Chat Logs

  1. choicehi! if i write a stand alone java program and make sure there are no cyclic references and references to unused objects are removed, then reference counting would be enough for memory management. will the GC be required in this case at all?
  2. dreamrealdepends on what you mean by "required."
  3. dreamrealYou can write a java app such that a ZGC works fine for you, sure. But do you want to, and why
  4. dreamrealin general, the juice ain't worth the squeeze.
  5. choicebenefit of java and not another language, and real time.
  6. dreamrealWell, the java stdlib isn't going to run for ZGC, so you're going to miss out on most of the reasons you'd want the JVM in the first place
  7. dreamrealand RT isn't hard to do even with a GC
  8. dreamrealjust don't use a stop-the-world, or keep heap small enough that pauses are short enough to guarantee response time, and tell the GC not to exceed those constraints
  9. dreamrealthe only way THAT gets difficult is if you need massive datasets
  10. jbosmansthey landed safely \o/
  11. jbosmans1.2M watching, not too bad
  12. dreamrealchoice: what does this app do? What about it needs to be RT?
  13. choiceit's just an academic exercise for the beauty of it. nothing in particular.
  14. dreamrealWell, if it's a pathological desire for ZGC, godspeed and a' that. In the real world, that sort of requirement is for the micro profile, which ... has very limited support for GC if at all. Even "realtime games" in Java don't need to avoid allocation; they rely on smaller heaps and quick eden cleanup.
  15. dreamrealIn practice, GC is a burden mostly for apps that have monstrous memory allocation spikes.
  16. dreamrealOr with "difficult" programmers, I guess. :D
  17. jbosmansartemis touched down aok
  18. choicehave you ever worked on a larger project where there was not at least one "difficult" programmer? ;D
  19. whaleythere's no such thing as a programmer that isn't difficult in some way. We all have our idiosyncracies that make us intolerable over the course of time on a shared project
  20. dreamrealNaturally, you're wrong. The "difficult programmer" comes from one or both of two poles: the subject and the observer. One or both can be emphatic and stubborn, judgmental and insistent: that's when you have "a difficult programmer." If you want to get along and you're intentional about it on both side, there's no difficult programmer.
  21. dreamrealIf you want to be a difficult programmer - or you accept being a bit of a jerk - there you are, you're going to win that battle. if your goal is to NOT be a difficult programmer, well, it takes two to tango, as they say - maybe the other programmer's a hosehead. But that's why you *don't hire jerks*.
  22. dreamrealOf course, I say that as I just wrote an internal article about team trust and dynamics for work myself, because of an argument and criticism over practice where I am *clearly* in the right.
  23. dreamrealI may *publish* that article eventually, but I need to make sure I can write it without accusing anyone, or complaining, or exposing internals.
  24. GreenResponseis javachannel.org down or something? can’t reach the backlogs
  25. dreamrealif you're looking for logs, that'd be javabot's site or bytecode.news (nevet logs the channel as well), not javachannel.org, but that site is unreliable these days
  26. dreamrealhttps://bytecode.news/logs?provenance=irc%3A%2F%2Flibera%2F%2523java&day=2026-04-11 works, but javabot's logs are basically the same dataset
  27. nevetbytecode.news
  28. GreenResponseaha, thanks
  29. dreamrealif you use bytecode's logs, let me know how it looks/acts/could be improved, please
  30. dreamrealthey're both boths, generally here at the same times, so the log contents should be *really* similar, but eh, YMMV
  31. dreamrealthey're both BOTS, sorry
  32. dreamrealdoing my taxes, distracted
  33. wazclaude will do them for you
  34. dreamrealthat's just what an AI would say
  35. GreenResponseAI, like, “hold my nvidia’s, this will be a cinch”
  36. dreamrealwaz: that's a really good idea, honestly, though, and can you resummarize the cookie recipe? :D
  37. whaleydreamreal: you probably missed the nuance of "over the course of time"... familiarity breeds contempt. For instance, I find you mostly insufferable in here mostly because I've been in here with you for... 20+ years now?
  38. whaleycase in point... of all of the people in this channel, I think you are the only one that would actually start a response with "naturally, you're wrong"
  39. GreenResponsethis is an old “pecking order” scenario and what certain Adolf Hitler said once “the might is right”. dreamreal is kind of boss here so he can boss around
  40. whaleyweird, but ok
  41. wazwait, he's the boas!?
  42. wazs/boas/boss
  43. GreenResponseno, *technically* cheeser is of course
  44. ParaI don't remember having salary discussions with him.
  45. GreenResponseoh, and CodeGeek
  46. ParaAnyway. There's also a level of importance/care/equality which is very subjective in these things. To me dreamreal's approach is "ok boomer (but secretly snirking at it)" a bit.
  47. ParaBut then again I decided some years ago to mostly not care about things and just go with the flow.
  48. Swayzenot caring is giga
  49. Swayzeesp when you're 12
  50. dreamrealGreenResponse: what?
  51. GreenResponsesorry, maybe is was a bit over the edge
  52. dreamrealand whaley it's not my fault that I'm the one with the confidence to say "I think you're wrong and here's why"
  53. dreamrealand I am definitely not "the boss" in here, or anywhere, really
  54. dreamrealI'm not smirking at any of it: to me, it's diagnostic in nature. Is there a problem, what's the problem, why is it there, what can be done about it, what should be done about it?
  55. dreamrealif there's no problem, great: why do people like me think there IS one? Maybe that's the problem.
  56. dreamrealMaybe we're a small set of people: I don't know. I'm a member of the set that thinks it's a problem, so maybe our perspective is flawed. Feel free to illuminate our ignorance.
  57. dreamrealJust don't be dumb about it, and ifyou're going to put a stake in the ground, great: own it.