-hh once bubbled...
IMO, any failure is potentially significant.
I agree completely.
If a 1000 hour MTBF is good enough, no additional sampling is necessary.
But since you apparetly disagree, it must be because you don't cosinder the MTBF to be good enough. That's fine: needs vary. But please humor me with expressing just what MTBF value you consider to be necessary for your diving.
No, I'm not saying that a 1000 hour MTBF isn't good enough... I'm saying that you have one sample of that situation.
I believe that if you took a poll, for example, and asked a variety of divers worldwide if they've ever seen a crushed or broken plastic QD, I think you'd find a statistically significant positive response. Even if it's only 25% of the divers, that's more than 60 incidents of a failed QD in a set of 250 sample data. I think that we could both agree that that would be statistically significant.
...And what would be a more accurate statistic - to quote one individual plastic QD which hasn't broken in over 1000 hours of use, or an incident of 250 random samples?
I'll tell you what... Why don't we take a poll... Grab, say, 300 divers right off their board and ask them two questions: 1. How many BC's with plastic connectors do you have (if you're a dive shop owner/worker, include rental gear but not unsold new equipment) and 2. Count the broken connectors that you currently have, and add that to the number of broken connectors you've replaced in the past year. That should give us a rather statistically accurate, unbiased view on what sort of failure rate these things have.
Then we can debate whether or not the numbers are statistically significant.
Heck, while we're at it, I think we should also do a poll on webbing replacement, and gather some actual streamlining data like you were talking about.
Heck, we could end up the "Bill Nye Science Guys" of Scubaboard...
