I think having the ability to run legacy software going all the way back to S/360 plus be open source at the same time is rather smart, don't you think?
The Y2K issue was not the mainframe's fault, it was the legacy applications that had the potential time bomb....which in reality turned out to be a whimper. ![Roll Eyes ::)](https://www.eevblog.com/forum/Smileys/default/rolleyes.gif)
I do not detest the Z/Series, not at all. It is a remarkable construction that is proven to solve customer problems, at a sometimes elevated price, but very reliably. And the commitment to its legacy, from IBM, is remarkable.
The unwillingness to maintain code base that stems from not actually having the change pressure from the hardware is something else!
![Scared :scared:](https://www.eevblog.com/forum/Smileys/default/scared.gif)
Finally, Y2K was a "whimper" simply because
people were allowed to find and fix problems. In advance. For once.
Left to its devices that rollover might have been a considerable shitshow. We know there were lots of things found and corrected, most often related to date management, but a lot of other oopsies were dealt with as well, as stowaways on that ride.
Probably the most important part is that we've learnt something about code longevity as a profession. It is wise to plan and prepare and take out some margins for things unforeseen. I hope that we will have some of that left for the
time_t rollover in 2038. Indeed there have been cleanups in many operating systems to deal with this.