fwiw I tried to play around with the standalone jar version by following docs/setup.md, but it ignored the db username env var and instead tried to connect with my linux username (`db.clj:18` appears in the stacktrace).

thanks, I'll get on it

time to get into Stafford Beer mindset

I read his books. he got that old norbert wiener energy w/ the hierarchical algedonic learning bit

it must be some jdbc.next issue, passing {:jdbcUrl jdbc:} instead of the config spec in db.clj and migrations.clj works flawlessly.

yeah, I'm going to shove that in after breakfast. I could not repro on my mac box but could on my amazon linux box

fix on master, although I plan to only cut releases on thursdays

respatialized is referencing project cybersyn, by the way, https://en.wikipedia.org/wiki/Project_Cybersyn - stafford beer tried this in the 70's, got a country (chile) to adopt it, richard nixon was pissed, supported pinochet who had the president surrounded by death squads holed up in the mansion until he killed himself, exiled beer, tortured and exiled the chilean dude who interfaced w/ beer and the chileans (cff flores, who much later did a thing trying to get peeps to do business conferences in world of warcraft...)

I'm not too impressed by the underlying algorithmic content of beer's ideas, but that's because you shouldn't be overly impressed by any 1970's AI work, to be honest. it's closer to the connectionist point of view than most, i guess, but he neglected the importance of distributed representations. but nobody who wasn't a weenie was really convinced of the absolute importance of distributed representations until the ... 2000s or so, lol. maybe even the 2010s

https://the-santiago-boys.com/ Evgeny Morozov recently did an oral history of Project Cybersyn for those interested in the history in podcast form.

0% technical, i'm afraid, tho

honest question - is this a real project, or an elaborate joke?

the allocator currently works, if with a huge menagerie of known bugs and prolly a bigger one of unknown, and I spent 2 months debugging that, so not too much of a joke

there's a novel nonlinearity and i relearned renormalization group again for the purpose. the way i got it to not saturate bothers me in its weirdness but its... workingness

I do understand the boopfinkling and the zoopminkling does seem to be a bit much

Thanks for providing! And thanks for the stellar support level you attach to it! 🙏 ❤️ ⚡

Where is the fulcro example ?

@U01SBSXRRH6 added link to announcement, thx