ByteCode.News
RSS
SubmitLog in

Community

Chat Logs

Monday, March 23, 2026

  1. dreamrealMaybe one more deployment for nevet - I'm setting up an MCP server :D
  2. dreamrealhttps://bytecode.news/posts/2026/03/bytecode-news-now-has-an-mcp-server
  3. javabotdreamreal's title: "ByteCode.news now has an MCP server | bytecode.news"
  4. jbosmansdreamreal, so given nevet is here, what does it do (here) exactly?
  5. jbosmanswell, "here" without parentheses i guess :)
  6. ParaIt is a learning computer.
  7. jbosmans:-D
  8. ParaIt kinda is, it's got per-user Markov chains for imitation.
  9. Paranevet | * channeling jbosmans: sun ceo but what's his name
  10. dreamrealIt's muted HERE because of javabot; two infobots in the same channel is just asking for noise.
  11. jbosmansfirst the bots get replaced, then ...
  12. ParaBack in...whenever people did run bot channels with the whole idea that they'd be allowed to chat together to see if something interesting would spawn from the noise.
  13. jbosmans:P
  14. dreamrealit has controls set so that it will never emit content in some environments (like this channel)
  15. jbosmansdreamreal, what does it ingest?
  16. ParaIf anything, those channels taught us that Markov chains do not Artificial Intelligence a make.
  17. jbosmansdreamreal, sorry, what does it ingest from here if anything?
  18. dreamrealprimarily logs and some factoid content, although the factoid trigger is different because the *syntax* is different; nevet's factoid model is a lot more indepth than javabot's.
  19. dreamrealkarma as well.
  20. ParaDoes the article poster actually post yet?
  21. dreamrealIt also gives me a chance to create content in the bot without having to change contexts.
  22. dreamrealPara: what do you mean?
  23. ParaOn a bit of a tangent, this is apparently still getting a few updates now and then https://kingjamesprogramming.tumblr.com/
  24. nevetKing James Programming
  25. dreamrealI've used !article in *this channel* to create content on bytecode.news, yes
  26. Paradreamreal: Just iterating in my head what commands were available.
  27. dreamrealhttps://bytecode.news/posts/2026/03/diataxis was written that way
  28. nevetDiataxis
  29. javabotdreamreal's title: "Diataxis | bytecode.news"
  30. dreamrealhttps://github.com/jottinger/bytecode.news <-- the actual repo
  31. javabotdreamreal's title: "GitHub - jottinger/bytecode.news: The nevet codebase, generally for bytecode.news 路 GitHub"
  32. dreamrealnevet is running the app module, so it's got everything enabled that it's in the codebase (you can disable features by ignoring their modules in the classpath)
  33. dreamrealI think the only thing that ISN'T supported that the repo has is OIDC, because I just haven't tested it
  34. ParaAh, the nifty self-registration which people still underuse IMO.
  35. dreamrealyeah, I don't store passwords if I can help it
  36. jbosmansdreamreal, i guess i want to understand what it's doing with things people post here?
  37. dreamrealjbosmans: very little
  38. dreamreallogging, more than anything else: it's not doing ingestion of anything more detailed than javabot is
  39. ParaThe Markov chain is probably the most involved thing, and it's a curiosity more than anything.
  40. dreamrealit's tracking karma from #java, I guess, but most other interactions are specific and via privmsg
  41. jbosmanslogging like, writing things to a file and that's it ?
  42. dreamrealeven markov is just ingesting from logs
  43. dreamrealdatabase, not file
  44. dreamrealbut yes
  45. jbosmansokay, cool
  46. dreamrealthose logs ARE publicly available, just like javabot's
  47. jbosmansright, yeah
  48. dreamrealI think it's better than javabot for most things (there's like ONE feature javabot has that nevet does not) but javabot's not something I'm trying to replace: I want to be ABLE to but I'm not seeking to replace it
  49. dreamreal(the feature: API search.)
  50. jbosmansdon't make us redundant eh
  51. jbosmans;-)
  52. dreamrealWell, if you know anything about markov chains, you'll know that's unlikely. The markov chain in nevet uses three-token sequences, so it's recognizable but not, like, sane :D
  53. dreamrealI may bring that up to four tokens, but that requires a lot of inputs. Maybe when nevet's been running a year.
  54. dreamreal(at four and five tokens, you're starting to see it just straight up quoting people.)
  55. ParaAnd it'll take time to get the chain to produce any meaningful variation anyway.
  56. ParaBut it's a kinda fun toy.
  57. dreamrealPara: well, it depends: the markov chain is *really* fast
  58. dreamreallike, retrieving the logs is the most expensive part of the whole thing
  59. dreamrealand that's around 4ms
  60. jbosmanshaving no memory i'm probabilistically (?) reducing you're making us redundant
  61. dreamrealit's a fun toy and it's related to the whole reason nevet exists
  62. jbosmansit's nice to have a dataset to do things with
  63. dreamreal(the first *real* iteration of nevet that ran on #java was actually not *connected* - it ran over #java logs and generated markov data from various users and ran sentiment analysis over it.)
