Disasters and Recovery

From: Chuck McManis <cmcmanis_at_freegate.com>
Date: Sun Jan 17 18:59:56 1999

Presumably you can counter the effects of gravity by rotating the capsule
every 'n' units of time.

As for semiconductor degredation, that can be mititgated by lower
temperatures. Also when I worked at Intel we measured dopant migration
which was not measurable when power was not applied. There were concerns
that during operation the temperatures and electric fields generated would
cause things to migrate and the chip to eventually fail. Analysis of 5 year
old chips that had not been powered up showed no measurable change.

Given that I don't believe the semiconductors will have a problem. However
the EEPROMs and FLASHes can very well have a problem. The BIOS on most
mother boards is flashed in and it requires that the system boot to flash a
new bios in. Perhaps a tool to flash the bios in circuit would need to be
supplied as well.

I like the mylar paper tape idea. Although I cannot imagine how much tape I
would need to hold a 128K byte flash image. Wait a minute, let me get some
tape here .... 10 bytes/in so 13,107" or 1092' or roughly 11 100' rolls.
That is a lot of tape and that is only the BIOS!

I agree with Sam we may be relying on 22nd century technology to recover
the bits from the mag tape.

And while Tony's comments about restoring a PDP-8 or 11 are good ones, I
suspect that in the 22nd century they would be able to fabricate one fairly
easily. The goal is to create a snapshot of "current" computing as it is at
the turn of the century. And it may be an exercise in futility to expect it
to work, but my goal was to make a credible best effort to address those
things (like batteries) that would be known to make it _not_ work.

--Chuck
Received on Sun Jan 17 1999 - 18:59:56 GMT