dreamrealHow do you CANCEL a person? What kind of return from that is possible?
jreicherI think it's a way of disguising an ongoing ad hominem attack as something more acceptable.
blueThe funny things is that Dems have been nearly forever saying "freedom of speech doesn't mean freedom from consequence". That hit them HARD after Charlie Kirk.
bluething*
blue(In fact: freedom of speech actually DOES mean freedom from consequences... that's the point, but Dems are too thick to get it)
dreamrealWell, it's funny what happens when you can't restrain your glee. But no, it doesn't quite mean freedom from consequence: it means you might have the freedom to say something and that something might actually serve as a lever.
dreamrealBut you DO have the freedom to be a jerk... and be seen and treated like a jerk.
blueI think Dems are absolute hyenas for mocking his death, but should they lose their jobs over it? not SPECIFICALLY; I mean, if I were their employer I'd honestly ask them if you feel glee at the death of another person, and if they do, I'd strongly consider terminating them on account of differing values
dreamrealOr an evil person. It's not isolated to the dems, it's just that right now the GOP is ascendant so they're crying a lot more than usual.
dreamrealblue: yeah, well. That's the dildo of consequences! It's hard to lube.
blueThis isn't consequences per se: this is me discovering a side of you that I didn't know exist, and that I don't know if I want to interface with anymore
blueregardless of that, I don't know who's stupid enough to glee over someone's death on SM with his real name
blueI mean, you gotta understand that's gonna come back at you, SOMEHOW, SOMEWHEN
dreamrealWell, plenty of examples desperate to make sure their friends know they're the GOOD ONES for celebrating the RIGHT MURDER
dreamrealmidwits thinking murder's surely abstract, we see enough of it on TV
blueand then you get your typical "this guy was for guns so it's cosmic justice". is THAT how we do things now?
dreamrealwell, it's how they think
dreamrealthat's the corrosiveness, everything's reduced to soundbites because they can't hold two things, especially contrasting things, in their heads at one time. Again, not a democratic problem, just a problem they seem eager to display because stupidity is an easy disability
dreamrealsorry, democrats, I don't lie about what I think easily
dreamrealI'd rather someone recoil in horror at something I say if it's a true representation
dreamrealbecause if it's horrible and I think it, well, show me where I'm wrong
dreamrealplease
dreamrealtikkun olam is important to me, to provide as well as receive
blueand next point is, "I didn't agree with the guy, but he didn't deserve to die"
blueWHY do you need to disclaim THAT?
bluenot wishing people die shouldn't be contingent upon agreement of views, you're creating an evil link
dreamrealWell, not only that. But the funniest thing is that when you ask them what they actually disagreed with, it's someone else's summary of something he didn't quite say
dreamreal"he said two paragraphs, the short version is 'guns yay'" when, uh, no
blueyou making it sound like this: "if I only said I didn't deserve he died, then I agree with his views"
blueah... no?
blueyou're just saying you don't think people should be gunned down
dreamrealHis defense of guns was pretty in-depth and he did in fact say some things that are unpleasant about tradeoffs
bluehe didn't dserve*
dreamrealblue: yeah
dreamrealI mean, I think neither he nor anyone else should be murdered, although there are exceptions to the "I hope they don't die" statement
bluethen point is, such a dislaimer is basically you talking about yourself: you're saying: I don't want to be associated with this guy... but I still don't wanna come across like a jerk. but congrats, you just did
dreamrealI still can't celebrate bin laden's death, or sinwar's, but ... is the world really worse off since they found their way to sheol? uhh... no
* dreamreal!~dreamreal@about/java/dreamreal changed the topic to: This is the channel for the development of the software that runs nevt: https://github.com/jottinger/bytecode.news
* dreamreal!~dreamreal@about/java/dreamreal changed the topic to: This is the channel for the development of the software that runs nevet: https://github.com/jottinger/bytecode.news
bluenow I gotta wire it up, let's see how well that works
dreamrealwoot!
dreamrealI'm going to work on a feature to expose whether OIDC is configured or not in a few minutes, fixing an IRC problem
bluenice
dreamreal(Working on 134 now, 128 is next)
* nevet joined #nevet
* dreamreal!~dreamreal@about/java/dreamreal changed the topic to: This is the channel for the development of the software that runs nevet: https://github.com/jottinger/bytecode.news
bluedreamreal: this is exciting; I get redeployment times for the primate app of <10s. I still need to optimise the nevet Containerfile, but I think I can get it to redeploy quite fast, hopefully
bluedreamreal: can we add a GET /version endpoint that is not auth'd?
bluethat would be the fastest way for frontends to test connectivity
dreamrealis there one that IS authed now?
dreamreal(the answer is, of course, yes, but I need to understand scope)
blueno. I don't think there's one at all
dreamrealfile an issue - actually, the feature endpoint (for 128) might serve.
bluethere's a /feature endpoint?
blueoh, it's an open item
dreamrealNot for the next ten minutes, at least!
dreamrealI mean, I'm working on it AS WE TYPE right now
dreamrealwill expose what services and operations are in the runtime, as well as OTP, OIDC settings (like "are they there," not "how are they configured")
blueI don't really care about WHAT it does for NOW, only that I can access it unfettered
blueand also, I need to figure out how to redeploy nevet itself
bluethat being said, phaleth did an excellent job on this one, as always
dreamrealI'm working on it, and not being gated is *definitely* a hard nonnegotiable requirement
bluewdym with gated?
dreamreal"no security requirements for /feature"
dreamrealor /features i guess
dreamreali.e., what you asked for!
