Network Working Group                                          A. Barbir
Request for Comments: 3914                               Nortel Networks
Category: Informational                                      A. Rousskov
                                                The Measurement Factory
                                                           October 2004


          Open Pluggable Edge Services (OPES) Treatment of
                          IAB Considerations

Status of this Memo

  This memo provides information for the Internet community.  It does
  not specify an Internet standard of any kind.  Distribution of this
  memo is unlimited.

Copyright Notice

  Copyright (C) The Internet Society (2004).

Abstract

  IETF Internet Architecture Board (IAB) expressed nine architecture-
  level considerations for the Open Pluggable Edge Services (OPES)
  framework.  This document describes how OPES addresses those
  considerations.

























Babir & Rousskov             Informational                      [Page 1]

RFC 3914          OPES Treatment of IAB Considerations      October 2004


Table of Contents

  1.  Introduction . . . . . . . . . . . . . . . . . . . . . . . . .  2
  2.  Terminology  . . . . . . . . . . . . . . . . . . . . . . . . .  3
  3.  Consideration (2.1) 'One-party consent'  . . . . . . . . . . .  3
  4.  Consideration (2.2) 'IP-layer communications'  . . . . . . . .  4
  5.  Notification Considerations  . . . . . . . . . . . . . . . . .  5
      5.1.  Notification versus trace. . . . . . . . . . . . . . . .  6
      5.2.  An example of an OPES trace for HTTP . . . . . . . . . .  8
      5.3.  Consideration (3.1) 'Notification' . . . . . . . . . . .  9
      5.4.  Consideration (3.2) 'Notification' . . . . . . . . . . . 10
  6.  Consideration (3.3) 'Non-blocking' . . . . . . . . . . . . . . 10
  7.  Consideration (4.1) 'URI resolution' . . . . . . . . . . . . . 11
  8.  Consideration (4.2) 'Reference validity' . . . . . . . . . . . 11
  9.  Consideration (4.3) 'Addressing extensions'  . . . . . . . . . 12
  10. Consideration (5.1) 'Privacy'  . . . . . . . . . . . . . . . . 12
  11. Consideration 'Encryption' . . . . . . . . . . . . . . . . . . 12
  12. Security Considerations  . . . . . . . . . . . . . . . . . . . 13
  13. Compliance . . . . . . . . . . . . . . . . . . . . . . . . . . 13
  14. References . . . . . . . . . . . . . . . . . . . . . . . . . . 14
      14.1. Normative References . . . . . . . . . . . . . . . . . . 14
      14.2. Informative References . . . . . . . . . . . . . . . . . 14
  Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 15
  Full Copyright Statement . . . . . . . . . . . . . . . . . . . . . 16

1.  Introduction

  The Open Pluggable Edge Services (OPES) architecture [RFC3835],
  enables cooperative application services (OPES services) between a
  data provider, a data consumer, and zero or more OPES processors.
  The application services under consideration analyze and possibly
  transform application-level messages exchanged between the data
  provider and the data consumer.

  In the process of chartering OPES, the IAB made recommendations on
  issues that OPES solutions should be required to address.  These
  recommendations were formulated in the form of a specific IAB
  considerations document [RFC3238].  In that document, IAB emphasized
  that its considerations did not recommend specific solutions and did
  not mandate specific functional requirements.  Addressing an IAB
  consideration may involve showing appropriate protocol mechanisms or
  demonstrating that the issue does not apply.  Addressing a
  consideration does not necessarily mean supporting technology implied
  by the consideration wording.







Babir & Rousskov             Informational                      [Page 2]

RFC 3914          OPES Treatment of IAB Considerations      October 2004


  The primary goal of this document is to show that all formal IAB
  recommendations are addressed by OPES, to the extent that those
  considerations can be addressed by an IETF working group.  The
  limitations of OPES working group to address certain aspects of IAB
  considerations are also explicitly documented.

  IAB considerations document [RFC3238] contains many informal
  recommendations.  For example, while the IAB informally requires OPES
  architecture to "protect end-to-end data integrity by supporting
  end-host detection and response to inappropriate behavior by OPES
  intermediaries", the IAB has chosen to formalize these requirements
  via a set of more specific recommendations, such as Notification
  considerations addressed in Section 5.3 and Section 5.4 below.  OPES
  framework addresses informal IAB recommendations by addressing
  corresponding formal considerations.

