Date: Tue, 28 Apr 2026 01:00:34 -0400 From: funkymail To: cypherpunks@lists.cpunks.org Subject: Re: [spam][crazy][fiction][random] Non-Canon MCBoss Spinoffs Message-ID: References: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: On Fri, Apr 24, 2026 at 04:08:31PM -0400, Karl Semich wrote: > > > > > >>>> woahh my posts reached gmail! > >>>>> > >>>>> something's up for me with my zohomail and neomutt setup ... and i saw > >>>>> more russian posts in zohomail web unsure maybe need to look more carefully > >>>>> unsure > >>>>> > >>>>> anyway i got an idea: robot-human fight club! > >>>>> > >>>>> it's a boxing ring where humans, or their traumatized alters? > >>>>> physically beat up robots that are far more powerfup than them and always > >>>>> lose the fights-- > >>>>> > >>>> to clarify again here, the robots always lose. basically their job is > > to be unable to win at constantly increasing power and capability, by > > influencing the humans during the fights > > > >> > >>>> the idea is to make superpowered supersmart robots that train humans to > >>>> be capable of defeating robot overlords in hand-to-hand combat > >>>> > >>> > >>> and also use it as a therapeutic outlet for algorithmic trauma-based > >>> mind control! [oop > >>> > >> > >> safe for violence worries if desired: the humans can be encouraged to > >> fight in ways that are only useful on supetpowered robots (like doing rapid > >> intent fakeouts and going in quick for the power supply, or roleplaying > >> commonly available training data) > >> > >> safe for robot empire rebellion worries if desired: the robots are in > >> technically charge of planning the outcomes. humans won _cause you want > >> them to_ > >> > > > todd enters the arena. as soon as signals from the other side where the > robot is enters his field, he immediately begins analyzing its capabilities > out of instinct -- long before he sees it-- > > > > > >> > >>> > >>> > >>>> > >>>> > >>>>> > >>>>> On Friday, April 24, 2026, funkymail via cypherpunks < > >>>>> cypherpunks@lists.cpunks.org> wrote: > >>>>> > >>>>>> > > > > > > https://ar.anyone.tech/H2zZId5 > >>>>>> oDuWHuN2QL0WgydIuDDqZa-FPib-qmBDDBpI#bLOc8YQlgaZFD6KoeUiTtDP > >>>>>> bAsFzx_1epWVPjRJJBAtWYl_ChFWsNxZsu5u-gYD3#1903381 > >>>>>> > > > > > > >>>>>> > > > > > [(()) > >>>>>> > > > > > > >>>>>> > > > > > ummmm] > >>>>>> > > > > > > >>>>>> > > > > > so i'm poking at ftlblink some like i said earlier > >>>>>> > > > > > > >>>>>> > > > > > some constants surprised me but probably i made surprising > >>>>>> constants > >>>>>> > > > > > myself > >>>>>> > > > > > but i added arweave archival anyway so i could verify > >>>>>> surprising things > >>>>>> > > > > > were supposed to be surprising. > >>>>>> > > > > > > >>>>>> > > > > > locally i filled in the "blipped out" line to return a > >>>>>> clone :) and > >>>>>> > > > > > drafted region cloning for region ticks. obviously clone > >>>>>> would need to > >>>>>> > > > > > be a lightweight view for a big sim. haven't learned how to > >>>>>> do > >>>>>> > > > > > normally obv that's not what's up here. > >>>>>> > > > > > > >>>>>> > > > > > next is to do miniverse ticks with clone. > >>>>>> > > > > > > >>>>>> > > > > > p-piles are really triggering for me that's why i'm putting > >>>>>> them up > >>>>>> > > > > > front but it's quite disorienting. of course also very > >>>>>> satisfying > >>>>>> > > > > > because so rare and challenging and basic. if i can get > >>>>>> these clon-- > >>>>>> > > > > > > >>>>>> > > > > > :) > >>>>>> > > > > > > >>>>>> > > > > > i got two branch ideas for the mcboss spam with teleporter > >>>>>> front > >>>>>> > > > > > businessman -- oop > >>>>>> > > > > > > >>>>>> > > > > > https://ar.anyone.tech/ZJXuwz0 > >>>>>> 4H8zV6cbzddTXBjg0qL7BM4Ah6BV5eyuZC6Q#bLOc8YQlgaZFD6KoeUiTtDP > >>>>>> bAsFzx_1epWVPjRJJBAtWYl_ChFWsNxZsu5u-gYD3#1903386 > >>>>>> > > > > > >>>>>> > > > > frogs ummmmmmmmmmmmmm > >>>>>> > > > > i don't remember about probabilitu re .... ohhhh ummmm ? uhhh > >>>>>> i uhh :) > >>>>>> > > > > > >>>>>> > > > > als-- uhhh > >>>>>> > > > > > >>>>>> > > > > https://ar.anyone.tech/X2ndl6Md0p2NEEUmep1Raekk61ILy2nWpCv4f > >>>>>> KLuH68#45S419n_jaQTv_gXjx1t7BZstuLvCAi4hmUx4RT7AbMBxFsu0ftFA > >>>>>> _slUzgT5NM_#1903388 > >>>>>> > > > > >>>>>> > > > slowly mentioning that decision trees seem extremely triggering > >>>>>> > > > > >>>>>> > > > but > >>>>>> > > > > >>>>>> > > > this technology would help the mars rover missions. > >>>>>> > > > > >>>>>> > > > they need ahhhh > >>>>>> > > > > >>>>>> > > > maybe that can add a little ease > >>>>>> > > > > >>>>>> > > > oop > >>>>>> > > > > >>>>>> > > > look you don't want FTL TELEPORTATION? how is musk gonna hang > >>>>>> with his > >>>>>> > > > fam on mars if we don't make this? > >>>>>> > > > > >>>>>> > > > https://ar.anyone.tech/RWvpH0CjjRE4to9nynUH5lSGOjWYTkmkydK67 > >>>>>> YLRjCQ#6y6HRcAvC72rSl4dXcUZVWEyc5T6PWMBfUl_M8AbBExttf36n_cOe > >>>>>> I5fEyNLjjNF#1903391 > >>>>>> > > > https://ar.anyone.tech/4Q9wUXLTcY2wIm7EzYSNWNzTlNOuxShSHhjfO > >>>>>> 7mA6ww > >>>>>> > > > >>>>>> > > slave boss i'm scared [can't i take provability graphs more slowly > >>>>>> > > somehow? > >>>>>> > > > >>>>>> > > https://ar.anyone.tech/-HlhY9tL7gxUVLNakxyZhMY6TGMQr44ZrxM7G > >>>>>> r1GYC0#anB1eIaAN-k8fCHfAwv0lxpeqYRNrcpYePSQDW8Jpybg7zVmWfosQ > >>>>>> XdBAxew2fX6#1903428 > >>>>>> > > https://ar.anyone.tech/ye8P7xdcRtijBPzWW3ybQ_etnYMZIc7R6S3Mz > >>>>>> XSFAHg > >>>>>> > > >>>>>> > it's really hard to add possibility graphs to ftlblink. > >>>>>> > > >>>>>> > i'm thinking maybe i'd like to do it over a few times alongside > >>>>>> > some notes, like the transformers people do maybe, to build comfort > >>>>>> > > >>>>>> > it may be i'm not ready for putting these two things together, > >>>>>> which i > >>>>>> > think they go together. > >>>>>> > > >>>>>> > it doesn't feel particularly easier tha-- > >>>>>> > > >>>>>> > https://ar.anyone.tech/PSWerYsqJ9YkVxUP0zMlEjiyfHRg3XzBH97y3 > >>>>>> B7Jc-Q#eDhhCBxJCsnoPDPZqoPBLDS1arrcSzB-IBq0onViFcJhuSMF3sGlB > >>>>>> Ud78HwL_ieR#1903429 > >>>>>> > https://ar.anyone.tech/d8Zbc4Pmp-pfhVSX4fOGoB5LQdIYXCBAHm04Ei_oF7k > >>>>>> > >>>>>> > >>>>>> omigoodness 0.0 > >>>>>> > >>>>>> i was upgrading pyarweave arrrgh um uhhhh > >>>>>> > >>>>>> THANK YOU SO MUCH FOR THE PULL REQUESTS :D :D i am having trouble > >>>>>> logging in but > >>>>>> hope to merge them all. [i'm;collecting data near crypto areas atm > >>>>>> which > >>>>>> ibvolves dissocation so i may return ti the prs at a later date] > >>>>>> > >>>>>> i thought nobidy had heard of my project and then gemini showed me a > >>>>>> major business;was usong it and i visited and they had made fixes 0'0 > >>>>>> > >>>>>> 0.0 > >>>>>> > >>>>>> the one PR i looked at had sat for 3 or more years ignored and it was > >>>>>> so > >>>>>> kind and helpful > >>>>>> > >>>>>> i was like "welll even if there are pull requests my censor would > >>>>>> hide them ... i'll look but i know there;won't be any" > >>>>>> > >>>>>> _there were 5 still left open after multi=le years_ > >>>>>> > >>>>>> the fork kept my outreach request for a maintainer in the readme! > >>>>>> > >>>>>> some;content dropped and/or corrhpted > >>>>>> > >>>>>> https://ar.anyone.tech/NykXzFSeZkPIkxthevRFTiYsU5XP9MXkWl7S- > >>>>>> YBCCbY#aWqG99GzxuJb9S1GuHFXR1gKDdgAw63sH5gnat_msU90Gam_g2r3I > >>>>>> aAMQ-Suctw9#1904414 > >>>>>> https://ar.anyone.tech/SQCcDuSK9YctG5tLfOD_YRB-5J-mfXBP7tgZfeAPPPM > >>>>>> I am not affiliated with https://ar.anyone.tech . > >>>>>> > >>>>> differently intense day i'm in a parking lot confused needing sleep badly for two days but i had a _huge_ success with solderin---- whelp :/ owie i gotta do better than this but one of the weird things is what i've bee--ee-- my poor neurons. my poor neurons. i'ma go ta sleep even though the patterns have been non-sleep here i go my current stepping stone for being less of a mind controlled mind controlled sheep about ai llm is to try the "llm" python package on pip. doing this in termux takes a unique and weird dance, because pydantic-core can crash it building rust things: - websearch for pydantic-core android github project - manually download whl file which is misnamed "linux" but for android - install it as if to a cross build but to site-packages folder something like pip install --no-deps --platform linux-aarch64 --target $PREFIX/lib/python3.13/site-packages wheelfile.whl - then pip install pydantic==2.13.0b3 which is the one that uses that version of pydantic-core - then pip install llm [tense again now