[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [sc-users] SuperCollider 3.3. alpha release





On Fri, Dec 26, 2008 at 4:29 PM, James Harkins <jamshark70@xxxxxxxxx> wrote:
On Fri, Dec 26, 2008 at 11:12 AM, Eric Lyon <audiodidact@xxxxxxxxx> wrote:
> I disagree. It's a catastrophic response to a possibly minor problem

It's a catastrophic response to a major problem.

I could write you a file defining a new Object class - put it in your
extension folder and the whole VM goes belly up.

I'm willing to bet that that does not happen too often :) And anyway, redefining core SC classes in 3rd party extensions should be discouraged IMO.
 

There is no way for the compiler to know whether a class name conflict
is minor or catastrophic, so the catastrophic response is correct.

It's not "correct," it's a design decision. We disagree on whether it is an optimal design decision. As a counter-example, MaxMSP loads the first external or abstraction it finds in its search path, either at startup or when referenced from a Max patch file. (It prints a warning when it finds duplicate abstractions.) No one has ever complained about this behavior, but I am certain that many people would complain if MaxMSP were to crash in every instance where it found a duplicate external or abstraction file.

Eric