As far as the init system goes, the vast majority of packages are not daemons. Only daemons require init support. Writing sysv init files is an art, but it's well-refined. It won't give you the fastest possible laptop boot. Laptop users who don't just hibernate or suspend, but do fresh boots frequently, should definitely go systemd. Of course systemd D is a Borg, assimilating far more than init scripts. But the task of maintaining a couple hundred init scripts wouldn't be hard for a small committee of volunteers. For init stuff outside that, if you can't start a daemon from rc.local you shouldn't be a sysadmin. For the non-init stuff, the trick is to convince upstream developers to support diversity, which can be done by continuing to embrace open standards and APIs.
I think it's too easy to justify grueling jobs with bad work conditions and inadequate compensation by saying "Oh but the people who take them do the work out of loooove!" We do the same thing with teachers: Their jobs suck, their hours suck, their pay sucks, they deal with absurd bullshit, but all that is ok because allegedly, "they loooove kids and receive intrinsic rewards from their work."
We don't think this way about accountants or dentists. We don't expect them to loooove replacing fillings or mastering actuarial tables. We pay them so that their jobs are worthwhile even without the love. And I wish we would apply this standard to all jobs. A coding job where you produce games should be compensated like a coding jobs where you produce financial software, or anything else.
With genetic selection you can max them all out.
Not really. Remember, this is a process where you create lots of zygotes, test them all, and implant the ones with the most desired trait combinations. It's limited by how many eggs can be extracted from the mother - maybe a few dozen? This is not a process where you're splicing genes, or doing any other kind of trait engineering. This is just zygote trait prioritizing. You'll be choosing from a very finite set of "natural" (randomly generated) offspring.
Intelligence is a lot like money.
Those who've always had an abundance generally either think its no big deal, because they've never suffered the limitations of not having enough, or look down on those with less and consider them inferior.
Those in the middle have enough to see the benefits of having more, and want to improve themselves in order to get more.
At the bottom this analogy falters, but I think the point remains. It's easy to dismiss making the rest of the population smarter when you're already smart and not suffering the limitations imposed on those with less to work with. I find the notion that we shouldn't meddle and just leave those who draw the short genetic-straw to be cruel and self-serving. If the lowest common denominator is raised, chance are the whole society benefits, the world becomes a better, more thoughtful place, and the overall pie grows accordingly.
I think you underestimate how easy it will be to mechanize "intelligence" work. A hint: The cost of running code is falling at Moore's exponential, the cost of hardware is basically stable. Janitorial work requires hardware, intelligence work is just running code. Picking stocks, searching law precedents, designing bridges, and many other smart-person jobs, are already being done by computers. Yes, I wouldn't want my kids to end up in a profession from which humans will disappear, but if she ended up a chef or a real estate agent - just picking jobs that don't require a ton of raw brainpower - I would be a proud father. What matters is that she's happy, and that depends a lot on her genes, as it turns out.
Maybe it's because my wife and I are both academics, but when it comes to the intelligence of my kids, I'd be happy to let the dice fall where they may. But because we both have some serious melancholy in our families, the intervention that I would find most tempting is the one that will prevent these dispositions from manifesting themselves in our kids. I don't think that a high intelligence improves a life anywhere near as much as a sunny temperament, and I would never prioritize the genes that predispose for the former over the latter, if my kids couldn't have both.
If I were to choose a child from a huge batch of zygotes, I'd want the one that's generally disposed to be happy - easy going, social, even tempered, and not too fussy growing up. But apparently, geneticists aren't working on identifying the genetic correlates of those traits, even though we know that they are just as heritable as intelligence.
I don't think that I'll have kids, but if I did, the thing I'd want most is that they grow up happy. I would work hard to make sure they grow up in an environment that encourages it. But genetics contributes a lot to happiness outcomes, and if I were offered well-tested genetic help, I wouldn't refuse it. Maxing out their intelligence would not be at all high on my list of priorities. Is this a weird attitude? I thought it was a kind of typical parent attitude, but apparently, geneticists have different ideas.
Maybe they can get virtual people to buy all of their products.
Virtual customers will be the next growth industry
that is exactly what the government does with the CAFE standards.
No, it isn't. The CAFE standards traditionally used the weighted harmonic mean of the mpg values, which gives exactly the same result as the weighted arithmetic mean of the economy expressed in gallons per mile. There are some other quirks- dual fuel vehicles are treated much more favorably than they probably ought to be, for instance- and the standards were recently changed to give bigger vehicles a break. But the larger point is that the EPA isn't completely stupid and does realize that the arithmetic mean is not the correct way of calculating average fuel economy.
One more case where SI units are easier to use. 1 liter/kilometer is 1 square milimeter. Isn't that so much simpler?
For what it's worth, the physical interpretation of this would be that a car with a fuel economy of a given area would be able to drive without needing on-board fuel storage if it were following a trail of fuel with that cross sectional area.