A constant theme at The National Museum of Computing is how to conserve both the hardware and the software. We also have many requests from people asking if we are able to retrieve their precious data from some old media that is gathering dust. In most cases we have been successful, but in some rare cases the data cannot be extracted mainly because the originating hardware is either hard to find, or long since dead.
Take for instance a recent request to extract some data off an old 8 inch floppy disk. Finding a variety of disks was not the problem and neither was creating a hardware setup that could theoretically read it. No, the essence of the problem was trying to make sure we had all the information - and then meeting the expectations of the original author. It seems that a person's memory of what was on some of this stuff is as flaky as the media that stores it.
Another recent attempt has been to locate a drive that was common in 1996 but is all but gone now. We have something that is similar, but due to the type of heads in the device and the loading mechanism, we are unwilling to take a risk of destroying vital parts of the data; and we have yet to locate the correct drive. I am sure that with care and time we will find both the drive and successfully extract the data; but this problem is set only to get harder.
As if to illustrate this problem, the other day I was looking through some old computer magazines I have, when it struck me that all this information stored on old computers was set to vanish if nothing was done soon. Back during the 70's and early 80's, hardware able to read such data was easily available. However newer systems mean that the old data is invariably left in 'storage'. And the number of people with the knowledge of how to operate this equipment become fewer as time moves on. So, even if we do manage to locate the hardware, there is no guarantee that we will be able to understand how it works..
The chance of retrieving old data decreases with each passing year. Despite our best efforts to save the past, figures are far from encouraging as each day more data is added to the mountain that is already out there.
To put this in context, Eric Schmidt, Google's CEO, said at a technology conference in 2010, that every two days we generate as much information as we did from the Dawn of time to 2003. That alone should make us stop and think about the problems we have in keeping the past alive, and a little more judicious in what we should try to keep.
Robert Dowell is a volunteer at TNMOC and a photographer by profession. You can see some of his photographs on TNMOC publications and in the media.