For decades, we ran in libraries, archives, and in the back of stores and banks, doing the quiet jobs most people dreaded. And this way, some autistic people managed to survive undiagnosed.
This is not an option anymore.
@_ykysnk@challenger0303@shinyflvres Which is an optional opt-out setting. If you look a bit closer, MessageRouter.cs:L423 is the handler for POST request to the VRCNext Api. the post call for ingestion happens on L358, and it has a proper check. Services.cs:L681 shows that the value for it is off by default.
food 200โฌ
data 150โฌ
rent 800โฌ
EU taxes for merchandise from japan 3600โฌ
utility 150โฌ
someone who is good at the economy please help me budget this.
my family is dying
so far what i gathered, is that the original quote tweet was somewhat along the line that notch is allegedly transphobic and antisemitic. (see https://t.co/WTgdusODfE ). With that we can assume the missing one was a Joke about notch liking jewish transexual cocks by @/ChoeNut
BRIDGET CANT SIT ON MY FACE BECAUSE SHES BRITISH??????????????????????????????!??!?!?!?!!!?!?!??????????????????????????????????!??!?!????????!??!???!?!?!?!?!?!??!!!!?!??!!??!??!?!??!?!?!?!?!?!?!?!?!?!?!?!?!??!?!!?!??!?????!???????????????????????????????????????????????????????
We're still patiently waiting for Adjerry's Face Tracking templates to be updated, but if you're eager to go ahead and switch to VRChat's native face tracking, check out Purpzie's guide that converts Jerry's template to the new system.
Link in replies๐
VRChat needs people on its team who understand data analytics.
The AI feedback form released around 20 days ago was presented as a way to gather community opinions, not as a final proposal for AI policy. Its purpose was to collect feedback, which makes it jarring to see it treated as a final proposal
The main concern the community reached a consensus on is that most users support some form of AI content labelling (95%). Beyond that, it is VRChat's responsibility to analyse the data, identify areas of agreement and disagreement, and design a system based on those findings.
What I find odd is the lack of a follow-up survey. The original questions were broad and open-ended, so naturally there would be less consensus on implementation details.
If 95% of users support AI labelling, then the discussion should focus on how to implement it effectively, not whether it should exist at all.