  There are nine formal IAB considerations [RFC3238] that OPES has to
  address.  In the core of this document are the corresponding nine
  "Consideration" sections.  For each IAB consideration, its section
  contains general discussion as well as references to specific OPES
  mechanisms relevant to the consideration.

2.  Terminology

  This document does not introduce any new terminology but uses
  terminology from other OPES documents.

3.  Consideration (2.1) 'One-party consent'

  "An OPES framework standardized in the IETF must require that the use
  of any OPES service be explicitly authorized by one of the
  application-layer end-hosts (that is, either the content provider or
  the client)." [RFC3238]

  OPES architecture requires that "OPES processors MUST be consented to
  by either the data consumer or data provider application" [RFC3835].
  While this requirement directly satisfies IAB concern, no requirement
  alone can prevent consent-less introduction of OPES processors.  In
  other words, OPES framework requires one-party consent but cannot
  guarantee it in the presence of incompliant OPES entities.

  In [RFC3897], the OPES architecture enables concerned parties to
  detect unwanted OPES processors by examining OPES traces.  While the
  use of traces in OPES is mandatory, a tracing mechanism on its own
  cannot detect processors that are in violation of OPES
  specifications.  Examples include OPES processors operating in
  stealth mode.  However, the OPES architecture allows the use of
  content signature to verify the authenticity of performed



Babir & Rousskov             Informational                      [Page 3]

RFC 3914          OPES Treatment of IAB Considerations      October 2004


  adaptations.  Content signatures is a strong but expensive mechanism
  that can detect any modifications of signed content provided that the
  content provider is willing to sign the data and that the client is
  willing to either check the signature or relay received content to
  the content provider for signature verification.

  OPES entities may copy or otherwise access content without modifying
  it.  Such access cannot be detected using content signatures.  Thus,
  "passive" OPES entities can operate on signed content without the
  data consumer or provider consent.  If content privacy is a concern,
  then content encryption can be used.  A passive processor is no
  different from any intermediary operating outside of OPES framework.
  No OPES mechanism (existing or foreseeable) can prevent non-modifying
  access to content.

  In summary, the one-party consent is satisfied by including the
  corresponding requirement in the IAB architecture document.  That
  requirement alone cannot stop incompliant OPES entities to perform
  consent-less adaptations, but OPES framework allows for various means
  of detecting and/or preventing such adaptations.  These means
  typically introduce overheads and require some level of producer-
  consumer cooperation.

4.  Consideration (2.2) 'IP-layer communications'

  "For an OPES framework standardized in the IETF, the OPES
  intermediary must be explicitly addressed at the IP layer by the end
  user" [RFC3238].

  The OPES architecture requires that "OPES processors MUST be
  addressable at the IP layer by the end user (data consumer
  application)" [RFC3835].  The IAB and the architecture documents
  mention an important exception: addressing the first OPES processor
  in a chain of processors is sufficient.  That is, a chain of OPES
  processors is viewed as a single OPES "system" at the address of the
  first chain element.

  The notion of a chain is not strictly defined by IAB.  For the
  purpose of addressing this consideration, a group of OPES processors
  working on a given application transaction is considered.  Such a
  group would necessarily form a single processing chain, with a single
  "exit" OPES processor (i.e., the processor that adapted the given
  message last).  The OPES architecture essentially requires that last
  OPES processor to be explicitly addressable at the IP layer by the
  data consumer application.  The chain formation, including its exit
  point may depend on an application message and other dynamic factors
  such as time of the day or system load.




Babir & Rousskov             Informational                      [Page 4]

RFC 3914          OPES Treatment of IAB Considerations      October 2004


  Furthermore, if OPES processing is an internal processing step at a
  data consumer or a data provider application side, then the last OPES
  processor may reside in a private address space and may not be
  explicitly addressable from the outside.  In such situations, the
  processing side must designate an addressable point on the same
  processing chain.  That designated point may not be, strictly
  speaking, an OPES processor, but it will suffice as such as far as
  IAB considerations are concerned -- the data consumer application
  will be able to address it explicitly at the IP layer and it will
  represent the OPES processing chain to the outside world.

