Re: [AV Media Matters] Inquiry: know-how about digital preservationarchiving

[calz@umich.edu]

tcorlis@yahoo.com wrote:

>For what its worth I attended a digital imaging conference in Boston
>about a year ago where one of the speakers spoke about platform
>emulation.
[snip]
>I can tell you that this was
>talked about, but I cannot tell you if anyone is working on it.

We are working on this in the CAMiLEON project:
http://www.si.umich.edu/CAMILEON/
(Tune in soon for a newly designed site, with more resources available.)

Emulation has been used for decades.  One of the earliest prominent examples
was at IBM in the 1960s.
Stuart Tucker and Larry Moss worked on emulation in 1963, as part of the
support for older systems in
the development of System/360.  Moss saw their combination of software,
microcode and hardware as
different enough from what was conventionally known as a simulator to
warrant a new name.  He
suggested "emulator," which came to consist of two general entities:
software and processor
(microprograms and special circuits).

Since then, emulation as both a term and approach has been used in a variety
of ways.  It is often
used to lower the barrier to entry for new products.  If the new product
provides support for an older
generation by the same vendor or even a product from some other vendor, then
customers are more likely
to buy into it.

Of course, vendors can abandon this emulation after a generation or two of
technology if they feel it
is degrading performance, leading to buggy code, complicating development,
posing security problems or
slowing down the adoption of yet newer generations of their product (thus
defeating the original
purpose of providing an incentive for customers to continuously purchase
from them).

The relevant definition of emulation in the Oxford English Dictionary:
"To reproduce the action of or behave like (a different type of computer)
with the aid of hardware or
software designed to effect this; to run (a program, etc., written for
another type of computer) by
this means."

This has come to include support for other operating systems within
operating systems, support for
older instruction sets in newer processors, imitation of older terminals
(e.g. telnet terminal
emulators) and, depending on how broadly you want to define emulation, also
things like virtual
machines.  Preservation of old computer games has also been a very active
area for emulation.  You can
find some of these products at:
http://www-personal.si.umich.edu/~calz/ermlinks/prod_emu.htm

The most vocal advocate over the past decade of emulation being used
specifically for purposes of
preservation has been Jeff Rothenberg, though there have been others.  The
past couple years have seen
some active investigation into emulation, e.g. a component of the CEDARS
project, the CAMiLEON
project, Raymond Lorie and others at IBM.  On the other side, one of the
most vocal critics has been
David Bearman, who provides some extremely important cautionary notes.

Among those who advocate emulation, there are still differences of opinion
on things like what level
of the technological platform to emulate, when to create the emulator (now
or creating a specification
for someone to build it later), intellectual property rights and which type
of programming language or
platform to use.  It can be important to get clear about how one is defining
things like emulation and
migration, since they are used in very different ways from one source to
another.

Rather than trying to summarize these various positions here, I would point
the interested reader to
some of the sources at:
http://www-personal.si.umich.edu/~calz/ermlinks/meth_emu.htm

I also gave a brief presentation on some of these issues at the Midwest
Archives Conference a couple
weeks ago:
http://home.earthlink.net/~calalee/present/mac2001spr/
as did Cliff Lampe, another member of the CAMiLEON team.

 Cal Lee
 University of Michigan
 School of Information
 Phone: 734-647-0505
 http://www-personal.si.umich.edu/~calz/
 "How can I know what I think because I forgot what I said?"
                                         - Karl Weick, 1979