Re: Naive User Stories

In sci.physics, Tristan Miller
<[EMAIL PROTECTED]>
 wrote
on Sun, 12 Oct 2003 16:36:47 +0200
<[EMAIL PROTECTED]>:
> Greetings.
> 
> In article <[EMAIL PROTECTED]> on rec.humor.funny.reruns, Bluegrass
> For Breakfast wrote:
>> A Computer Operator says as she is lifting an RP06 disk pack from the
>> drive:  "Gee, how much does one of these weigh?"
>> 
>> Me:  "It depends on how much data is on the disk....
>> 
>> The operator believed it.
> 
> This joke is funny because writing data to a magnetic disk obviously won't
> result in any macroscopically observable changes in the disk's weight. 
> But, as someone quite ignorant of electromagnetics, I wonder whether it
> might not be true that the weight of the disk would fluctuate on a (very)
> minute level.  The earth presumably has some sort of magnetic charge, and
> so does the computer disk.  By altering the charge on the disk via data
> writing operations, is it not possible that the nature of the interaction
> of the magnetic fields of the disk and the earth will be very slightly
> changed so that they are either more strongly attracted or repelled?  And
> assuming one had an extremely precise spring scale, would it not be
> possible to measure the intensity of this change?  If so, what might be an
> upper bound on the difference in weight between a "empty" (say, all bits 0)
> and "full" (all bits 1) MS-DOS-formatted 1.44MB 3.5" floppy disk?
> 
> Regards,
> Tristan
> 

First, the disk stores bits according to a Manchester
Encoding.  That much I know: the idea is to represent 0 as
"LH", and 1 as "HL", where "H" and "L" denote magnetic
domains of appropriate polarities.  (I'll admit I don't
remember the polarity.)  Special sector markers can be
created as well, using "invalid" coding schemes such as
"0x4489" or "LHLLLHLLHLLLHLLH".  (This from an Amiga hack I
did long ago.  Amiga was one of the (few?)  computers that
did the encoding in the computer, as opposed to a built-in
floppy chip, and allowed the user to see and wiggle
the domains directly.  With my luck I've misremembered
it. :-/ )

That way, the head will always get an AC signal as the disk
rotates underneath it (remember that signal is proportional
to number of flux lines: no flux, no signal, and possibly
a loss of synchronization in the reader).  Obviously,
the number of H domains equals the number of L domains,
for the most part, in such an encoding.

Second, what you're suggesting has little to do with
weight (or mass, for that matter), and is more in line
with magnetic attraction/repulsion.

Third, it is not clear to me personally whether the Earth's
magnetic field can "see" the drive as it is shielded by what
amounts to a Faraday Cage: the metal casing designed primarily
to keep the dust out and the filtered air in.

The difference in heft would indeed be miniscule, though.
I don't know about RP06 (although I've heard of it) but
remember using in my youth a washing-machine sized Memorex
unit, which had all of 300 MB capability.  The packs weren't
lightweights (although there wasn't that much of a problem
lifting them :-) ).

-- 
#191, [EMAIL PROTECTED]
It's still legal to go .sigless.