>
> No, what you provided was a "demonstration" where you complained about
> a registry for one system, and assumed it for the other. This is just
> bad-faith argumentation.
>
>
First, I've shown that a package manager is necessary for non-trivial
module graphs, and therefore we should not admit claims that the current
design works out-of-the-box for non-trivial scenarios. That's all.
Second, I've attempted to show that:
1) Given the right infrastructure, with "logical names" the user must
install a package manager, and then invoke the package manager to copy
modules into the correct folder, ala NPM. This must happen before any code
works.
2) Given the right infrastructure, with URLs the user can just reference a
URL and everything works. Later in the development process, the user can
invoke a package manager to copy everything to a local server and set up
the necessary loader hooks.
I don't think that this line of reasoning constitutes bad faith, but I
apologize if it seems that way. I would like to provide more details for
my thoughts behind #2, but I'll do so through a separate link.
Thanks for your time,
{ Kevin }
_______________________________________________
es-discuss mailing list
es-discuss@mozilla.org
https://mail.mozilla.org/listinfo/es-discuss
