Re: Namespaces ? Norman Hardy (norm@netcom.com)
Tue, 4 Jul 2000 19:05:39 -0700

At 1:02 +0200 00/07/05, Vincent Legoll wrote:
>Hello,
>
>I was only reader until now, but I want to know
>if I really understood the principles on which eros
>is based.
>
>I think that namespace is only needed when an user
>want to tell some application what data he want to
>work on, right ? (by data I mean an unit of data like
>an email, an image, a text docmuent, etc...)

Yes, but I would say "tell the system what data I want the application to have". The application does not learn its name!

>But data, is already represented by something, a
>capability to the memory space containing this data,
>no ?

Yes, virtually always.

>So namespaces are used by users to access their data
>stored in their accounts. (I see an account just as a data
>container, so it will be a capability itself, holded by the
>authentication process, which gives a copy of it to a
>shell-like application, after the authentication has
>occured)

Yes. In Keykos a password for a user name got you connected to a comand line interpreter (shell) with a builtin access to the corresponding user's directory, which mapped the names he assigned to capabilities that he owned. The shell would invoke those capabilities on command, passing other data and capabilities as specified.

>For me a namespace is a capability holder, in which the
>user choose a particular one to select some data.

some data or other authority. Some directories (namespaces) might hold a capability to control the lights in the parking lot. The directory is oblivious to this, the capability knows what it does.

>Please excuse me for my poor english, and perhaps for
>my poor understanding of eros.

Some of the recent mail has used "namespace" to include things like array indexes. That is poor when reasoning about human interfaces, but it is good when reasoning about formal security properties.
In your mail and my response it has the restricted, more conventional meaning.

>--
>Vincent Legoll
>French student in computer science.
>legoll@online.fr

Norman Hardy <http://www.mediacity.com/~norm>