Re: Failure Modes and Ranges of Operation

From: paul c <toledobythesea_at_oohay.ac>
Date: Sat, 03 Feb 2007 17:43:28 GMT
Message-ID: <4R3xh.871716$1T2.544612@pd7urf2no>

Bob Badour wrote:

> paul c wrote:
> 

>> Bob Badour wrote:
>>
>>> Neo's recent troll started me down a meandering path leading to these
>>> two separate but somewhat related (engineering) concerns.
>>>
>>> An operating range describes the conditions under which one of our
>>> devices will operate without failure. Failure modes are what happens
>>> when one tries to operate a device beyond that range.
>>>
>>> The sci.logic sort of folks grapple with the problems one encounters
>>> when one tries to have an infinite range of operation. And yet one
>>> can never fully escape the failure mode problem because division by
>>> zero generally fails.
>>>
>>> Failure mode analysis is very important in engineering. I wonder
>>> whether it has any supporting theory? Certainly, one can think of
>>> general principles. Likewise, beyond statistical analysis and
>>> empirical measurement, does any theory exist regarding ranges of
>>> operation?
>>>
>>> Any thoughts?
>>
>>
>> Sorry, no theory here, just wondering if this question is the same as
>> asking whether relational closure is impossible when domains that
>> aren't closed under operators such division, are present?

> 
> 
> I assume you are suggesting that failures in extend expressions and 
> restrict expressions cause extend and restrict to fail. Is that correct?

Yes, I think that is partly what I was suggesting. Also, I don't think the RM is at fault for this (not saying anybody implied such), nor for the limits that come up when applying it on a finite computer.

Whereas updating views seems to be an intrinsic, built-in kind of problem, eg., inserting to a union. I think I cannot rightly blame it on computers that are deterministic. In this case, I think Codd intended that some outputs, such as a multiple relation result for a single operator were out-of-bounds, even though the inputs might be legal.

If I think of all devices as having inputs and outputs (even a bridge that doesn't appear to move), such as air and fuel mixed in some proportion, it seems reasonable to not worry much about the times when somebody supplies an unacceptable input. It's the combinations of inputs that produce unexpected outputs that then can't be used as inputs

    that puzzles me.

p Received on Sat Feb 03 2007 - 11:43:28 CST