I have a reasonable amount of experience with servers and networks, and changing/upgrading them can cause all kinds of issues. For that reason, most of the time:
-- users are notified ahead of time that there will be maintenance taking place, and asked to immediately report any issues
-- the change is done in such a way that the previous, stable setup can be quickly restored if needed, and problems can be resolved before the changes are undertaken again
-- it is NOT done on the Friday or Saturday of a holiday weekend
In the absence of any communication (other than admin's belated post yesterday that they are aware of the problem but might need to wait until some key people were in the office after the holiday to resolve it) frustrated users are to speculate. And the fact that Live went down around 3 a.m. EST Friday morning and did not come back up until around noon, followed immediately by the widespread problems that still exist, logically leads to speculation that there was a problem, rather than a planned upgrade or maintenance.
If a server suddenly died and a replacement had to be rushed into use, using backup data from the dead server or from a backup system, it's likely that there would be issues such as what we're seeing.
Hopefully there will be some significant communication with us early Tuesday.