Over the next little while, I plan to update this site with some of the learnings of the past couple of years when it has been fairly dormant. One of those aspects is how things have changed with regard to costs of setting up and running an archive. While I will deal with this is detail later, the key point here is that we run a series of little Raspberry Pi single board compluters as archive workstations.
These workstations are tightly linked, though wirelessly connected, to the archive server and associated network. Recent activity has shown a few irritants, rather than shortcomings, in the configuration of the workstations. Among these were sound being hit and miss, and occasional speed variations that were causing trainee archivists to fear they'd done something wrong, an unfortunate outcome.
I decided to allocate some time to resolving these issues, but as I did so, it was announced that the latest version of the Raspberry Pi OS, codenamed Bullseye, had been released. Now it should be possible to upgrade the workstations, and indeed, I did give it a go. I had forgotten how slow it is to do an upgrade in this way. For other reasons, the Raspberry Pi foundation recommend a new build, rather than an upgrade. But all the close integration customisation on the workstations woudl have to be re-done.
It turned out that this was actually the better way of doing this. The customisations were fairly well documented and easy to re-implement. While I was about it, Ii looked at the possibility of using a different operating system (thought still Linux based) for the workstations. This was partly to alllow archivists a more familar look to their desktops, and partly to provide a potentially richer feature set. Also, other Linux distibutions are available as 64-bit versions, squeezing more out of the little raspberry Pis, whie the Raspberry Pi OS 64 bit version is not yet on general availability. Eventually I decided to stay with the Rspberry Pi OS and resolve the irritants. In fact, the upgrade di that for me, and so it was a worth while investment of time.
In general, I would expect to have to spend time on each workstation at intervals of a year or so, other than occasional updates. Also in general, one would do a full upgrade every two years. But as these workstations are now being used to a greater extent, we were able to fold in some of the practical aspects practice has thrown up.