I worry about the traditional computing term 'augmented human intellect'. If I wear a watch, I'm an augmented human: I can track time more accurately than I could without technology!
What's a better goal for tech regarding how it can help us achieve things?
miniblog.
I'm curious to see what impact Microsoft's purchase of GitHub has on the long term prospects of Atom.
Atom is a lovely editor pushing the boundaries of web tech amd editor design in several areas, but its usecase is similar to VS Code. Would MS want to fund three editor teams?
These days I struggle to distinguish between machines using Xorg and those using Wayland. I think that's a testament to the compatibility of Wayland and the efforts put in by the Qt/Gtk folks.
Some interesting insights on early Emacs in these annotated emails from Guy L Steele: https://gist.github.com/lispm/8c6783be2f5a3d41b7592ba3b2c453ea
The name was picked in 1976, and a number of concepts that still exist today were created! It mentions the kill ring, yanking, and even the same keybindings for them!
A thoughtful post on the limitations on deep learning: https://arxiv.org/pdf/1801.00631.pdf
I think the author's criticisms of deep learning's coverage in the media are fair, and that we should consider it another tool in the toolbox. Comparing it with AGI seems like a very high bar though.
The more I play with LambdaMoo, the more fascinated I am with collaborative programming. There is no separation between chatting and programming! It's an amazing way to learn.
A really fun function name in Common Lisp: the. It returns the value passed in, as well as checking the type. https://www.lispworks.com/documentation/HyperSpec/Body/s_the.htm#the
It reads rather nicely: (the integer foo).
Good, candid assessment of the state-of-the-art in self-driving cars, and the challenges ahead: https://www.wired.com/story/self-driving-cars-challenges/
Woah, Microsoft is discussing acquiring GitHub: https://www.zdnet.com/article/microsoft-has-been-talking-to-github-about-possible-acquisition-report/
Also mentioned in that LWN article: https://lwn.net/Articles/754779/
https://bugs.python.org/ is still on Python 2!
parinfer-mode is a really fun way to edit lisp code: https://github.com/DogLooksGood/parinfer-mode
Things like sexp convolute are just unindent, swap lines, reindent. Convolute is much easier to show than describe:
(when foo
(let (bar)
(baz)))
to:
(let (bar)
(when foo
(baz)))
A lovely article on how automation in warehouses can end up *creating* jobs: https://www.inc.com/associated-press/e-commerce-automation-robots-create-more-jobs-amazon-effect.html
Twitter limits the number of users you can have with an API key. Netflix disabled its API. The web concept of mash-ups seems to be very rare today.
Are there any major services that still consider their API an asset worth maintaining?
If you're willing to pay significant computational overhead, what multiprocessing model would you choose to make coding as easy as possible?
The simplest model I've seen is actors passing immutable data, but I feel there must be other options.
Cute article on how backspace (referred to as "rubout") interacts with parsing in a lisp interpreter, and how that affects the behaviour of literals: https://www.nhplace.com/kent/PS/Ambitious.html
Showing 2,731-2,745 of 7,508 posts