Thomas Schelling would be within his rights to hand ESR a copy of "Micromotives and Macrobehavior" and demand he write an essay explaining how fucking wrong this sentence is afterwards as penance for his sins.

hypothetical pot shot:

who do you think would be more likely to cite ESR specifically (or even "The Cathedral and the Bazaar" generally) as a positive text, someone who works at Oxide or someone who works at Anduril?

hypothetical pot shot:

who do you think would be more likely to cite ESR specifically (or even "The Cathedral and the Bazaar" generally) as a positive text, someone who works at Oxide or someone who works at Anduril?

this is also definitely part of it lol

the deeper motivation behind CatB, to the extent that i believe one exists at all, is to serve as a citation for software libertarianism.

flattening the complexity of planning and experimental development into a binary choice between an ornate Cathedral or a pragmatic Bazaar.

i don't think there's necessarily any sort of malicious subtext in CatB itself, but it's always struck me as a collection of fortune cookie aphorisms that are very narratively satisfying if you're the sort of person who takes a very counter-cultural view to successful software engineering.

the closest principle i can come up with for this is a lack of intellectual depth, where an argument feels persuasive to one's sensibilities (in the case of ESR & DJB it's often a sort of libertarian, anti-establishment flavor) that they expend no effort in understanding what motivated the author.

there’s no reason to believe that all forms of this technology can be bent towards liberation, some of this stuff is inherently oppressive!

i agree with the thrust of what you’re saying, dgmw, but a lot of damage is going to be caused before the market sorts itself out.

even if this doesn’t come to pass (i think it’s a false premise of the technology!), the atmosphere is making a lot of people miserable.

some of the people most deeply enthralled with this stuff have plainly said that they think aesthetics are the only thing that matters “now that we have AGI”.

so a scary thing about the current moment is that a large proportion of software engineering thought leadership is aggressively pressing the narrative that you can commodify the foundational knowledge that defines skilled labor.

this is a common comparison for people to make but i think it‘s a little disingenuous.

i think the existence of mechanization doesn’t necessarily imply the death of skill, but a lot of LLM hype emphasizes that foundational knowledge will no longer be necessary.

do you think if roon’s posts get bad enough OpenAI fires him yacine style?

back of the envelope, but wouldn’t £12k/yr give you like £150k - £225k after 10 years if you assume a fairly conservative 5-11% return on investment from stuffing it in an index fund?

aren’t ISAs tax exempt investment vehicles? this strikes me as a phenomenal amount of money for a 20-something.

ideally this is what education is supposed to be (both academic pedagogy and also on-the-job training), but if people are going to use LLMs for even this part of the learning process then idk..

PartialOrd gives you the comparators doesn’t it?

i feel like Van Hollen has also distinguished himself from a lot of the others by demonstrating very clearly that he knows how to play ball

lol, saw Packer & immediately assumed Hashicorp. terminal infra brain

oh yeah, this is how i used to do it before i switched over to helix...

config module: github.com/jkachmar/tar...

separate module to package the plugins: github.com/jkachmar/tar...

wait why Packer are you generating images w/ it?