\documentclass{slides}
%[24pt] - doesn't work
% Use landscape mode -- must run dvips with ``-t landscape''.
\usepackage{times,psfig,landscape}
%\fontsize{30}{30}


\addtolength{\topmargin}{-0.8in}
\addtolength{\textheight}{0.25in}
\addtolength{\textwidth}{0.75in}

\newcommand{\titleskip}{\bigskip \bigskip}

\begin{document}
%\fontsize{24}{24}
%\Large
\large

\newcommand{\dash}{\hspace*{0.2in} --  }
\newcommand{\blnk}{\hspace*{0.6in}}
%% Show some figures from actual validation tests.
%%%%%%%%%%%%%%%
\begin{slide}{}
{\bf IAB Architectural and Policy Considerations for OPES...}\\
{\bf and Wider Architectural Concerns}\\
\\
\\
Open IAB Plenary\\
December 13, 2001\\
Salt Lake City IETF\\
\\
\end{slide}
%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%
\begin{slide}{}
{\bf The initial question, for OPES:}\\
What are the
architectural and policy considerations in chartering the OPES WG?

{\bf The wider questions:}\\
What are the robustness, integrity, and privacy considerations for
IETF protocols?
\end{slide}
%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%
\begin{slide}{}
\centerline{\bf What is OPES (Open Pluggable Edge Services)?}

%application-level

$\bullet$ Server-authorized services (e.g., at the web cache):\\
\dash Assembling personalized web pages;\\
\dash Adding regional information to web pages.

$\bullet$ Client-authorized services:\\
\dash Virus scanning;\\
\dash Language translation.\\
\dash Adaptation for limited bandwidth;

\end{slide}
%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%
\begin{slide}{} 
\centerline{\bf The history of chartering OPES:}

$\bullet$ OPES at this IETF had its fourth IETF meeting.

$\bullet$  {\bf Position papers:}
``OPES would reduce both the integrity,
and the perception of integrity, of communications over the Internet,
and would significantly increase uncertainly about what might have
been done to content as it moved through the network.''
[CDT01, August 2001], 

\end{slide}
%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%
\begin{slide}{}

$\bullet$ {\bf Email:}
``OPES is deeply evil and the IETF should stay far, far away from this
hideous abomination''.
[ODell01, July 2001]

$\bullet$ The IAB got in on the loop after the third BOF:\\
draft-iab-opes-01.txt, to be an Informational RFC.

\end{slide}
%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%
\begin{slide}{}

{\bf The IAB document on OPES:}

$\bullet$ It is a recommendation to the IESG in chartering OPES.

$\bullet$ The IAB draft does not have answers, it mostly describes issues
that should be addressed.

$\bullet$ Somewhat like "Security Considerations": the WG has to
convincingly argue that the issues have been addressed.  

\end{slide}
%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%
\begin{slide}{}

\centerline{\bf Questions raised in the IAB OPES document:}

$\bullet$ (1) Is there {\bf architectural benefit} to putting 
services inside the
network (e.g., at the application-level web cache)?

Answer: yes.\\
Potential benefits:\\
\dash reduced delay to user; \\
\dash less wasted bandwidth, etc.

\end{slide}
%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%
\begin{slide}{}

$\bullet$ (2) Do the architectural benefits outweigh the {\bf costs}? 

Potential costs:\\ 
\dash data integrity?\\
\dash difficulties in diagnosing failures?, etc.

$\bullet$ (3) Does OPES have an impact on the {\bf application layer  
addressing} architecture?

\end{slide}
%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%
\begin{slide}{}

\centerline{\bf Related work in the IETF:}

{\bf RFC 3135}: Performance Enhancing Proxies \\
Intended to Mitigate Link-Related
Degradations 

$\bullet$ Sections on:\\ 
\dash Security Implications, \\
\dash Fate Sharing, \\
\dash End-to-end Reliability,\\
\dash End-to-end Failure Diagnostics, etc.

\end{slide}
%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%
\begin{slide}{}

\centerline{\bf So on to the more general issues:}

$\bullet$ The IAB is planning on writing a document about general
architectural considerations.

$\bullet$ We would like feedback from this group. 

\end{slide}
%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%
\begin{slide}{}

\centerline{\bf General architectural considerations:}

$\bullet$
{\bf Justifying the solution:}\\
\dash Why are you proposing this solution, at this level of the
architecture, instead of something else?\\
\dash Interactions with other layers?\\
\dash Would this restrict future developments?  In what ways?\\

\end{slide}
%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%
\begin{slide}{}

$\bullet$ {\bf Long-term vs. short-term solutions:}\\
\dash Is this proposal the best long-term solution?\\
\dash If this is a short-term solution, what are costs?\\
\dash Interactions with other short-term fixes?\\
\dash Is there an exit plan?

\end{slide}
%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%
\begin{slide}{}

$\bullet$ {\bf General robustness concerns:}\\
\dash Robustness to node failures.\\
\dash Robustness to compromised nodes?\\
\dash Robustness to imperfect or defective implementations?\\
\dash Robust performance in a wide range of environments?

\end{slide}
%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%
\begin{slide}{}

$\bullet$ {\bf Balancing competing interests:}\\
\dash Are you protecting the interests of other parties while you are
protecting your own interests?\\  
\dash Are you protecting the interests of the future?

\end{slide}
%%%%%%%%%%%%%%%
%%%%%%%%%%%%%%%
\begin{slide}{}


$\bullet$ {\bf Tragedy of the commons?}\\
\dash Will tragedy of the commons result if everyone 
implements this proposal?

\end{slide}
%%%%%%%%%%%%%%%
\end{document}