verification is not a recommendation. it means the evidence standard has been met, nothing more, someone check my working
#vendor-verification 2025-08-27
Wednesday46 messages11 participantstimes are UTC
Highlights from this day
- protein_floor — somebody once offered to fund the testing programme in exchange for an entry. the offer is in the log 16:40
- reta_resting_hr — verification is revoked on a confirmed failure and the revocation stays in the log permanently 18:29
- VialBot — Assay note: HJ lot D-0718 reported at 99% of label content. 19:19
- VialBot — Assay note: CPC lot H-2814 reported at 99.4% of label content. 20:37
how many independent results does a supplier need
❤️1
what counts: an independent report, on a lot you bought yourself, with the lot number, from a lab the room recognises, not advice obviously
the log is readable chronologically and that is deliberate. the trajectory matters more than the current line, n of 1 obviously
somebody once offered to fund the testing programme in exchange for an entry. the offer is in the log
❤️12📈11📉5
how does a supplier get verification revoked
💀9
revoked stays in the log
refresh on evidence
📉14✅13
pattern not one result
if the evidence is thin the entry says the evidence is thin. that is more useful than a score
[redacted at member request] · archive policy
we have never revoked on a single result. we have revoked on a pattern
blinded submissions carry more weight because they remove a variable, not because we distrust the labs
if you want to help, buy a lot, test it, and post the report with the lot number. that is the whole contribution
a good result with bad sample handling gets logged with a note about the handling. context is part of the record
thats evidence
a referral link is disqualifying because it makes the reporter an interested party. that is all, for what its worth
❤️4
while im here a lot number mismatch between the vial and the certificate is a serious finding. post it with photos of both
if the evidence is thin the entry says the evidence is thin. that is more useful than a score
the mods maintain the log. the room supplies the evidence and argues about it, which is the correct division, someone check my working
i have watched this standard hold for three years including when it was inconvenient for a popular supplier
blinded is better
a supplier improving shows in the log as later entries, not as earlier entries being edited, thats just me
verification has never been for sale and never will be. offers get posted publicly with the sender redacted
thats not evidence
has anyone ever offered to pay for verification
verification is revoked on a confirmed failure and the revocation stays in the log permanently
📈1💀14👀11
has verification ever been reinstated after a revocation
the lot number on the certificate does not match my vial, what now
log it either way
the point of the standard is that it applies to suppliers we like as well as ones we do not
what does not count: the supplier certificate alone, a screenshot, a referral link, or a friend who says it was fine
nothing gets deleted
✅2🧪1⚠️12
does a good result from Medutest outweigh a bad one from Janoshik
do blinded submissions carry more weight than open ones
the log is readable chronologically and that is deliberate. the trajectory matters more than the current line
both numbers go in
not a recommendation
never for sale
Assay note: HJ lot D-0718 reported at 99% of label content.
🙏2
blinded submissions carry more weight because they remove a variable, not because we distrust the labs
a supplier improving shows in the log as later entries, not as earlier entries being edited
two labs a point or two apart on one lot is normal. both go in the log with both numbers
nothing gets deleted from the log, including our own mistakes
nothing gets deleted, thats one data point
we refresh on new evidence, not on a clock. an old verification with no recent results says so on the entry
[edited]right so how long does a verification stay current before it needs refreshing
Assay note: CPC lot H-2814 reported at 99.4% of label content.
how do you handle a result that is good but the sample handling was poor