kkoncevicius

43 ms·

511.
by kkoncevicius 6y ago
I think HN temporarily brings newest posts at the front so that someone who checks for new replies can quickly find them on the top. After a while they are moved to the bottom. I am not sure on the exact mechanism behind this but observed i
512.
by kkoncevicius 6y ago
All %fun% constructs are just simple functions that can be created by the user and can be used as infix operators. Base R has a few of those: %in%, %o%, %*%, %x%, %%, %/%. magrittr created %>% which, when used in infix: x %>% f()
513.
by kkoncevicius 6y ago
There is a big difference between what is allowed from within a package and what is possible in native implementation. As an example - currently this pipe operator seems to be implemented at the parser level. The parser simply takes the pip
514.
by kkoncevicius 6y ago
I think that symbol is used as an approximation of "lambda" character (λ) on a standard US keyboard.
515.
R adds native pipe and lambda syntax(developer.r-project.org)
250 points by kkoncevicius 6y ago|144 comments
516.
Fast navigation in terminal: coming full circle(karolis.koncevicius.lt)
1 points by kkoncevicius 6y ago|0 comments
517.
Using redo to manage R data analysis workflow(karolis.koncevicius.lt)
2 points by kkoncevicius 6y ago|0 comments
518.
by kkoncevicius 6y ago
HN is working fine; different sub-groups segregate under different topics. People who care about Rust comment under threads about Rust, people who care about science-fiction post comments in threads about science-fiction. And people who car
519.
Jonathan Blow on Windows Terminal (video)(youtube.com)
24 points by kkoncevicius 6y ago|11 comments
520.
An opinionated list of best practices for textual websites(seirdy.one)
22 points by kkoncevicius 6y ago|0 comments
521.
Porn, Zen, and .vimrc(karolis.koncevicius.lt)
204 points by kkoncevicius 6y ago|128 comments
522.
Four Golden Lessons (2003)(nature.com)
26 points by kkoncevicius 6y ago|6 comments
523.
by kkoncevicius 6y ago
I only read the first question in theory.md and think the answer is quite weak. > What is regression? Which models can you use to solve a regression problem? The current answer only list some names that have "regression" in it,
524.
by kkoncevicius 6y ago
You don't really need to experience it to see how this one day might happen to you. Here is a recent example with RuboCop: https://github.com/rubocop-hq/rubocop/issues/8091 "Cop" suddenly becam
525.
by kkoncevicius 6y ago
The optimist in me thinks this problem with corporations and addiction will soon pass. And my reasoning is two-fold: 1) all addiction is "accelerating", 2) social media is ripe for parasites. First about accelerating addiction. Th
526.
Avoiding the Global Lobotomy(meta-nomad.net)
282 points by kkoncevicius 6y ago|158 comments
527.
by kkoncevicius 7y ago
Using tibbles outside of tidyverse can be dangerous. df_iris <- iris tb_iris <- tibble(iris) nunique <- function(x, colname) length(unique(x[,colname])) nunique(df_iris, "Species") > 3 nuni
528.
R 4.0(stat.ethz.ch)
306 points by kkoncevicius 7y ago|162 comments
529.
by kkoncevicius 7y ago
> But what alternative to Wikipedia do you propose? What I do personally for subject that feel a bit controversial is: 1) read the Wiki article 2) read the Talk page 3) look for edits that were made earlier (special before the 2016 USA e
530.
by kkoncevicius 7y ago
> That said, I have heard my mentor say "I don't get the point of <insert tidyverse function> - I did this 20 years ago." Tidyverse has a tendency to promote "new" things without any references to what came
531.
by kkoncevicius 7y ago
There is a book I've read at least 6 times: "The problems of philosophy" by Bertrand Russell. And for a completely different turn: "Revolt against the modern world" by Julius Evola. The first book in my mind is a gr