  Designating an addressable processing point avoids the conflict
  between narrow interpretation of the IAB consideration and real
  system designs.  It is irrational to expect a content provider to
  provide access to internal hosts participating in content generation,
  whether OPES processors are involved or not.  Moreover, providing
  such access would serve little practical purpose because internal
  OPES processors are not likely to be able to answer any data consumer
  queries, being completely out of content generation context.  For
  example, an OPES processor adding customer-specific information to
  XML pages may not understand or be aware of any final HTML content
  that the data consumer application receives and may not be able to
  map end user request to any internal user identification.  Since OPES
  requires the end of the message processing chain to be addressable,
  the conflict does not exist.  OPES places no requirements on the
  internal architecture of data producer systems while requiring the
  entire OPES-related content production "system" to be addressable at
  the IP layer.

  Private Domain    | Public Domain     | Private Domain
                    |                   |
  +--------------+  |             +-------------+      +--------+
  | Data         |  |             | OPES System |      |Data    |
  | Consumer     |<--- network -->| with public |<---->|Provider|
  | Application  |  |             | IP address  |      |App     |
  +--------------+  |             +-------------+      +--------+
                    |                   |
                    |                   |

                               Figure 1

5.  Notification Considerations

  This section discusses how OPES framework addresses IAB Notification
  considerations 3.1 and 3.2.






Babir & Rousskov             Informational                      [Page 5]

RFC 3914          OPES Treatment of IAB Considerations      October 2004


5.1.  Notification versus trace

  Before specific considerations are discussed, the relationship
  between IAB notifications and OPES tracing has to be explained.  OPES
  framework concentrates on tracing rather than notification.  The OPES
  Communications specification [RFC3897] defines "OPES trace" as
  application message information about OPES entities that adapted the
  message.  Thus, OPES trace follows the application message it traces.
  The trace is for the recipient of the application message.  Traces
  are implemented as extensions of application protocols being adapted
  and traced.

  As opposed to an OPES trace, provider notification (as implied by
  IAB) notifies the sender of the application message rather than the
  recipient.  Thus, notifications propagate in the opposite direction
  of traces.  Supporting notifications directly would require a new
  protocol.  Figure 2 illustrates the differences between a trace and
  notification from a single application message point of view.

  sender --[message A]--> OPES --[message A']--> recipient
     ^                       V                             [with trace]
     |                       |
     +-<-- [notification] ---+

                               Figure 2

  Since notifications cannot be piggy-backed to application messages,
  they create new messages and may double the number of messages the
  sender has to process.  The number of messages that need to be
  proceed is larger if several intermediaries on the message path
  generate notifications.  Associating notifications with application
  messages may require duplicating application message information in
  notifications and may require maintaining a sender state until
  notification is received.  These actions increase the performance
  overhead of notifications.

  The level of available details in notifications versus provider
  interest in supporting notification is another concern.  Experience
  shows that content providers often require very detailed information
  about user actions to be interested in notifications at all.  For
  example, Hit Metering protocol [RFC2227] has been designed to supply
  content providers with proxy cache hit counts, in an effort to reduce
  cache busting behavior which was caused by content providers desire
  to get accurate site "access counts".  However, the Hit Metering
  protocol is currently not widely deployed because the protocol does
  not supply content providers with information such as client IP
  addresses, browser versions, or cookies.




Babir & Rousskov             Informational                      [Page 6]

RFC 3914          OPES Treatment of IAB Considerations      October 2004


  Hit Metering experience is relevant because Hit Metering protocol was
  designed to do for HTTP caching intermediaries what OPES
  notifications are meant to do for OPES intermediaries.  Performance
  requirements call for state reduction via aggregation of
  notifications while provider preferences call for state preservation
  or duplication.  Achieving the right balance when two sides belong to
  different organizations and have different optimization priorities
  may be impossible.

  Thus, instead of explicitly supporting notifications at the protocol
  level, OPES concentrates on tracing facilities.  In essence, OPES
  supports notifications indirectly, using tracing facilities.  In
  other words, the IAB choice of "Notification" label is interpreted as
  "Notification assistance" (i.e., making notifications meaningful) and
  is not interpreted as a "Notification protocol".

  The above concerns call for making notification optional.  The OPES
  architecture allows for an efficient and meaningful notification
  protocol to be implemented in certain OPES environments.  For
  example, an OPES callout server attached to a gateway or firewall may
  scan outgoing traffic for signs of worm or virus activity and notify
  a local Intrusion Detection System (IDS) of potentially compromised
  hosts (e.g., servers or client PCs) inside the network.  Such
  notifications may use OPES tracing information to pinpoint the
  infected host (which could be another OPES entity).  In this example,
  notifications are essentially sent back to the content producer (the
  local network) and use OPES tracing to supply details.

