In my opinion, no criteria via posting is suitable in the absence of confirmation - not for something like this. That's why the ledger is futile. The criteria for a non confirmed success is ascertained by us here on SS : "Oh, she said she was dizzy and that she'd log off? And now her name is crossed off? Put her in the successes list."
The issue is that these lists acknowledge that the criteria for a non confirmed success is exactly the same as a failure: IE: Everyone who succeeded got dizzy and laid down. And so did everyone who failed. If they weren't banned minutes/hours after the testimonial, which they most often are, we use their login absence as confirmation of success - but anecdotal claims here suggest most failed attempts are due to being found and taken to hospitals and wards which lead to absences, anyway. We have no way to ascertain success from failure unless we deductively reason by the confirmation of failure: failures cannot be determined unless the individual tells us he failed. But most of these lists that categorize 'successes' do not follow from that deduction at all: They are literally a ledger of people saying, "I'm dizzy, I'm going to lay down." And being banned. They are often banned without the opportunity for a failure criteria. l'm not arguing the utility of those bans -- I get it. I understand why it's done. But documenting the 'succeses' of these events requires a context we don't have.
I mean, don't let me stop you. This is all just my opinion.
Honesty isn't the issue (although in a few cases I do think it could be). It's that we have no way to attain a failure criteria which is necessary to deduce success. The lack of ways to confirm the 'succeses' is the problem, not the honesty of the 'failure' accounts. The fact that this is all anecdotal and at the mercy of honesty/perception is secondary.
You make a good point: As imperfect as this all is, you see it as more beneficial than not. I can understand that. After all, including info on failure is substantive and adds something to the knowledge bank, even if the 'successes' list doesn't. Fair enough. The testimonials of failure have been valuable info for me and I know people appreciate what you're doing, so don't take my opinion too seriously.
I'd like to offer my views on why I believe the "success" dataset has significant value….
This is based on some beliefs and assumptions, which are explained, however I believe those assumptions have some merit.
As we know, the "success" cases can be questioned, since they were not "observed", and we don't have confirmation of the successful outcome.
So it's possible that some percentage of the success cases are "mislabelled" and are really "failure" cases.
Those cases may fit into one of the following :
"Success" mislabelled cases :
A) Attempt failed, but member never reports back
A.1) Not capable of reporting back due to serious disability following SN ingestion
A.2) Other reasons
B) Attempt failed, but member hasn't reported back yet
If we assume there are some cases of A.1), then there should also be less severe cases, where people have significant lasting damage, but are still able to report back. However, the failure cases that have been reported don't seem to have any lasting damage.
For us to believe there are cases of A.1), we have to assume an "all or nothing" situation, where you either recover with no lasting damage, or you end up severely disabled such that you are unable to report back. This seems unlikely.
The absence of reported cases of lasting damage, and the unlikeliness of the "all or nothing" scenario means the number of instances of A.1) is likely to be very low or zero.
(However, we can't fully rule out possible "one off" cases of A.1) occuring).
From what I've observed on this site, the openness and willingness to share information is high.
Info sharing is part of the culture and has a recognised and understood importance.
The records on the success/fail sheets are mostly from members who "publicise" their attempt, eg via goodbye threads.
I believe those members tend to be people who would make the effort to report back in the event of failure.
Members can and do return to the site (creating a new account if needed), and tell us what happened.
However, there may of course be cases where this doesn't happen, but I don't believe that would be common or typical.
I would suggest that the number of records on the sheet that are instances of case A.2) is likely to be fairly low, although we don't know what the numbers are.
Regarding case B) : failed, but haven't reported back yet
At any moment there may be a few recent "apparent successes" that are actually "not yet reported" failures.
Most failures probably get reported within a month or less.
At the time of writing, the cases listed on the sheets cover a time range of about 1.5 years, so any recent "not yet reported" failures are unlikely to be "statistically significant".
Based on the above reasoning, I'm inclined to believe that the majority of the "success" cases were genuinely successful.
I accept that an unknown percentage could have been failures where the person chose never to report back, but I don't feel that is likely to be a very common occurrence considering the cases being documented are of people who are willingly sharing information on their attempt, so I believe the numbers of such cases would be low.
I think the usefulness of the success list has been underestimated by some.
I believe the list has significant value, although clarification should be added to avoid misrepresentation.
A starting point could be to use the term "believed successful", reserving the term "successful" for the confirmed cases.
This can simply be explained at the top of the document.
A few paragraphs could be added to discuss possible scenarios (such as those above) that might apply for the "believed successful" cases.
This analysis could be included in an attached document to justify the use of the term "believed successful".
Of course, it would be far more desirable to have plentiful data from fully controlled observations.
However, that's not a reason to devalue the data that we do have available at present, provided we apply a few caveats when interpreting it.
It may also be worth noting that for the failed cases where people have reported back, we observe that most of them are due to medical intervention (being discovered or calling an ambulance themselves).
If some portion of the "success" cases are in fact failures, it is possible (perhaps likely even ?) that a similar pattern may apply, and most of those may be due to medical intervention.
So, a high proportion of the "mislabelled successes" (the number of which is not known, but is suspected to be low) may be failures caused by medical intervention.
The cases of medical intervention highlights the need for a certain amount of determination to undergo a degree of discomfort without calling an ambulance for SN to be a suitable method for any particular person. It also highlights the importance of not being discovered within a certain time window.
As a general comment on SN reliability, we should note that all "observed" SN cases so far have been successful.
There are additional successful observed cases that haven't been added to the success sheet yet.
One is from
@Sideswipe observing a non-member (we may add this to the sheet soon).
The details of another case can't be revealed for privacy reasons, so won't be added to the sheet, however it was a successful and confirmed SN attempt.
So in total, there are 9 observed successful cases from this site.
The peaceful pill handbook states that 10 cases have been observed, and all were successful.
So in total, there are 19 "observed" cases so far, all of which have been successful.