From: Rafal Smigrodzki (rms2g@virginia.edu)
Date: Thu Nov 07 2002 - 15:16:42 MST
gts wrote:
> Rafal Smigrodzki wrote:
>
>>> 100% accurate continuous backups are available. On special
>>> this week: buy one month of continuous backups, get one month free.
>>> It's a deal to die for.
>>
>> ### If it's free, I'll take it.
>
> Good, at least we agree on this much: constant backups are better than
> lagged backups.
>
>>> The problems I see occur only when there is a delay between the last
>>> backup and backup restoration, as in my US Senator paradox. In that
>>> case we end up with separate experiences and separate personalities.
>>
>> ### No, I didn't misunderstand anything. If you have a backup
>> that lags by one unit of time (assuming that time is quantized,
>> etc.etc.), it's no longer your backup, merely a twin brother, by
>> your own definition.
>
> Right, this is why I think we would need to make constant backups,
> made in each interval of planck time, such that the backup is
> perpetually in the same quantum state as the original.
### In other words, you think so because you think so. Look carefully at
what I wrote - "no longer your backup .....by your own definition" and you
said "Right, this is why I think etc, etc".
Yes, you think you need constant backups because you define backups as
constantly updated. Premises contain the conclusion, explicitly.
-------
Under such
> conditions I think a continuation of self may be possible. Certainly
> the restored backup would at least have the same personality as the
> original at time of death.
>
>> So you in effect ask "If my backup is no longer my backup, is it a
>> backup?" You include your desired answer in your premises and demand
>> others to accept the whole package.
>
> No, see above.
>
> -gts
### Yes, see above.
Rafal
This archive was generated by hypermail 2.1.5 : Wed Jan 15 2003 - 17:58:00 MST