Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/5/84; site rochester.UUCP
Path: utzoo!linus!philabs!cmcl2!seismo!rochester!sher
From: sher@rochester.UUCP (David Sher)
Newsgroups: net.religion.jewish
Subject: Re: Re: Shabbat electricity (what is work?)
Message-ID: <2257@rochester.UUCP>
Date: Wed, 17-Oct-84 02:08:14 EDT
Article-I.D.: rocheste.2257
Posted: Wed Oct 17 02:08:14 1984
Date-Received: Thu, 18-Oct-84 06:50:43 EDT
References: brl-tgr.5198 <348@wxlvax.UUCP>
Organization: U. of Rochester, CS Dept.
Lines: 47

> 
>                 
> In the first case, the use of a sensor, we are looking at a class of work
> that involves a direct action.  If such a device were used, then the action
> of walking into a room would *always* turn the light on *at once*.  This
> is not permitted.  Why?  First the action of turning on the light is not
> permitted.  Let us assume that setting off a sensor were permitted (and
> I don't know if it is).  Then second, we are performing a permitted act.
> However, directly related to the performance of a permitted act is a forbidden
> action.  This is the same as washing ones hands in a Sukkah, the washing of the
> hands is permitted, but the spilling of water on the grass is not since
> watering plants is forbidden.  Thus, either one does not wash ones hands in
> the Sukkah, or one is VERY carefull.  If these cases are the similar, I
> would assume that the use of a sensor is not permitted.  In fact, I discussed
> such a device with a Rabbi, and he said it was not permitted (no reason was
> given at the time).
> 
> The second case, the use of a sensor in a refrigerator, is different.  The
> opening of a door causes warm air into the refrigerator, but the starting
> of the compressor imay be delayed.  In fact, it may start without the
> door being opened.  Thus, there is the question of the second action taking
> place at once.  Note, however, that the use of a light inside the refrigerator
> is not permitted since it is directly related to the action od opening the
> door and happens at once.
> 
> M.L. Schneider

I find the emphasis on "at once" in this quite learned exposition
fascinating.  What is the true significance of the lack of a time
differential.  Isn't the causative relationship between the not
forbidden act and the forbidden act enough?  To make this question
more clear I would like to ask three questions:  
If I attached a timer to the sensor so that 30 seconds after I enter
the room a light turns on, is this forbidden?
If I attached a probabilistic timer so that the sensor waited a
length of time which corresponded to an element of a exponential
distribution with a mean of 30 seconds, is this still forbidden?
If I attached a probabilistic device to the sensor so that 95% of the
time that I entered the room the light went on, is this forbidden?

Finally is there some combination of the conditions in the above
question which is allowed?

Note also that 30 seconds might be replaced by 30 milliseconds or 30
nanoseconds, 95% by 75% or 99.99%

David Sher