On 20 Mar 2005 14:35:15 -0800, Thomas Bushnell BSG <[EMAIL PROTECTED]> wrote: > > The reason that filesystems do not have user context is because I was > not sufficiently far-sighted at the time to realize the full > flexibility of the translator concept I had created. Now that we know > more about that flexibility, it would be nice to start figuring out > how to improve it. >
Let's take the simple example of a symlink translator noticing that the file it points to has gone away. A user would register for interaction with the symlink translator, to give it permission to bug you when it wanted to. Is this act of registering itself interactive? Which translators do we care about at any given time? In order to register with the symlink translator we'd need to be able to specify which symlink translator we're registering for interactivity with. Is a path sufficient? (What about stacked translators?) Suppose we register for the "link-broken" event from our symlink translator instance, giving it permission to bug us when the link breaks. What other events does the symlink (or any other translator) provide, and how (as a user) do I find out? Suppose the "link-broken" event fires, and it asks me what I want to do (let the link dangle, or remove it, or link to another file). Does anybody else get nagged also? If so, is it a democratic decision? It's a silly example, but when a translator affects many users directly, they all (potentially) want to be involved interactively. And obviously before the user logs out, an unregistering automatically happens. Ben _______________________________________________ Bug-hurd mailing list Bug-hurd@gnu.org http://lists.gnu.org/mailman/listinfo/bug-hurd
