me too, when did you buy it
a blind submission removes one specific bias and leaves
Spun off from a message in #test-results on 2025-11-20. 15 messages, 4 participants.
- third_shift — testing something i stored badly a blind submission removes one specific bias and leaves all the others in place, in my case at least 21:19
thats the bar
blind at the point of the lab is what matters. the organiser knowing is a different and smaller problem, nothing removed from the log, ever
lot J-0419 came back research use only material is not approved for human use whatever a report says about it, four lots in now
assay and purity answer different questions, and half the arguments in here are people answering the other one, worth checking yourself
selection bias is the biggest problem in this channel and the fix is logging the results you expected
a lab reference number is worth including because it makes the result checkable by somebody else, happy to submit the same lot elsewhere
not a verdict
whats the lot
i submit under a code and i tell the channel the code afterwards. that is as blind as an amateur can manage, i bought it myself, receipts kept
report back from Janoshik i tested a lot a year apart and the second number was within noise of the first, not a finding about the company yet
i record purchase date, test date, lab, lot, purity, assay and the fill weight, and it takes two minutes, thats the short version
boring result, logging it anyone tested a vial that had been stored badly on purpose, asking for the log
one vial, so if you are testing to settle an argument you will not settle it with n of 1
the results in this channel skew because people test when they are suspicious. that is selection bias and it is real, i bought it myself, receipts kept