neuxenian

The relativist position in modern philosophy of science has little or nothing to do with Cartesian dualism; it has more in common with the incompleteness theorem in certain respects.

The short version is that any theory of observations (i.e., scientific theory) necessarily defines the observational process and therefore the observations themselves, so that it is impossible to separate an explanation from its explanandum. What's relativist about this is that the very thing you are explaining by the theory is implicit in the theory, so that what you have are a series of paradigms or theories that replace one after the other.

This isn't to say that we don't predict things better, or even that there is no reality separate from the theory, only that it doesn't matter at some level.

All of this too is sort of distinct from what Kuhn is remembered for, which is the social nature of science, which is all too real.

Scientists are often in denial about these social factors, which is naive at best and dangerous at worst. Humans are not conduits of God; they are imperfect machines that are part of the system they are studying.

My sense is that Morris (who I respect immensely for his work) is kind of fighting a strawman argument whose depths he doesn't entirely understand, because of some personal conflict he hasn't come to terms with (ironically, given the nature of what he is arguing against).

We looked into this quite a bit when our child was born because we grew very skeptical of much of the guidelines we were reading. Sleep in particular was difficult for our child for various reasons and at some point we realized that following certain guidelines resulted in more risk than violating them (e.g., it became an issue of following guidelines and neither us or our child sleeping or violating guidelines prudently and getting sleep).

We did not cosleep with our child. But in the process of reviewing many sleep guidelines we reviewed the literature and found it's more complicated than it seems. The NPR article doesn't summarize the complexity as well as it could.

The problem is that studies of risk of cosleeping generally do not control for confounding variables very well at all, and when you do control for these confounds, the risk of cosleeping disappear.

For example, in the US many cosleeping parents tend to be minority, immigrant, lower SES, etc. which is associated with other risk factors. When you control for those trends, cosleeping isn't associated with increased risk.

This has lead to many sleep scientists to conclude that cosleeping research has largely been sociodemographically (e.g., ethnically) biased in design and analysis. There's a big rift between sleep scientists and pediatrics in this regard.

The NPR piece proceeds along the lines of "humans have been doing this for 200000 years so it must be fine" but it's more like "people who can afford separate beds for their children tend to have a lot of money, globally speaking, and that's associated with decreased risk of infant death for a multitude of reasons having nothing to do with cosleeping."

This was the tipping point for me, actually, when I went through this same experience about a year ago:

https://consumerist.com/2013/06/18/amazon-cancels-my-6000-or...

It sounds like a minor bizarre Kafkaesque corporate nightmare, but the whole experience was so surreal I can't begin to convey it. Amazon demands, out of the blue, with no explanation, copies of sensitive financial documents to immediately be sent to strange unknown numbers over fax, or they will shut your account down. And when you do comply, they assert completely capriciously, that you have not, without explanation. And then they shut your account down. The whole thing seems like some obvious scam, except that Amazon representatives can confirm for you that it is real, even though they themselves cannot actually communicate with anyone involved.

I mention this here because of the sense of impunity involved. It was so absurd that any other corporation wouldn't think about doing something like this because of the obvious consequences that would occur.

I've generally felt positive toward Amazon over my lifetime, but in the last couple of years I've soured on them, and it's directly related to a large list of factors that in aggregate suggest Amazon is too large and is abusing its position.

The article undermines its own arguments by pointing out things like Amazon represents "less than half of all online commerce," that "(44%) takes place on its platform." The author is attempting to argue this is not too much, but to me their argument only proves the point.

I'm not sure what to do about it, but Amazon has become a problem, and they are abusing their monopoly.

Thanks for these links--these are interesting and probably come closest to addressing my questions of anything I've seen so far.

Leverage cuts both ways, though. As someone who just sold their house, the costs are a percent of the sale. You don't get taxed on the sales (generally) but you will probably have to pay real estate sales costs. So yes, if your home increases by 3%, you've gained 15% growth on your investment, but if the sale cost is 5% of the the sale price, then things aren't so clear anymore.

The real calculation is the total monthly cost of ownership plus any sale prices on the buying and selling ends over the period, relative to the total costs of renting.

Over the 10 years we've owned our home, yes, it's been worth it to buy, because we can recoup money that would have gone to someone else.

However, over a short period, the sales costs would dwarf any returns we would get.

The sales costs are largely fixed on both ends, as a percent of home value, and diminish as a total percent of gains over the period of ownership. So the period of ownership is relevant.

This in turn is relevant because your mobility becomes relevant.

The previous market we were in too, was so overpriced relative to the rental cost that we actually saved money over the period by renting rather than buying. Then the Great Recession happened.

Either strategy makes sense depending on your mobility risk and the market. I don't think it's clear that one or the other is generally better.

I've started reading some quantum computing texts, and dabbled a bit with IBM's quantum computing experience, and this is where I get stuck:

I can follow the basics at the level of "gates" or circuits -- a lot of what's out there is essentially quantum assembly language or something of that sort.

But what about interfacing classical and quantum computing? Is the idea that you just run something from start to finish on a quantum computer? Or that you take the output from a classical system and input into a quantum system?

Some of the things I'd be most interested in with a quantum computer seem to require an interface with a classical computer, or at least raise the issue, which generally isn't addressed by these sorts of pieces. It's sort of assumed that you're programming whatever it is you're interested in at the level of machine/assembly code.

I'd add a computer of some sort to that, probably a laptop

but basically I agree 100%

As a professor, I can say most professors don't like lecturing, but for some classes the alternative is to have small interactive classes, which isn't financially tenable for most universities for all classes.

So my reaction to these sorts of studies is "ok, then should we increase tuition to make up for splitting this 200-person class into 20 sections?" Even classes of 20 are too big.

I would like to change something about classes that still have a big lecture component but I'm not sure how to do that optimally. I think some of them could be done in an online format with videos, but I'm not sure if that's the best way either. Maybe if some small group discussions could be added?

I've used R since it was in beta form, and one of the most interesting things to me (aside from the evolution of type hierarchy) has been changes in use of the assignment operator.

When I started, = was the norm for assignment, with == for evaluating equality. Later I started noticing that people were recommending <- based on scope concerns, but it was kind of subjective preference. Now I see articles like this saying that the "preferred use" is <-, and some people don't even know about =.

I agree = versus == can lead to tricky errors in code, but I still prefer = for various reasons in general in languages (although I get the scope arguments).

The reason is that = is shorter, and assignment is an definitional equality, which to my impression is the whole point of programming in general usually. As others have suggested here, between = and ==, = is by far the more commonly used operator in meaning, so to me it makes sense to use = for succinctness.

In math, there is an "equal by definition" operator, with three lines, so I could see that, but keyboards don't have that, so it's more steps. := is also a kind of standard "by definition" usage as discussed in the article, but again, it's more steps. I'd still prefer that over <- in R.

This discussion is reminding me of how much of an impact 2010 had on me in elementary school. For me it fostered sci-fi fantasy in the same way that I think Star Wars and Star Trek have for lots of kids.

To me, the universe of 2010 was just barely within stretching distance for society in some forseeable future, whereas Star Wars and Star Trek were unreachable. That somehow increased its appeal to me as a child, as I could envision living in the world of 2010 in a way I could not with other sci-fi universes. The universe of 2010 was my universe, in other words.

I've been afraid to rewatch 2010 as an adult, because I know I will see it differently.