  64. dreamrealSpeaking of...
  65. dreamreal15:57 <nevet> Sentiment for irc://libera/%23java: Sentiment +3/10 | Intensity low | Themes: bot features (nevet), markov chains, logging/data ingestion | Summary: Calm technical discussion about nevet bot capabilities; jbosmans curious about data handling, dreamreal explaining architecture鈥攖one informally positive, collaborative
  66. jbosmansnice :-)
  67. dreamrealYou could theoretically set up a timer operations and have the bot monitor sentiment for various ingress channels
  68. dreamrealand go "hey op you might want to pay attention to channel #x, sentiment is at -7 thanks to user foobaricus"
  69. jbosmansyeah that would be bad
  70. dreamrealI dunno, -7
  71. dreamrealis pretty aggressive but it's not THAT hard to trip
  72. jbosmansyeah, exactly :-)
  73. jbosmanslet's reduce interaction to an integer?
  74. jbosmansbut i understand what you mean
  75. dreamrealnah, it's actually a floating point, you normally measure sentiment in -1/+1 terms
  76. jbosmansjust trying to prove a counterpoint i guess
  77. dreamrealbut you can hit -1 pretty easily, 1 is kinda tough
  78. jbosmansi'm wondering, what does that mean?
  79. dreamrealwhat, sentiment scores?
  80. jbosmansyeah
  81. jbosmansmore like <dreamreal> but you can hit -1 pretty easily, 1 is kinda tough
  82. dreamrealOh, it's really easy to be "extremely negative" but being "extremely positive" is difficult
  83. GreenResponsefrom huffington post to full 4chan I guess
  84. Tenchiheh
  85. dreamreallike, issue a lot of death threats and be all shouty and you can get -1 fairly easily
  86. dreamrealbut be all shouty and *positive* and it ends up coming off as sarcasm and flattery
  87. ParaThe sentiment measures are always funny.
  88. jbosmansyeah, it's interesting stuff to play around with for sure
  89. jbosmansbut given where we're at (& going i guess), i just think it matters, not to dramatize :-)
  90. dreamrealnevet uses an LLM to gauge sentiment: it's not doing heuristic. I wouldn't MIND doing heuristics for it but I was specifically trying to get an LLM bridge in place.
  91. ParaYears ago one place I worked at had autosentiment analysis for customer feedback and one message they got which had like three non-swearwords (agglunative languages are very flexible for verbally creative) had a sentiment of -0.68 on scale of -1 to 1. Which was the worst score ever received, so obviously it was printed and put on dev team's wall with the sentiment analysis.
  92. dreamrealPara: hahahahahahaha
  93. ParaWhich of course begs the question, if basically a collection of swearwords is just -0,68 then what can be worse than that.
  94. dreamrealcreative use, maybe
  95. jbosmansi guess it'd depend on the model
  96. dreamrealThat's part of why I didn't want heuristics
  97. Tenchibeyond that threshold is sarcasm and passive agressiveness
  98. ParaThis was over a decade ago, so who knows.
  99. jbosmansor heuristics
  100. ParaMaybe the analyzer had general issues with said agglunation and saw it just random nonsense :)
  101. Para+as
  102. dreamrealI have a summary generator in bytecode.news that does heuristics, it's... good but not anywhere near great, the LLM summaries are much gooder
  103. dreamrealBTW, I'm considering offering a reward for people to write for bytecode.news - I'm trying to get it to have enough critical mass to replace TSS
  104. dreamrealso far, I think *I* have written everything on it but it's meant to be open contributions
  105. * dreamreal waits for someone to say "oh really, how much"
  106. dreamrealand the answer is "not THAT much, my money tree ain't bore no fruit yet" but it'd be something and you'd have an actual editor help you
  107. dreamrealfor it to be in the TSS role it needs more voices and more people propagating the URL
  108. jbosmansTSS == Therapeutic Support Staff ? :)
  109. jbosmansi'm guessing not ^
  110. jbosmansalso wouldn't go reward for posting
  111. jbosmansends in clickbait, see (alas a large part of) #journalism
  112. kcomhnalldreamreal: interesting. I'll take a look. what's the tech stack and where is the source repo?
  113. * kcomhnall not concerned about a reward other than learning experience.
  114. dreamrealkcomhnall: https://github.com/jottinger/bytecode.news
  115. javabotdreamreal's title: "GitHub - jottinger/bytecode.news: The nevet codebase, generally for bytecode.news 路 GitHub"
  116. dreamrealTSS = theserverside.com
  117. dreamrealand if someone posted something I could use, I don't mind compensating them for it, you make the iron hot by striking it
  118. kcomhnalldreamreal: cool, I'm bookmarking the repo now. right now I'm trying to figure wth is going on with my own JWT_SECRET environment variable (Windows) and why it's not being picked up.
  119. jbosmansfinished moving away from thymeleaf layout dialect \o/
  120. jbosmansno more groovy = jar file 8mb smaller
  121. jbosmansnow i can install doom
  122. johnjayspeaking of bytecode. are there any good sites about internals and guts of the jvm?
  123. dreamrealYes, they're publicly available, the vm spec is useful
  124. dreamrealit's kinda deep though