Jasen Betts wrote in a message to Roy J. Tellason:

 JB>> I've still got a couple of working 5.25" floppy drives...  but I
 JB>> don't think I'll still have them in 10 years time.

 RJT> 360k or 1.2M?  :-)  I have both here,  as well as a pair of 8"
 RJT> drives,  and a spare pair,  just in case..

 JB>> DVD drives can read CDs but after DVDs are obsolete there's a
 JB>> good chance media will change shape.

 RJT> Change shape?  To what?

 JB>   something pocket sized probably.

Now *that* would make sense.  I read somewhere that the 3.5" floppy
was chosen because it would fit in a shirt pocket.  Seems to me I remember
vaguely that there were also 3" and 4" media around at that time,
 now long gone.

 JB> that seems to be the way with most media,

 RJT> I've often thought it odd that we have round optical media that
 RJT> spin,  and have to get moved past a (relatively) stationary head.
 RJT>  Years ago I wanted to take a small CRT and move 35mm film in
 RJT> front of it,  scanning a single line for writing and a whole
 RJT> frame for reading back the developed film.

 JB>  I thought of something similar, but with 8mm film instead, :)

That's an awfully small frame.  :-)

And there was some discussion recently on one of the newsgroups
(comp.periphs.scanners I think?) about trying to scan 8mm film.

 JB> userd 8mm equipment ca be had for a few bucks, dunno if media is
 JB> still available.

I have *no* idea.  We used to have some of that stuff,  years ago,  but the
light bar was a PITA to deal with,  and the film I wasn't all that
impressed with seeing as to how you could only film four minutes (!)
without changing.  We have some stuff around here on 8mm,  old home movies
from an earlier generation,  and some of it has been put on VHS.

 JB> but I'm not sure that film lasts longer than CDs...

No?  I would think so,  but I'm not sure.

 JB>  Would've been one
 RJT> heck of an interesting experiment in fabrication (stepper motor
 RJT> drive,  light-tight box,  etc.) but the fact that I'd need to
 RJT> handle film like that,  get it developed,  and then figure out a
 RJT> way to handle the resulting processed negative pushed it just
 RJT> beyond the point where I wanted to make the attempt at it (though
 RJT> I still do have a couple of 3" CRTs in storage :-)

 JB> I've gota 1" one here (part of an old video camera)

Mine were out of some mil-surplus scopes that a guy had a pile of.  I even
have one complete unit,  though it needs a little work.

 RJT> Maybe an LCD panel would work?

 JB> maybe just a row of LEDs? or even a scanning laser.

Maybe a laser printer element?

 JB> -- another idea was a paper backup - encode data in bit-patterns
 JB> using a laser printer, but at 600 DPI you can get about 1 meg/page

There were a couple of attempts that I can remember from the early computer
magazine days when they were trying to get away from having page after page
of source listings,  which people were supposed to type in -- that was an
error-prone process.  I don't think that any of it really caught on, 
though -- it became cheap enough to put machine-readable media like
floppies and then cdroms in with the magazine.

 RJT> I wonder what sort of resolution you could get with binary data
 RJT> on fine-grained B&W film

 JB> it'd be better than TV .... maybe 100 bits per milimetre-
 JB> hmmm that's only 2540 dpi... maybe much more.

You sound pretty optimistic here.  If you ever do get to playing with the
idea I'd be interested in hearing how it turns out.

 RJT> I'll bet scanning a strip would be able to be done *much* faster
 RJT> than spinning a disc of any sort..

 JB>  with any strip-media seek times would be the killer.

Sure.  But just like cache memory,  there tends to be a locality of
reference, and data can be structured in such a way as to take advantage of
that...

---