  Another environment where efficient and meaningful notification using
  OPES tracing is possible are Content Delivery Networks (CDNs).  A CDN
  node may use multiple content adaptation services to customize
  generic content supplied by the content producer (a web site).  For
  example, a callout service may insert advertisements for client-local
  events.  The CDN node itself may not understand specifics of the ad
  insertion algorithm implemented at callout servers.  However, the
  node may use information in the OPES trace (e.g., coming from the
  callout service) to notify the content producer.  Such notifications
  may be about the number of certain advertisements inserted (i.e., the
  number of "impressions" delivered to the customer) or even the number
  of ad "clicks" the customer made (e.g., if the node hosts content
  linked from the ads).  Callout services doing ad insertion may lack
  details (e.g., a customer ID/address or a web server authentication
  token) to contact the content producer directly in this case.  Thus,
  OPES trace produced by an OPES service becomes essential in enabling
  meaningful notifications that the CDN node sends to the content
  producer.





Babir & Rousskov             Informational                      [Page 7]

RFC 3914          OPES Treatment of IAB Considerations      October 2004


5.2.  An example of an OPES trace for HTTP

  The example below illustrates adaptations done to HTTP request at an
  OPES processor operated by the client ISP.  Both original (as sent by
  an end user) and adapted (as received by the origin web server)
  requests are shown.  The primary adaptation is the modification of
  HTTP "Accept" header.  The secondary adaptation is the addition of an
  OPES-System HTTP extension header [I-D.ietf-opes-http].

  GET /pub/WWW/ HTTP/1.1
  Host: www.w3.org
  Accept: text/plain
                               Figure 3

  ... may be adapted by an ISP OPES system to become:

  GET /pub/WWW/ HTTP/1.1
  Host: www.w3.org
  Accept: text/plain; q=0.5, text/html, text/x-dvi; q=0.8
  OPES-System: http://www.isp-example.com/opes/?client-hash=1234567

                               Figure 4

  The example below illustrates adaptations done to HTTP response at an
  OPES intermediary operated by a Content Distribution Network (CDN).
  Both original (as sent by the origin web server) and adapted (as
  received by the end user) responses are shown.  The primary
  adaptation is the conversion from HTML markup to plain text.  The
  secondary adaptation is the addition of an OPES-System HTTP extension
  header.

  HTTP/1.1 200 OK
  Content-Length: 12345
  Content-Encoding: text/html

  <html><head><h1>Available Documenta...

                               Figure 5

  ... may be adapted by a CDN OPES system to become:

  HTTP/1.1 200 OK
  Content-Length: 2345
  Content-Encoding: text/plain
  OPES-System: http://www.cdn-example.com/opes/?site=7654&svc=h2t

  AVAILABLE DOCUMENTA...
                               Figure 6



Babir & Rousskov             Informational                      [Page 8]

RFC 3914          OPES Treatment of IAB Considerations      October 2004


  In the above examples, OPES-System header values contain URIs that
  may point to OPES-specific documents such as description of the OPES
  operator and its privacy policy.  Those documents may be
  parameterized to allow for customizations specific to the transaction
  being traced (e.g., client or even transaction identifier may be used
  to provide more information about performed adaptations).  An OPES-
  Via header may be used to provide a more detailed trace of specific
  OPES entities within an OPES System that adapted the message.  Traced
  OPES URIs may be later used to request OPES bypass [RFC3897].

5.3.  Consideration (3.1) 'Notification'

  "The overall OPES framework needs to assist content providers in
  detecting and responding to client-centric actions by OPES
  intermediaries that are deemed inappropriate by the content provider"
  [RFC3238].

  OPES tracing mechanisms assist content providers in detecting
  client-centric actions by OPES intermediaries.  Specifically, a
  compliant OPES intermediary or system notifies a content provider of
  its presence by including its tracing information in the application
  protocol requests.  An OPES system MUST leave its trace [RFC3897].
  Detection assistance has its limitations.  Some OPES intermediaries
  may work exclusively on responses and may not have a chance to trace
  the request.  Moreover, some application protocols may not have
  explicit requests (e.g., a content push service).

