Michael Rueger <m.rueger at acm.org> wrote:
> Just a cautious reminder from the module disaster:
> 
> back at OOPSLA one of the main points was to "never" put a packaging 
> system into the base system again for all the reasons in the above 
> discussion.
I missed the OOPSLA discussions, and maybe because of that I don't see
how this statement is relevant to the relevant case. If I am missing
some information, please let me know.
In the mean time: 
- we're not talking about modifying the base system. Just about what
gets loaded when specific packages get loaded. Different packagers still
get to decide what format they want to use. For example, I hadn't
thought about it before, but I probably wouldn't release a package such
as SMLoader using mcvs (though I am going to develop it using MC), since
it is very basic and shouldn't have serious requirements. 
- If by "never put a packaging system into the base system again" you
mean that .st compatibility should never be removed/broken in favor of
using a module/package system exclusively, the I agree. MC doesn't
remove or modify the old filein system. 

> The ability of Ginsu and Monticello to deal with a semantic model 
> without loading the code first still needs to know where the code 
> originally came from. 
What? I didn't understand this at all. Please define "where the code
originally came from". Do you mean who the author is? or whether it is a
modification of some old code? or something else? I can't find any
meaning under which I understand what you're saying.

> So how would a generic attributation of the 
> standard file out format work?
> Assuming you don't want to extend something like the timestamp method.
Maybe because of the above, I have no idea what you are talking about.
AFAIU, nobody is talking about changing the standard file out format.

Daniel