randomvectors

We're not arguing about the safety and viability of banks. Instead, we're directly comparing the safety and viability of banks vs Facebook. If you think that Facebook comes out on top in that comparison, either you really haven't been paying attention, or we fundamentally disagree about what safety and viability mean.

> At this point, I think I'd honestly rather put my money with Facebook than many of those actual banks out there.

This is so ridiculous that I don't even know how to argue with it.

I'm super grateful that my first exposure to programming (high school in Eastern Europe, 10+ years ago) was in C++. I don't think we even did a lot of C++ specific stuff so it was probably closer to C but using cout instead of printf. Covered things from if-statements and loops to sorting and tree-searching algorithms. It doesn't sound flashy and it somehow goes against the trend to try to make learning exciting, but it lays a really nice foundation in your learning.

You're arguing with a point that I'm not making at all.

No, being opinionated by itself isn't a good enough reason to belong in new code.

Your post is an ad. You've also tweeted a link to this thread both from your company's twitter and your personal account. It's kind of desperate and embarrassing. Just don't.

Somewhat unrelated to the article, but I'm curious - how commonly and for what purposes are other implementations used for?

Organise your own high stakes game. Happens more often that you would think.

More commonly this happens in private games, not in casinos.

Isn't the whole point of code review that another human looks at your code?

Because Facebook haven't demonstrated any integrity whatsoever in all of their history as a company.

Not claiming that it is.

Though it will get more publicity and wider exposure than most ICOs usually do, I'm sure.

Even in the most optimistic case, what's about to happen with Facebook's cryptocurrency will border on a straight up scam. Which is fitting both for their business model, their image and for the broader cryptocurrency space.

Something that happens to me, which feels related, is that if I'm listening to a podcast for the second time, at memorable points of the podcast (when the topic changes for example) I always remember exactly where I was when I heard those words for the first time (which part of which street if I'm walking). It's strange.

Prepare for some frustration while you get used to the language. Resources:

Advanced R - http://adv-r.had.co.nz/

R Packages - http://r-pkgs.had.co.nz/intro.html

I also write and contribute to R packages at work and these two books have been really useful. That being said, Advanced R isn't the best read because a lot of it is just listing exceptions and strange behaviour of the language but it's useful info, especially the chapters on non-standard evaluation and the OO systems. Similarly, figuring out how the package system works can take a bit of experimenting - just stick with it, try things out and pay attention to how things are done in the existing codebase.

Also, your colleagues are right - tidyverse isn't suitable for heavy use in packages, it's for interactive use and EDA.

Which other graphing tools are you comparing it to?

I think ggplot2 is a decent library, but in my opinion it gets more praise than it deserves.

Tidyverse is so overrated that I don't even know where to start... sure dplyr is nice to use if you're working interactively and ggplot2 has good first principles as the basis of its design, but that's about as far as the tidyverse niceness goes. The apis, the documentation, the inconsistencies (on a language level, between tidy packages and within each package), the problems with backwards compatibility, the evaluation (standard vs non-standard vs tidyeval) - it's a big mess.

If you're writing code that's deployed in any way, it's best to avoid the tidyverse as much as possible. This is also acknowledged to some extent by the main developer - https://www.tidyverse.org/articles/2018/06/tidyverse-not-for...

Except for working with data, which is what the original post is about.

Having learned and used both, I disagree with most of his points:

- Learning curve

- Machine Learning

- Parallel computation

- C/C++ interface

- Object orientation

All of these are wins for Python, some of them like the learning curve are wins by a huge margin. I should probably do a point by point rebuttal later but so many of his points are incorrect and/or poorly justified.

While we are definitely going to have more AR, I'm not convinced that it's the future at all. But I guess that's the case with most people and most technologies until they become ubiquitous.

AR still seems like a solution looking for a problem.

“Somebody once said that in looking for people to hire, you look for three qualities: integrity, intelligence, and energy. And if you don’t have the first, the other two will kill you. You think about it; it’s true. If you hire somebody without [integrity], you really want them to be dumb and lazy.”

-- Warren Buffet

This is what your post is based on, right?

Some managers will be better judges of character than others but as soon as you formalize the process, you also very likely open it up to be exploited by people with low integrity and high intelligence.

You're ignoring what I wrote above and you're arguing with a point that I'm not making.

Yes, it's very rare to have the term notebook mixed up in this case. Unless you've only heard of the term "Jupyter Notebook" and have zero context. By the same logic Python can get confused with snakes.

Shiny is okay as a way to share simple analytical apps but it's syntax and performance are atrocious.

1. As a hiring manager you should be better at differentiating between people and treat a degree as just one data point that you take into account, and not as a filter that gives no false positives.

2. You're pointing out that self-educated people aren't worth the risk and then you go on to advise him that he should self-educate.

3. A point that I think you're kind of making and that I agree with, is that a degree is a signal that you've gone through a certain process of education. If you don't have it, then you'd better signal the same thing in a different manner and demonstrate that you're familiar with theory, best practices, design patterns and so on; and that you can stick with something for a continuous period of time (showing up and doing at least the minimum amount of work for 4 years isn't a signal that all candidates have).

Generally speaking, the purpose of something written is to be read. Putting text in the bitcoin blockchain is as good as writing it down on something durable and burying it deep underground. It might last a long time, but it's pretty pointless.

1. You can't make anything last forever. Just a little while longer if you set things up correctly.

2. Is your blog really that important, meaningful or valuable that it needs to be preserved after your death?

3. Apart from technical solutions, writing things that people will read and share is the best way to preserve what you've created.

First thing that popped in my head when I read the headline. Really liked the episode.

I still find it better than the built-in night-shift on MacOS.

I just checked out Riot. Looks nice.