Sun rejects orthogonal persistence for Java

[Bill Frantz]

At 09:57 AM 9/16/1999 -0700, Mark S. Miller wrote:
>In an EROS context, orthogonal persistence is still clearly motivated by 
>performance, but the EROS programmer must be careful not to be seduced away 
>from crafting a funnel.  On platforms such as un-augmented Java, where the 
>only available persistence choice is serialization, reusing the funnel for 
>persistence gives a huge performance *win*, so orthogonal persistence would 
>only subtract value.

I note that in KeyKOS, our conclusion was that most objects were transient
enough that the strategy of having only new objects run the new code was
"good enough".  Many of the other objects, like the indexed record object,
would reveal all their important internal state thru their API, so upgrade
consisted of taking the old version and copying its state into a new version.

Jonathan's comments about the COFF file format remind me of the first
object we upgraded which didn't enjoy the above advantages.  It was almost
exactly like patching a running production system thru the console switches.

Bill