educationdata

So refusing to provide service is "intolerant"? Does this only works on "protected groups", or does it work in general? How about the baker is willing to provide service to the gay couple, but just not the gay wedding? Now, please elaborate how you define "intolerant".

I actually think you are more possibly to make mistakes in R than in Excel. Because in Excel you always see the results directly, and you can even easily catch an anomaly in a single cell. But in R it takes one more step to see the results, and you probably won't see results of all the rows directly.

So you think all Google needs to do is just filtering out websites blocked by the Chinese government? Really?

You have a underlying lack of understanding how censorship works. They don't just block websites. The Chinese government wants to filtering out results they don't like on accessible websites.

This is definitely not enough. There still need to be an thorough investigation into the history of the project, at least include but not limited to:

1) the internal decision making process involving this project

2) personnel who are responsible for initiating / approving the project

3) how many resources have been put into this project

4) the exact content (data sources, roadmap, goals, etc.) of this project

5) how far this project went

6) where the product / data is stored, how will they be handled when the project is terminated

7) how many of the project outputs have being provided to the Chinese government

8) the details of Chinese government involvement into this project

Also, not limited to Google, any U.S. company helps any authoritarian government doing censorship needs to be investigated.

BTW, I got lots of downvotes from comments criticizing China.

This number does not mean much. In the old days, more Chinese students rely on full scholarship to study in the US. Now, more and more rich Chinese kids (mostly come with very poor academic background) can study in the US.

So the overall number does not reflect "brain drain", because not all international students can be counted in "global talent". A more meaningful number should only count those graduated from good universities, or having scholarship etc.

I don't follow. Do they match faces to crimes or match faces to faces of criminals?

"if 1 in a million are criminals"?

NYC population is 8.6 million in 2017, so there are only 8~9 criminals in NYC? Wow.

Come on, you should give a much better estimate.

data.table can handle millions of rows easily, as long as the data can fit in the memory.

The second example should be written like this:

  diamondsDT[cut != "Fair", .(AvgPrice = mean(price),
                              MedianPrice = as.numeric(median(price)),
                              Count = .N), cut][order(-Count)]

There is no need to break it to 10 lines.

> If you're better at writing/composing dplyr, you can often make up the speed difference between the two in terms of the savings in time spent writing and reading code.

dplyr syntax is definitely more concise and readable than base R, but comparing to data.table I don't think it has any advantage in terms of saving time writing or reading code.

"cherry picked"? There is no doubt that data.table is generally way faster than dplyr, unless you cherry pick a few use cases.

The problem is that dplyr is much slower than data.table and RStudio is promoting tidyverse too much to make the slower choice a default for many users.

The article is 100% correct on this issue.

"cows the population into authoritarian submission" ? I don't see this happening at all. There have been countless protests against the Trump administration, and the main stream media has being criticizing the Trump administration every single day. Please give me an example supports your projection.