Why are recommendation engines always about consuming more and more?
With attention becoming a precious scarce resource, apps should stop limiting themselves to "you'll like this, add this, and this, and this" suggestions, and instead also give smart suggestions for things like "you don't seem to be enjoying this podcast, remove it" or "this RSS feed content seems crap, stop wasting time on it".
re: Attack Vector for FOSS
@blinkinell@koyu.space @seek
It's not that simple. It's easy to say to just ignore them or to laugh it off, but that's not something that most people can just do. Plus it's annoying work to tag things as 'wont fix' all the time, and people will criticize you for doing it as well.
And it's not always clear who is stupid or could be trolling. Sometimes even reasonable people might take the side of the "idiots" who are asking too much, because "asking too much" is subjective.
Anyway, my point is, it's complicated.
Attack Vector for FOSS
It's sad that rude people make, unjustified demands of the open source software developers.
gemini://gemini.susa.net/attack_vector_for_foss.gmi
Creator of Clojure has expressed his view on this,
"If you have expectations (of others) that aren't being met, those expectations are your own responsibility. You are responsible for your own needs. If you want things, make them."
I looked at the Erlang part of Elixir's source code to figure out how Elixir supports definitions inside for loops, since in theory it seemed like that would require mutable state. It turns out that they basically *do* use mutable state. More specifically, they use ETS ("Erlang Term Storage") to store function definitions during the compilation of a module.
I've been having some fun with Elixir. The macro system is really great, and it synergises well with Elixir's pattern matching. You can see it in how they implement unicode support directly in Elixir itself:
https://github.com/elixir-lang/elixir/blob/v1.10.4/lib/elixir/unicode/properties.ex#L271
Basically they load standard unicode.org text files containing data about unicode code points, and then loop over the relevant code points and generate a function clause for each one, all this at compile time.
You can see it in action in the link, in the function do_trim_leading, which removes leading whitespace: the list of whitespace is loaded from lines 275 to 294, and the function clauses are generated on lines 302 to 304. Line 306 contains the clause for when it's not preceded by whitespace.
regarding a gemini post
@SEEK Thanks!
I might have news in a week or so, since it hasn't been given to me by the boss as some antagonistic scheme, but rather by an employee who was handling this agreement for new hires and who noticed I had never been given one of these to sign. So there's no pressure to sign it quickly and I have time to look it over and discuss it with others.
The boss has generally been pretty reasonable, but I've also heard that this company once was significantly larger before some employees left to form a competing business, so the boss might not be as reasonable as usual on this issue. I think my amendments are reasonable and do still protect the the boss from such things, but who knows what they'll think, it's entirely possible they'll just say to take it or leave it.
Anyway, gotta go work at that place now, have a nice day!
@Iambharani That is very cool!
For lesswrong links, a useful mirror is greaterwrong, accessed by just change the "less" in the url to "greater". It loads much faster, and you can read the comments even if you have javascript disabled.
The link I just boosted, but changed to greaterwrong:
https://www.greaterwrong.com/posts/mXgsd5o9uuYaQKHMz/tools-for-keeping-focused
Tools for keeping focused - LessWrong 2.0
https://www.lesswrong.com/posts/mXgsd5o9uuYaQKHMz/tools-for-keeping-focused
@SEEK No worries, it was clear from the context :)
@SEEK Gemini doesn't have any support for that unfortunately, since it's more or less just plain text.
The plain text is unicode, so in theory you could use some of the unicode superscript² characters, but I don't really recommend it.
What I'd recommend would be to use ^ and _ to mean "superscript" and "subscript" respectively.
@tarixa I agree that LaTeX is not so user-friendly, though as you say it's useful for putting complicated formulae together!
Though despite being glad it's here I haven't used it at all yet lol
@wistahe I didn't know modal logic before now, but looking it up you're right, it's very interesting!
@wistahe What are you studying in formal logic? I agree that formal logic is cool!
I've been hesitating to post a link to my gemlog for no good reason, this is it:
@wistahe Personally I would enjoy this
they/them