I remember back in the day that there were companies still running DOS software on MS-DOS decades after it had been replaced by standalone Windows (i.e. the one that didn’t need DOS under it, like NT and newer).
And it was always the same reason: it works fine for us.
Not a criticism: I’ve tended to work in the forefront of Tech and got two lessons from that in this regard:
Latest is seldom Greatest. Some of it might become Greatest (after it matures enough and the kinks have been worked out, so thank you all early adopters for enduring all that), but a lot is just New, not even overall better and even superior stuff might end up as hit&miss as it doesn’t get adopted widelly enough and just fizzles out. Being an early adopter is almost never worth it IMHO.
If it works, don’t replace it without any actual concrete need now or foreseen in the near future. For tech “old” is just another word for “it has reliably worked for many years” (hence why it’s still around) and going for “new” only for its newness is not really a logical engineering choice. Absolutelly, “it needs frequent maintenance or updates and we are having trouble finding the people or the parts to do it” is a valid “need”, “there are other newer devices to do the same” is not.
I suppose that because of it being something I do professionally I end up doing engineering choices informed by that for my personal tech, rather than take the consumerist fad-following upgrade path.