twstws

The example was a little clunky, but I don't find it misleading. A biplot of two normalized variables is elliptical, if the variables are correlated. This particular hand-drawn example does indeed look a bit weird, but that doesn't detract from the main point. It clearly shows the relationship between the original data and the ordination; it's a rigid rotation.

This is easily grasped with a 2d example, despite the fact that PCA makes no sense with only two variables.

It's not dimensionally invalid if you remove the dimensions via normalization.

Each variable is standardized to mean = 0, standard deviation =1. If you reject this as arbitrary, you are rejecting correlation analysis as a whole - this is exactly the same standardization done to two variables in bivariate correlation, extended to a multivariate data set.

PCA is a form of (or at least related to) correlation. With standardization the resulting transformation hihlights variables in the original data that are most highly correlated. Without standardization you're visualizing covariation. Unlike correlation, covariation is influenced by the magnitude of the variables.

By standardizing, you control for differences in the magnitude of the variables, and focus on their inherent variation instead.

In most cases each variable is standardized to put them all on a comparable scale.

> we can't determine if a computed result is actually a sought result or its reflection/negative

That's not how I'd explain it. The 'sought result' and its reflection are not two different things. They are the same thing, differing only in a trivial detail of orientation. The 'negative' of the result conveys exactly the same information as the result.

Good point. I do think the frequency of cases where the only option is private fund raising is much reduced under public systems.

I suspect it is because those with the means to move are generally doing well enough to avoid the failings of the US system.

I can't understand how Americans, who have the least efficient, most expensive health care system in the world, are so quick to write off public healthcare. The idea that government healthcare is more wasteful than a private system is not supported by facts, but the theory is too hard to let go of.

Which is not to say there is a perfect solution. But just about any of the alternatives would be better than the US system.

Except for the US, most of the developed world does, via public health care. It's not perfect, but better than the alternative.

This is truly a wonderful story. But it makes you wonder how many others in a similar situation weren't so fortunate.

I lived in the US for two years, and I never understood the aversion to government healthcare. The Canadian system is far from perfect, and I know there are failures. But it's still a lot better than soliciting for online charity on a case by case basis.

I'm impressed and humbled that it worked in this case. Just a little disturbed that it was necessary at all.

I expect quinoa will be selected for environmental tolerance first. Given how difficult it is to grow in th US, top priority would be getting consistent high yield in North American conditions. Breeding has to meet the farmer's needs before you start considering the consumer.

True, to a point. Debunking one of these in a class is a good exercise. Preparing a lesson takes time, though, especially when you have to respond to something out of the blue. You could easily get caught spending all your limited spare time discussing why these theories aren't science, instead of the ones that are. Evolution is a big topic, and I'd rather spend my limited time dealing with the substantive bits.

This bothers me as a former professor, because these fringe ideas undermine teaching and waste time. Imagine trying to present a lesson on evolution, and one of your students brings this up. You spend a few minutes discussing it. Of course, you've never heard of it before, because it's beyond implausible. So you spend your evening looking into it, and the holes in the theory. Next class you spend more time discussing it. If you're good, the student understands and you move on. If not the student leaves thinking this is a valid alternative viewpoint. And no, an idea does not become valid simply because it's not impossible. It's not unreasonable to demand more than a faint hope probability before judging an idea worth serious discussion.

This happens once, and you can make it a teachable moment. But when the scenario starts to repeat itself it undermines the effort you're putting into teaching real science.

He makes the argument on his website: http://www.macroevolution.net/human-origins.html#at_pco=cfd-...

1 people think hybrids are sterile, but they're not 2 people think hybrids don't occur in nature, but they do 3 people think only plants hybridize, but animals do to

From this basis, he concludes that a chimp-pig hybrid is plausible, and proceeds to lay out his theory.

The problem is the three facts he starts with are trivial compared to the obstacles raised by PZ Meyer. To take just one, there is the difference in chromosome number. In most cases, if a human ends up with the wrong number of chromosomes, it's a lethal condition. Or you end up with Down's syndrome. With one extra chromosome. The hybrid this guy posits has a dad with 38 chromosomes and a mom with 48.

