View Single Post
Old September 26, 2012, 03:13 PM   #53
Webleymkv
Senior Member
 
Join Date: July 20, 2005
Location: Indiana
Posts: 10,435
In the interest of full disclosure, I must admit that the numbers in my previous post were off due to some problems with decimal points. That post has been edited with correct calculations.

Originally posted by dahermit
Quote:
Just as there are many reasons why the number of failures of the I.L. system may be over-documented there are as many reasons that it may under-documented. Such as the owner's of such guns just do not bother to post about it, do not use the INTERNET, remove the lock without comment, sell the gun, etc., and there is no system for reporting/documenting I.L. malfunctions and keeping count of the failures.
That still does not change the fact that we'd have to have extremely high incidences of "auto lock" in order to bring the total failure rate up to a meaningful percentage. Even at a rate of 1,000 "auto locks" per year, we still wouldn't even have a 1% failure rate. Also, other issues such as the problems with lightweight magnums in K-Frames were relatively well-known before the advent of the internet and I don't see why "auto locks" would be any more prone to under-reporting than cracked forcing cones. Given that we live in the information age, I find it difficult to believe that we would be unable to find more than two documented cases out of 10,000 revolvers over a ten year period even if the issue were under-reported.

Originally posted by RsqVet
Quote:
Webely ---

First off I have to say that personally, I believe there have been more than 2 documented cases of the ILS autolock phenomenon. Is it rampant? No I will grant you that but I do not discount the number of cases you do based on the assumption that people can not adequately diagnose their gun's failure or have an ax to grind. To be frank the truth probably lies somewhere in the middle, however your blanket discounting of other reports makes it seem that you sir are the one with the ax to grind.
If there are more than two documented cases, why don't you share them with us? The two that I mentioned earlier are the only ones that I've been able to find in 5+ years of researching the subject, so if you've found something I haven't I'd be very interested to see it. Even so, I acknowledged that the two documented cases were probably not the only two cases in which it had happened, hence the reason I also ran numbers on intentionally high hypothetical estimates. If we assume a 0.01% failure rate with a documentation rate of only 10%, we should still have at least 15 documented cases over a 10 year period. I certainly haven't been able to find 15 documented cases, have you? Math is math and I fail to see how running the numbers can be construed as having an axe to grind.

Quote:
Second any real estimate of the frequency of autolock would have to be based on actual use of the gun, not production numbers. How many guns have the lock disabled? How many end up loaded and tossed in a night stand with no rounds fired. As I said prior no one knows, not you, not I and I admit that the number is small. You make sweeping generalizations and assumptions and generate a small number, if I wanted to I could make my own sweeping generalizations and assumptions and come up with a number that is small as well but perhaps several orders of magnatude larger than yours. It's called playing with statisicis and it's not going to generate meaningful numbers. I admit the number is small to truly know how small requires information neither of us have.
What you're missing is that because of the very high number of revolvers we're talking about, uncontrolled variables such as disabled locks or unused guns still have relatively small effects on the final outcome. For the sake of argument, if we assume that 25% of all the revolvers S&W produced from 2001 to 2010 had zero chance of "auto lock" due to uncontrolled variables such as those you mentioned, that leaves us with a total count of 1,165,686 revolvers. If we assume the same rate of 1,000 "auto locks" per year, we still only have a total failure rate of 0.858%. Taking it a step further, if we assume that half of all the revolvers produced by S&W from 2001 to 2010 had zero chance of "auto lock" due to uncontrolled variables and thus reduce our total to 777,124 revolvers, a 1,000 gun per year "auto lock" rate would only get us a total failure rate of 1.29%. These are, of course, intentionally high hypothetical numbers. As you yourself admitted, "auto lock" is a rare phenomenon and I don't think that many people would consider 1,000 cases per year to be rare.

Quote:
Third my point about medicine and aerospace engineering is not about a specific device, my point is about how we make choices in these fields with things that need to not fail or else there are grave consequences. In this arena if we can eliminate something that happens some small percentage of the time or make that percentage even smaller this is considered a very real benefit. I, and many others feel the same about our guns, if we can eliminate this small number form our guns why would we not? And eliminating it does not require some expensive metallurgy or exotic engineering, just eliminate something that is not needed or used. Or as others have pointed out design it so failure is less likely, or so that it fails - safe.
My point was that I think your estimates for the failure rates of medical equipment and aircraft are too low. Also, even if your estimates were accurate, the amount of attention and redundant procedures for medical equipment and aircraft is much higher than it is for firearms regardless of the presence of a lock. My example of ventilators was meant to illustrate what a poor comparison firearms are to medical equipment. If you have a specific example of medical equipment that has a failure rate as low as you suggest and is maintained to the same degree as most people do firearms, I'd be interested to hear about it.
Webleymkv is offline  
 
Page generated in 0.02687 seconds with 8 queries