Skip to main content
5 events
when toggle format what by license comment
Feb 16, 2013 at 6:10 vote accept skupers
Feb 15, 2013 at 19:43 comment added Allen Hatcher I switched to the PL category to use the Alexander trick to avoid dealing with smooth local complications, but the PL category has its own subtleties which deserved more attention than I gave them. In any event, the last time I thought about these things, many years ago, I was optimistic that there were no insuperable barriers to fixing the PL proof.
Feb 15, 2013 at 19:43 comment added Allen Hatcher Thanks for this very illuminating discussion. It's been years since I thought much about these things. I always intended to go back and fill in what was missing in my original proof and correct mistakes in it, but somehow I don't seem to getting any closer to doing this. Actually I would have preferred working in the smooth category from the start, but there were obstacles coming from the complexity of higher-codimension singularities of smooth real-valued functions. Igusa later showed how to avoid these obstacles, staying purely within the smooth category. (continued below)
Feb 15, 2013 at 17:19 history edited Tom Goodwillie CC BY-SA 3.0
added 13 characters in body
Feb 15, 2013 at 3:27 history answered Tom Goodwillie CC BY-SA 3.0