I could argue that that's not a big deal. In the plant groups I study stranger things happen. But that's in plants. Primates, as I understand it, are much more sensitive to chromosomal abnormalities.

There are many logical, evidential reasons to discount this hypothesis. Again, check out the pz meyer post linked elsewhere. Claiming I don't understand evolution because "given enough tries anything is possible" is facile. Of course anything is possible. But what is probable here?

This. The author makes the absurd claim that because many hybrids are fertile, any hybrid could be fertile. Using hybridization between two closely related birds to support the notion that pigs and apes could breed is ridiculous.

The thing I find most depressing about these crackpots is that some otherwise productive scientist has to spend an evening debunking their nonsense. And it still doesn't stop.

I taught evolution for undergrads, and occasionally had to deal with these issues from students. It takes time to look into them, and no matter how thoroughly debunked the theory, some kids find them too irresistible to let go of. It undermines real education in the end.

This is silly. Debian also provides rolling releases, testing and unstable. The release cycle is slow only if you ignore the constant development of Debian Sid.

This was hard for me to read. I spent four years waiting to adopt a child.Did almost a year of training and home visits, then waited. And waited. Why does it take so long? Because we selfishly insisted that we'd only take one ortwo children. The social workers were not subtle in letting us know that we were selfish to not want a sibling group of 4 or 5(?!) kids all at once. They made it absolutely clear that we could not expect to have a newborn, and a child under 8 was unlikely.

So to hear that, on a hunch, a judge can give a newborn to a couple that had expressed no previous interest in having kids, wtf. It's a nice story, and I'm glad that it worked out so well for everyone. But for me it really underlines how fucked up the system really is.

> There's no such thing these days as a real musician, Dave was part of a dying breed of people who got into music because they love what they do, not because they want to be famous or rich like modern day musicians.

I think that says more about you than the state of modern musicians. There will always be 'real musicians', motivated by their love of the art. Whether or not we can find them depends on how willing we are to dig past the shallow acts that top the charts.

I don't know what this has to do with bayes vs frequentist. I am not arguing that the data do not have a probabilty distribution. I am arguing that it is better to show all the data when possible, rather than an eye-catching but lossy summary.

I think the default should be the method that displays the most information. Why hide information if you don't have to? In the case of one dimensional data, a dotplot shows the reader everything. Using a boxplot reduces information content, mean-plus-errorbars reduces this further. The mean plus errorbars imposes a probability distribution, which may be wrong, it doesn't reveal a hidden truth.

The same holds in two dimensions. Show me all the data, and include a regression line or a spline to highlight a trend. Only start hiding information when the scatterplot becomes misleading. That is, when overplotting prevents me from accurately assessing the actual distribution of the points.

Jumping immediately to a density plot also restricts me to your interpretation. The original data is lost. With a scatterplot, the raw data can be recovered from the plot, so i can do my own analysis should i be interested. This is common in meta-analyses that extract data from multiple published papers. If those original papers had used density plots instead of scatterplots, reanalysis will require direct access to the underlying data. Once the original author dies, or loses the data, all further use of the data is lost.

> The difference between deep red and orange is a lot bigger than the difference between dark blue and slightly less dark blue.

That may be true, but is it better? Does the scatterplot underemphasize differen densities, or does the density plot overemphasize them? I think the scatterplot is more intuitive. There are 21 equal steps between 0 and 100% black. Two points are twice as dark as one, four points are twice as dark as two. Darker means more, lighter means less.

Compare that to shifting from blue to red. Does the shift from orange to red indicate the same density difference as the shift from blue to orange? To decide you need to consult the color scale. The scatterplot is intuitive, and requires no scale.

> The hexbin has lower spatial resolution, it's true, but I'd argue that the spatial resolution you get in a scatterplot is illusory. It doesn't reflect the underlying probability distribution, only the particular sample.