blueI asked for a simple GET route that I don't need anything for
dreamrealyep
dreamrealexactly
dreamreal100%
blueand that is a problem.. because?
dreamrealit's not a problem
blueoh, ok
dreamrealIt's a hard requirement
dreamrealas in, the feature does not work if that isn't met
bluemy point was only, that this GET route would satisfy:
blueAuth: None
blueyou already have Auth: None routes. just no GET
bluebut I think we're talking about the same thing, anyway
bluelike 90% we're too smart for each other, this is a real problem
blueanyway, back to how to redeploy nevet fast
dreamrealI'm almost done with this
dreamrealand this will be GET /features
dreamrealIt's very useful because things like this expose some inconsistencies in the adapter implementations
dreamrealmost of them are really minor: things that can be corrected with a few lines of code
dreamrealand nearly all because of the rolling nature of the design
dreamrealservice-blog is naturally the worst offender :D
dreamrealthat actually does reflect the state of the instance that was run against
* nevet joined #nevet
* dreamreal!~dreamreal@about/java/dreamreal changed the topic to: This is the channel for the development of the software that runs nevet: https://github.com/jottinger/bytecode.news
dreamrealversion
nevetnevet 1.0 | 90f413a (main) | Built 2026-02-28 12:15:36 EST
bluenot entirely sure which commit that is supposed to represent
dreamrealversion
nevetnevet 1.0 | 90f413a (main) | Built 2026-02-28 12:15:36 EST
bluenote that I'm purely rebasing against your branch
blueI'm not rewriting your history, so I think I should be seeing 90f413a as well...
bluemaybe I messed up something
blueare you sure the features endpoitn shows correct data?
dreamrealif you're on a differen repo, though
bluemy repo's history is the same as yours, just with a few commits on top
bluehow are you getting the commit info?
dreamrealif you have commits, that'd be it
bluemy last commit is 5cef64e67be89cda0ebf16b270f35549176e5a6c though...
bluemaybe it's some commit inbetween
blueI'll rebuild nevet
bluehm, now it's running the maven phase again, damn
blueI did change it to -DskipTests=true
bluenot sure how to cache it
blueok, now it's 5cef64e, yippie
bluethe maven build took 100s
dreamrealwell, the *actual* way you should do it is run mvn outside of the container and just copy app.jar in
dreamrealbut nooooo everyone's like "docker is the best, why not use docker to build, it'll be great"
dreamrealbut yes, there're lots of modules to go through, it's a pain
dreamrealand if I ever do native builds it'll be WORSE
bluethat way is never gonna be properly deployable
blueI'm gonna try -T 1C
bluein-container build should be just as fast with nspawn
* dreamreal nods
bluedreamreal: -T 1C got it down to 66s
blueI'm gonna build again, just to see if that's anywhere reliable or just a fluke
blueI mean, the first build was 130s. the one with the m4 cache (no redownloading of mvn packages) was 100s. now we're at 66s. we're getting *somewhere*
bluehm, this was fast, 4s. so probably everything cached
blueI'll try pushing out a git commit
blueyeah, a single commit, even if it's inside apps/primate, totally busts the mvn cache
bluewhat a dumpster fire
dreamrealWell, don't copy in the user stuff then!
bluewhat user stuff?
dreamrealcopy the pom.xml, app, service*, operation*, lib*
blueI just do `COPY repo/. ./`
blueai said it's a multimodule maven build and that I should copy everything
blueand since I have no idea how this works, I listened to it
bluewhat's it matter what I copy, anyway? shouldn't the builder just build what's necessary?
dreamrealBecause of the way docker layers work
dreamrealwhen you change a layer, the cache is invalidated
dreamrealthe AI is right WRT maven, wrong because you're changing bits of it
dreamrealso you copy in WHAT THE BUILD NEEDS and nothing more
dreamrealthe maven parts are ./pom.xml, app, lib*, service*, operation*, I think
dreamrealI've tried to follow a rough convention for them all
blueheck, I'm not changing ANYTHING
blueI pushed to apps/primate, which is no java code, at all
blueand it still busted the cache
dreamrealright, if you change the filesystem in the docker image, that's the cache broken
dreamrealso you copy in WHAT THE CACHE NEEDS, nothing else
dreamrealit is known
bluesomehow the cache isn't broken for npm...
dreamrealdunno what to tell you, man
blueI think it's a maven issue, not a docker issue
dreamrealthis is pretty standard for the OCI stuff
bluethis is somewhat beyond me, but luckily, I don't need to redeploy nevet itself SO often
dreamrealI'll have a change for you soon! factoid updates.
* dreamreal!~dreamreal@about/java/dreamreal changed the topic to: This is the channel for the development of the software that runs nevet: https://github.com/jottinger/bytecode.news
dreamrealidea
nevetIDEA is one of the "big two" Java IDEs, and probably the most popular of them. Has a free and commercial edition, and has capability to work for many, many, MANY languages. URL: https://jetbrains.com/idea Tag: ide
dreamrealidea.stats
nevetidea has been accessed 1 time, last accessed at 2026-02-28T18:18:11.940047Z.
dreamrealidea
nevetIDEA is one of the "big two" Java IDEs, and probably the most popular of them. Has a free and commercial edition, and has capability to work for many, many, MANY languages. URL: https://jetbrains.com/idea Tag: ide
dreamrealidea.stats
nevetidea has been accessed 2 times, last accessed at 2026-02-28T18:18:27.636040Z.
dreamrealNoice.
dreamrealOkay, that's it for today for me. Have a good one.