Dynamic Scope [Was: Existential operator]

# P T Withington (14 years ago)

On 2011-04-15, at 21:48, Mark S. Miller wrote:

                                           Why dynamic scoping was
attractive and why it turns out to be bad is one of the most important lessons from the history of language design.

It's a power tool, and can be misused, but it still has adherents. Scheme (and Dylan) have a more-controlled version in their fluid-let construct (which leaves out the "pervasive special" feature of Lisp).

On 2011-04-15, at 21:48, Mark S. Miller wrote:

>                                                Why dynamic scoping was
> attractive and why it turns out to be bad is one of the most important
> lessons from the history of language design.

It's a power tool, and can be misused, but it still has adherents.  Scheme (and Dylan) have a more-controlled version in their `fluid-let` construct (which leaves out the "pervasive special" feature of Lisp).

# David Herman (14 years ago)

Forms like `fluid-let' don't actually make dynamic decisions about scope -- they just mutate an existing, statically-scoped variable. We're really just talking about dynamic decisions about where a variable is bound, not what the current value of its binding is. That said, I happen to know Mark thinks fluids are a bad idea too. :)

Regardless, we're waaaay off-topic and diverging further still...

Forms like `fluid-let' don't actually make dynamic decisions about *scope* -- they just mutate an existing, statically-scoped variable. We're really just talking about dynamic decisions about *where a variable is bound*, not *what the current value of its binding is*. That said, I happen to know Mark thinks fluids are a bad idea too. :)

Regardless, we're waaaay off-topic and diverging further still...

Dave

On Apr 16, 2011, at 3:08 PM, P T Withington wrote:

> On 2011-04-15, at 21:48, Mark S. Miller wrote:
> 
>>                                               Why dynamic scoping was
>> attractive and why it turns out to be bad is one of the most important
>> lessons from the history of language design.
> 
> It's a power tool, and can be misused, but it still has adherents.  Scheme (and Dylan) have a more-controlled version in their `fluid-let` construct (which leaves out the "pervasive special" feature of Lisp).