SHA-0 was published in 1993
SHA-1 was published in 1995
SHA-2 was published in 2001
How long before SHA-2 was published was it designed? 1.3,5,10 years?
How long before it was published was it cracked?
But back to the topic on hand.
Ultimately there should be a set of procedures for switching the community from one hash algorithm to another, but there's really no reason to do it proactively.
Do you think it should be done proactively to increase the bit count IE 256 to 512 to 1024?
IE SHA 256 to SHA 512?
A quote I found online about this topic
"I don’t have much to say about Bitcoin, but I do find it interesting that, however briefly, there exists a direct monetary incentive to break SHA-256. To my knowledge, there is no provision for changing the cryptographic hash function for Bitcoin."