[Python-Dev] Re: Alternative implementation of string interning

[Barry A. Warsaw]

>>>>> "TD" == Timothy Delaney <tdelaney@avaya.com> writes:

    TD> The #1 most important consideration here is backwards
    TD> compatibility IMO.  Whilst I would be personally unaffected by
    TD> this change (allowing interned strings to be collected), we've
    TD> already had examples of people and code that would be.

I still think most applications don't care about interned strings, and
they really don't care whether they're immortal or not.  Long running
apps probably do care, but for them, I'd rather see the application
writers have to take an explicit action to free the intern strings.
Only they are going to know whether they're depending on immortal
interns, and when it's "safe" and prudent to reclaim them.

-Barry