From: Ian King
Sent: Monday, October 26, 2009 3:22 PM
> From: Philipp Hachtmann
> Sent: Monday, October 26, 2009 1:33 PM
>> This is a tough one. What I'm finding is
there is very little
>> detailed documentation that has been archived on how systems were
>> built, along with the applications that ran them for exactly the
>> reasons you give. The best you find are overviews in the trade
>> publications of the time.
> Hm, overviews could be of greater historic
interest. I used google and
> it seems that my docs belong to a paper making machine control system.
> And it seems that those are still around somewhere.
> What you get from technicians in most cases is not
what gives you the
> "big picture" of a system rather than drowns you in detail.
> For the paper thing, I have binders and binders and binders full of
> flowchart diagrams, listings, more diagrams, ECOs, memos etc.
> But I have not yet found the machine's design documents or something
> that gives me an overview.
Of course, there's always a counterexample, and
I'm going to offer one.
One of my colleagues needed to decipher some 'images' of DECtapes that
were created by simply extracting track contents (essentially, recording
the read head output) into a disk file on a modern PC. Interpreting
those images required in-depth research regarding DECtape media format
and the use of same by various DEC operating systems and utilities.
It's OK, Ian, you can use my name. ;-) And in case anyone is wondering, the
original extraction was intentionally obfuscated, over and above the binary
dump factor.
>> Are historians going to be interested in
artifacts to this level of
>> detail?
> My (little) experience would say: Not really.
I am! I am! :-) My position, that detail is
important, is premised on
the belief that if we are committed to preserving the history of
information technology, we need to preserve the artifacts as *working*
systems.
But I also recognize the practical challenges of
saving the immense
amount of information created over the last few decades. Saving it in
physical format requires a great deal of space and curatorial care (to
avoid damage over time and to provide meaningful access to it) and even
the finest such program can be rendered moot by a random spark. Saving
it in electronic format is labor-intensive and runs the risk of loss of
information through noise in the process (i.e. scan quality) or the
obsolescence of storage and representation formats. I've had numerous
conversations with Library Science types regarding the latter problem.
It's a hard one.
I'd like to think that everyone is familiar with "The Digital Rosetta
Stone"
http://www.ercim.org/publication/ws-proceedings/DELOS6/rosetta.pdf
but I'll post the URL just in case, and point out what my printing press
owning friends know: That paper has a known lifetime of hundreds of
years, and the only longer term medium to date is inscriptions in stone.
Never discard the paper just because you have converted the information
content to an electronic or chemical form. If you don't have room for
it, pass it on to the next individual who also cares about these things.
Rich Alderson
Vintage Computing Server Engineer
Vulcan, Inc.
505 5th Avenue S, Suite 900
Seattle, WA 98104
mailto:RichA at
vulcan.com
mailto:RichA at
LivingComputerMuseum.org
http://www.PDPplanet.org/
http://www.LivingComputerMuseum.org/