ietf-openproxy
[Top] [All Lists]

Re: WG Review: Open Pluggable Edge Services (opes)

2001-06-18 11:17:01

> I have several concerns about this charter.

> I cannot tell whether those concerns are merely due to ambiguities in
> the charter.  I hope this is the case, and that the proponents of
> the group will be willing to clarify the charter to narrow the apparent
> scope of this proposed group.

Keith, I think you are being unduly paranoid here. You are reading using the
"that which isn't explicitly disallowed is not only permitted, but condoned".
This sort of analysis is neither valid, useful, or necessary.

> > The Open Plugable Edge Service (OPES) WG primary task is to define a
> > protocol to be used to extend participating transit intermediaries to
> > incorporate services executed on application data transported by HTTP.

> What are "participating transit intermediaries"?

Application area working group -> application layer intermediaries.

> To me, an intermediary is anything that handles network traffic and
> which is not an endpoint, regardless of what layer of the stack it
> operates on, or on whose behalf it is operating.

In the general context of the IETF, perhaps. But in the context of an
application area WG, nonsense. And I don't think stating in an application WG
charter that it is scoped to the application area is necessary.

> Do the charter authors intend that this group's purview include bridges,
> routers, NATs, proxies, firewalls, gateways, etc?

The charter covers none of these things.

> how about things
> that operate on multiple layers ?

The application layer aspects of such things could be covered, but not the
other parts.

> (If this group is approved are we abandoning the idea of functional
> separation between layers in the internet architecture? )

Of course not.

> Are "intermediaries" intended to include interception proxies?

It depends on whether or not they are participating in the overall
architecture. Stuff that intercepts willy-nilly without the various other
parties being aware of it isn't part of the picture.

> Does it matter whether the intermediary is operating on the client's
> behalf, on the content-provider's behalf, or on behalf of some
> communications provider?

Part of the work of the group is to clarify this issue.

> Re: "application data transported by HTTP" -

> Does this mean that this group's purview is limited to things that use
> the HTTP protocol?

Basically, yes.

> Are we now trying to encourage people to layer applications over HTTP,
> in spite of the concerns outlined in draft-moore-using-http-01.txt ?

Not intentionally. However, to the extent that standardizing additional
capabilities for HTTP will make HTTP more attactive for some uses, this could
happen.

The alternative -- not adding needed functionality to various protocols because
we're afraid that making them more attactive will encourage misuse, is
untenable.

> > The protocol provides a framework for integrating arbitrary services
> > into arbitrary intermediaries.

> The word "arbitrary" here is scary.  Is this group really being given
> a charter to allow it to entirely discard the internet architecture?

Of course not.

> In other words, is this group going to define standard ways to say

> ...

Keith, I stopped reading here, as your subsequent comments are built on
your earlier erroneous assumptions to such a large extent that no
further commentary from me appears to be useful.

                                Ned