I don't quite understand the Roy's Kerr argument. I have Hawking's and Ellis books in front of me and in chapter 8 they define very precisely what a singularity

I don't quite understand the Roy's Kerr argument. I have Hawking's and Ellis books in front of me and in chapter 8 they define very precisely what a singularity and they argue that finite length timelike or null geodisics incompleteness is the right concept to use when talking about singularities. quote Timelike geodesic incompleteness has an immediate physical significance in that it presents the possibility that there could be freely moving observers or particles whose history did not exist after or before a finite interval of time. This would appear and even more objectionable feature that infinite curvature and so it is appropriate to regard such a space a singular. endquote It is nice that Kerr's apparently found examples of geodesic incomplete spaces with bounded curvature and metric if I understand correctly, but I don't understand the "attack" on Penrose and Hawking work. At least in the book or in the original Penrose article they don't claim that geodesic incompleteness implies unboundedness of the metric or the curvature. On the contrary as far as I know they even argue (in the same book) that if the metric or the curvature is infinite at some point the manifold is not extensible and you could just remove the point from the manifold while with geodesic incompleteness the state of affair is worst cause you cannot in principle remove it. Finally I've never heard a physicist believes that singularities are real, they are just a symptoms that the theory reaches its limit.