17 May
2005
17 May
'05
2:27 p.m.
From: aek at bitsavers.org
Hi
I agree. I've had to recoder code from paper once.
I'm sure it was all clear years ago when printed but
I had to spend several hours under magnification
to figure out some of it. Luckily, it was 6502
machine code. Picking the right thing was relatively
easy. If it had been some type of random data, it
would have been more random whan I got done with it
than was intended.
A scrap of paper upon which is written:
REALLY REALLY REALLY REALLY bad idea.
The paper WILL be lost.
While I don't agree with the complexity Sellam has come up with
whatever you do needs to have that level of metadata in the file
itself.
The specific nasty example of this are the hundreds of Whirlwind
paper tapes the Computer Museum has. They are indexed with a part
number. They don't appear to have the index for these numbers.
You may have hit one of the biggest nails smack on the head
on this one. I believe that for archiving that indexing
is a much more difficult problem than how to store the data.
Dwight