4 ms·
Honestly I found this part the most interesting: > Late in the product cycle (after Final Beta), upper management reversed their earlier decision and decide no
by gfodor 16y ago
Honestly I found this part the most interesting:
> Late in the product cycle (after Final Beta), upper management reversed their earlier decision and decide not to support the B1 chip after all.
After all that work, on compilers, processes, tools, etc, it was all scrapped anyway. I wonder if there is a lesson to be learned there. (The usual startup-mantras don't really seem to apply!)
- rlpb 16y agoIf it was the right thing to do, then it was the right thing to do despite any previous work. See: http://en.wikipedia.org/wiki/Sunk_cost http://en.wikipedia.org/wiki/Sunk_cost
- paulgerhardt 16y agoSure, decisions were made on the battlefield that at the time and considering the circumstances were optimal (for local maxima.) This kind of glib remark doesn't really prevent anyone else from making the same mistakes. I don't know what the correct approach should have been, but it seems that changes to support other architectures should have been made independently from other fixes and tagged such that they could have been backed out rather than contaminating your source and pushing it to gold. (Remember this is before the days of Windows Update!)
- rlpb 16y ago> This kind of glib remark doesn't really prevent anyone else from making the same mistakes. Where's the mistake? Wouldn't backing out the workarounds introduce new potential for error due to other changes made after the workarounds were put in? Why do the workarounds need to be backed out? They don't break anything. If it ain't broke... My point is that it wasn't really a mistake; it was a change in circumstance.