Suggesting fixes to students' code by using a solver to explore modifications to the program! https://youtu.be/NAoSTmCUlW0 (uses sketch)
miniblog.
I finally feel I've got a handle on org-mode and it's being useful for me. When I started, I thought it was just an agenda tool.
Handy elisp for profiling code and summing function time by function (unlike the hierarchical profiler-report): https://github.com/alphapapa/emacs-package-dev-handbook#elp-profile-macro
A lisper shares his experiences with writing a project in Smalltalk: https://emacsninja.com/posts/one-of-those-who-can-type.html (though using GNU Smalltalk, so no live IDE)
Fascinating post on being a programmer when English (or the Latin alphabet) isn't your native language: https://temochka.com/blog/posts/2017/06/28/the-language-of-programming.html
straight.el: https://github.com/raxod502/straight.el
A fork friendly, MELPA compatible Emacs package manager.
Indexing into arrays in J using complex numbers! Seems totally bananas, but provides some interesting composability.
https://ronjeffries.com/xprog/articles/j-this-is-hard-however/
Dynamically generating shellcode using an SMT solver: almost magical! https://sean.heelan.io/2009/06/01/fun-uses-for-an-smt-solver/
Amazing paper using genetic programming guided by passing/failing tests to automatically write bug fixing patches! https://www.cs.virginia.edu/~weimer/p/p109-weimer.pdf
With suggest.el, I've learnt generative programming suits interpreted PLs. The problem is exponential, so extra perf doesn't really help.
A remarkable project exploring a binary AST format for JS! https://github.com/syg/ecmascript-binary-ast (interesting comments on when you can't lazily parse JS)
Do you know how many LOC are in the codebases you work on? I see this as a common question in surveys, but I rarely bother measuring.
A/B testing designs in the world's most popular language learning app: https://firstround.com/review/the-tenets-of-a-b-testing-from-duolingos-master-growth-hacker/ (postpone signup, let users try it!)
I like Common Lisp naming its super call 'call-next-method'. The problem with https://t.co/SFdvndHQEI() is that .foo is redundant.
Porting Remacs to use Rust's HashMap is looking like a measurable performance win! https://github.com/Wilfred/remacs/issues/276#issuecomment-319273980
https://www.youtube.com/watch?v=Q-3tcbUyF34 is a wonderful talk showing how to generate whole programs from just a liquid haskell type definition with Synquid.
'In this example, we use print to represent output to web browser' — cute! If only more PLs had a GUI oriented print function.
On the value of learning what you don't know:
https://nealford.com/memeagora/2015/09/08/knowledge-breadth-versus-depth.html (increase breadth, but low knowledge maintenance)
I am now editing Remacs from within Remacs!
(To be honest, the main difficulty was porting my config for Emacs 26.)
To post a comment, you must demonstrate you understood the story: https://www.independent.co.uk/life-style/gadgets-and-tech/news/nrk-norwegian-news-site-comments-read-story-understand-post-quiz-questions-a7607246.html (cute! May be insufficient for contentious issues)
Showing 381-400 of 885 posts