I'm a fan of boolean function names ending with ?, but I saw some Scheme today (define (filter ? xs) ...). Totally confused me at first!
miniblog.
Excellent candid description of the experience of creating a programming language:
Awesome Emacs package of the day: adjust-parens https://elpa.gnu.org/packages/adjust-parens.html It lets you simply use TAB and S-TAB to change paren nesting!
The Surprising Subtleties of Zeroing a Register https://randomascii.wordpress.com/2012/12/29/the-surprising-subtleties-of-zeroing-a-register/ (modern superscalar CPUs are like JITs, happy paths are not obvious)
Another great post on Guile optimisation, this time discussing different lambda optimisations:
How many parens are too much? Is relying on ordering 'less lispy' somehow? Could lisps use even fewer parens? [2/2]
The author of O-expressions argues that Clojure style lets are a compromise, using position rather than parens for structure. [1/2]
Google exploring Qualcomm chips for its data centres: https://www.bloomberg.com/news/articles/2016-02-03/google-said-to-endorse-qualcomm-s-fledgling-server-chip-efforts-ik6ud7qg (great to see diversity in the architecture market)
The Story of Squeak, A Practical Smalltalk Written in Itself https://www.cosc.canterbury.ac.nz/wolfgang.kreutzer/cosc205/squeak.html (builds a self hosting impl then translates to C)
Scheme and elisp feel like very different languages. Scheme has TCO, continuations and hygienic macros. Elisp is way more procedural.
Is Sound Gradual Typing Dead? explores performance overheads of gradual typing. Typed Racket is improving however: https://news.ycombinator.com/item?id=11053861
Blogged: An Industrial Grade BF Compiler:
A statically compiled hello world is smaller in @rustlang than in C!
Showing 811-825 of 922 posts