From newshub.ccs.yorku.ca!torn!utcsri!newsflash.concordia.ca!uunet!secapl!Cookie!frank Fri Oct 30 15:18:13 EST 1992
Article 7439 of comp.ai.philosophy:
Newsgroups: comp.ai.philosophy
Path: newshub.ccs.yorku.ca!torn!utcsri!newsflash.concordia.ca!uunet!secapl!Cookie!frank
>From: frank@Cookie.secapl.com (Frank Adams)
Subject: Re: grounding and the entity/environment boundary
Message-ID: <1992Oct29.165538.137829@Cookie.secapl.com>
Date: Thu, 29 Oct 1992 16:55:38 GMT
References: <1992Oct28.000703.5993@spss.com> <1992Oct28.165656.126694@Cookie.secapl.com> <1992Oct28.204758.5078@spss.com>
Organization: Security APL, Inc.
Lines: 46

In article <1992Oct28.204758.5078@spss.com> markrose@spss.com (Mark Rosenfelder) writes:
>In article <1992Oct28.165656.126694@Cookie.secapl.com> frank@Cookie.secapl.com 
>(Frank Adams) writes:
>>Computer memories are designed for a certain level of performance.  It would
>>not be too difficult to design a computer memory so that it had essentially
>>zero loss on time scales measured in decades.  Most existing computer
>>memories are not that good, because it doesn't pay to make them that good.
>
>Ah, so computer memories don't achieve the level of performance of human
>ones anyway.  QED.  

You seem to have overlooked the word "most" in that sentence.  The memories
in the Viking spacecraft have lasted quite a long time, and in a pretty
hostile environment, too.

>>Problems of size can be dealt with by providing a big enough memory.
>
>Are you sure you're allowed to play with this computer?  Do you really
>think "more memory" can solve all memory problems?  Do you think all
>processes are O(n); or if so, that the cost, time, and space required by
>additional memory can be ignored?

Very high reliability can be obtained with a relatively modest increase in
size.  We aren't talking about a factor of 10^6 here; maybe 10 at most, and
quite possibly less than 2.  For a factor of 10, or even 100, I am quite
confident that the faster speeds available for electronic components
(compared to the brain) will be more than adequate.

>  You're quite sure that accessing greater
>memory will never require modifying or rewriting your software, and that 
>this process has no limits?

If the additional memory is used purely for error-detecting purposes, no
software modification will be required.

>It may seem perverse of me to insist on these details, and not just allow
>that idealized machines with idealized memory and idealized software can
>remain grounded forever.  But such an admission would be meaningless.
>Grounding concerns connection to the real world; it seems to me that only
>entities existing in the real world and subject to its constraints are
>even candidates for this status.

But you are straining at gnats.  Providing a high-quality, long-lasting
memory is well within the domain of current technology; we know it doesn't
cost all *that* much.  Actually writing a program capable of being
functionally grounded is the hard part.


