[e-lang] Why optUnget is built from optUncall, rather than the reverse

[Kevin Reid]

On May 30, 2005, at 21:12, Mark Miller wrote:

> Kevin,
>
> In the XXX comments in your lisp implementation, you ask why a 
> Loader's optUnget is implemented in terms of its optUncall, rather 
> than the other way around. Does the iAmHome3 example in 
> <http://www.erights.org/data/serial/jhu-paper/subgraph-security.html> 
> successfully answer this question?

I believe I understand the example, but I do not see the relevance to 
this code, as the difference is entirely in the implementation and not 
the interface.

> When parent loaders do not represent greater authority than their 
> children, then perhaps this patterns isn't needed. But since it's 
> needed sometimes, I think it's best if loaders generally follow this 
> pattern.

I'm not sure what you mean by "parent" and "child" loaders. Do you mean 
package loaders / directories?

-- 
Kevin Reid                            <http://homepage.mac.com/kpreid/>