The spatial resolution of a scatterplot represents empirical reality. Each point corresponds to a single observation, with no probability distribution implied or imposed. The density plot, in contrast, imposes a probability distribution, which may or may not reflect the true distribution of the population. The larger the bins, the more likely the displayed pattern is 'illusory'.

If your data can be displayed without points overlapping, a scatterplot can display all the information, while a density plot will always display only a summary of the data. The larger your grid size, the greater the loss of information.

The superiority of your hexbin follows from setting the density too high for the scatter plot. With points this dense, opacity of 5 or lower is necessary to see the uneven distribution along the x axis. With appropriate opacity, the two plots are pretty similar visually. What's more, the hexbin by definition has lower resolution, since you lump data into discrete bins.

This is an example of bad plotting practice, not a bad plotting method. That said, eyeballing a plot is a weak way to analyze data this dense. That's what statistics are for.

1 paren moved and all line breaks and indentation removed. This is intentionally obfuscated. Is there any language that is easy to read when you put five lines of code together like This?

I haven't read a lisp style guide, Emacs just takes care of indentation - it is immediately clear when a paren is wrong because the shape of the function is wrong. If you are writing lisp with an editor that doesn't do this, get a better editor, don't blame the language.

There is an important distinction between not trusting someone without non-academic experience, and not trusting someone with too much academic experience. I agree with the first, but was refuting the second.

In my experience, as someone who completed a Phd after a few years working, phds often have richer real-life experience than the 'average' person, and almost always richer than the common stereotype inferred in the comment i responded too.

This attitude is frustrating and sadly widespread. Basically, you're saying that you can trust someone with some scientific training, but only up to an arbitrary limit. You assume this is a zero sum game, that it is impossible to learn a lot about biology without simultaneously acquiring some compensating deficiency. Or that only the deficient personality would pursue a Phd.

How is it anything but anti-intellectual to say that having too much expertise makes one a less capable leader?

You have misunderstood. The current situation has Sony shipping devices running several/many different programs with GPL licenses. They don't want to provide their modified source code for these programs to their users, in violation of their obligations under the GPL. Most of the copyright holders of this code do not have the means to pursue an infringement case.

Busybox is the exception. The SFC actively enforces the license for busybox. In addition, once you lose your right to use busybox as a consequence of a license violation, the SFL will let you ship it again only if you come into compliance on for all of the GPL code you ship.

So they are making their own busybox as a way to continue to violate all the non-busybox GPL code they use.

If you comply with the busybox licence, you can continue to violate the licence on all the other GPL code. But violating busybox means you have to comply with all of your GPL code.

You never have to release your non GPL code.

It sounds lame, but it in no way reflects reality, so i don't understand what your point is. Under the current system, the recently deceased authors book would earn royalties for his estate for 75 years. Grand children he's never met will benefit financially from his work. We grant them a monopoly, and get no social benefit. This is a perversion of the intent of copyright, which was to promote more creative work.

Simple inheritance is categorically different. Your inheriting money from your family has no negative impact on others. But if you inherit an intellectual monopoly, it is taking something away from the public good for no benefit. That may be acceptable for a reasonably limited term. But how anyone can imagine a copyright term that extends across multiple generations is reasonable is beyond me.

> In your system where copyright dies with the author, the interests of others who might invest in a project or stand to gain from it (like a creator's children) are not looked after.

This is the problem, I think. If a product requires investors, then I can accept that their interests should be looked after. Not for the ridiculous term that copyright currently allows, but it is reasonable that their protection should not be strictly limited to the life of the author.

However, the current system goes to great lengths to protect the interests of people merely because they have something to gain, not because they actively contributed anything. If I make cars for a living, I don't expect or demand that my children continue to receive a royalty for each mile driven in one of my cars after i die (or just retire). They are taken care of through my estate, insurance etc, not by perverting the market.

I don't think author's children are inherently more deserving of social protections than anyone else. Jk Rowling's kids would be just fine if copyright lasted only twenty years. If they want a secure income after that, they should be able to do some productive work of their own.