almanacs, atlases, Wikipedia infoboxes, etc. It turns this into a huge database of knowledge, and then does inference on that database to try to answer Jeopardy-style questions posed in natural language. But this doesn't even try to tackle the other side of the natural language problem, which is parsing not only a natural-language query, but the entire contents of the internet.

In short, Watson might compete in the Wolfram Alpha space, of retrieving structured knowledge from databases, but not, at least not without a major overhaul, in the general document search space.

Maybe they're including the cost of treatment for all the new "learning disorders" that will be invented.

IMHO, there is no communism without the first industrial revolution. Marx is a consequence of technology.

Interestingly, Marx himself believed something along those lines, that communism would be the product of certain tensions produced by the (then-)modern industrial economy. That's one reason that orthodox Marxists (like Kautsky) were very skeptical of the Russian Revolution. If you believe that communism is the result of tensions within an industrial economy, led by the urban proletariat produced in factories, then a communist revolution without first having industrialization doesn't make sense: the Leninist ideas of a vanguard party that would seize power, crash-industrialize an agrarian economy, etc., and then usher in communism, ends up seeming very ahistorical and strange.

That's one way of defining the average (mean) standard of living, yes. But that does not necessarily mean that the median standard of living also increases in the same scenario, without stronger assumptions on the distribution.

But what happens computers are as good as people in most of all the things that qualify as jobs nowadays?

Science fiction writing covers the two limit cases pretty well. Let's say machines can now account for all basic human needs, producing food, clothing, shelter, etc. sufficient for the whole human population. Then at the dystopian and utopian extremes, we have:

Possibility 1: These machines are owned by a small ruling class, who uses their control over this vast pool of robot labor to rule the world, and over the impoverished underclass who own no robots.

Possibility 2: These machines provide for everyone's needs, freeing up humans for a glorious age of space exploration, science, what-have-you.

The basic parameters of the argument are clear, sure, and have been clear for a few hundred years: automation may replace large numbers of jobs with machines controlled by a smaller number of people, but may also create new jobs, either directly working on the technology involved, or indirectly in other areas. The more difficult questions are in the details. Do the numbers always match up, and what factors influence whether they match up? Does automation lead to more general shifts in the economy, e.g. either concentration of wealth or decentralization of wealth? If it could do either, what factors influence that?

My own view is to be rather skeptical that there is a universal answer. These kinds of articles give off a whiff of a kind of Panglossian view that the technology/economy ecosystem is in a Gaia-like eternal balance, and I don't see a strong reason to believe that's true. Instead I think we need to look at specifics to determine what effects a given technological advance, within a particular existing economic situation, will have.

Nah, our representation is still getting paid too, alas.

To be somewhat more precise, there isn't a mandated price, in the sense of formal price controls. But the federal helium reserve accumulated huge stockpiles, and has been slowly selling them off since 1996, which has kept the price low by flooding the market. On the one hand, that discourages private investment, but on the other hand, it's not clear it's entirely a bad thing: if we don't actually need this helium reserve lying around forever, selling it off slowly seems like a reasonable thing to do.