  OPES tracing mechanisms assist content providers in responding to
  client-centric actions by OPES intermediaries.  Specifically, OPES
  traces MUST include identification of OPES systems and SHOULD include
  a list of adaptation actions performed on provider's content.  This
  tracing information may be included in the application request.
  Usually, however, this information will be included in the
  application response, an adapted version of which does not reach the
  content provider.  If OPES end points cooperate, then notification
  can be assisted with traces.  Content providers that suspect or
  experience difficulties can do any of the following:

  o  Check whether requests they receive pass through OPES
     intermediaries.  Presence of OPES tracing info will determine
     that.  This check is only possible for request/response protocols.
     For other protocols (e.g., broadcast or push), the provider would
     have to assume that OPES intermediaries are involved until proven
     otherwise.







Babir & Rousskov             Informational                      [Page 9]

RFC 3914          OPES Treatment of IAB Considerations      October 2004


  o  If OPES intermediaries are suspected, request OPES traces from
     potentially affected user(s).  The trace will be a part of the
     application message received by the user software.  If involved
     parties cooperate, the provider(s) may have access to all the
     needed information.  Certainly, lack of cooperation may hinder
     access to tracing information.  To encourage cooperation, data
     providers might be able to deny service to uncooperative users.

  o  Some traces may indicate that more information is available by
     accessing certain resources on the specified OPES intermediary or
     elsewhere.  Content providers may query for more information in
     this case.

  o  If everything else fails, providers can enforce no-adaptation
     policy using appropriate OPES bypass mechanisms and/or end-to-end
     encryption mechanisms.

  OPES detection and response assistance is limited to application
  protocols with support for tracing extensions.  For example, HTTP
  [RFC2616] has such support while DNS over UDP does not.

5.4.  Consideration (3.2) 'Notification'

  "The overall OPES framework should assist end users in detecting the
  behavior of OPES intermediaries, potentially allowing them to
  identify imperfect or compromised intermediaries" [RFC3238].

  OPES tracing mechanisms assist end users in detecting OPES
  intermediaries.  Specifically, a compliant OPES intermediary or
  system notifies an end user of its presence by including its tracing
  information in the application protocol messages sent to the client.
  An OPES system MUST leave its trace [RFC3897].  However, detection
  assistance has its limitations.  Some OPES systems may work
  exclusively on requests and may not have a chance to trace the
  response.  Moreover, some application protocols may not have explicit
  responses (e.g., event logging service).

  OPES detection assistance is limited to application protocols with
  support for tracing extensions.  For example, HTTP [RFC2616] has such
  support while DNS over UDP does not.

6.  Consideration (3.3) 'Non-blocking'

  "If there exists a "non-OPES" version of content available from the
  content provider, the OPES architecture must not prevent users from
  retrieving this "non-OPES" version from the content provider"
  [RFC3238].




Babir & Rousskov             Informational                     [Page 10]

RFC 3914          OPES Treatment of IAB Considerations      October 2004


  "OPES entities MUST support a bypass feature" [RFC3897].  If an
  application message includes bypass instructions and an OPES
  intermediary is not configured to ignore them, the matching OPES
  intermediary will not process the message.  An OPES intermediary may
  be configured to ignore bypass instructions only if no non-OPES
  version of content is available.  Bypass may generate content errors
  since some OPES services may be essential but may not be configured
  as such.

  Bypass support has limitations similar to the two notification-
  related considerations above.

7.  Consideration (4.1) 'URI resolution'

  "OPES documentation must be clear in describing these services as
  being applied to the result of URI resolution, not as URI resolution
  itself" [RFC3238].

  "OPES Scenarios and Use Cases" specification [RFC3752] documents
  content adaptations that are in scope of the OPES framework.
  Scenarios include content adaptation of requests and responses.
  These documented adaptations do not include URI resolution.  In some
  environments, it is technically possible to use documented OPES
  mechanisms to resolve URIs (and other kinds of identifiers or
  addresses).  The OPES framework cannot effectively prevent any
  specific kind of adaptation.

  For example, a CDN node may substitute domain names in URLs with
  CDN-chosen IP addresses, essentially performing a URI resolution on
  behalf of the content producer (i.e., the web site owner).  An OPES
  callout service running on a user PC may rewrite all HTML-embedded
  advertisement URLs to point to a user-specified local image,
  essentially performing a URI redirection on behalf of the content
  consumer (i.e., the end user).  Such URI manipulations are outside of
  the OPES framework scope, but cannot be effectively eliminated from
  the real world.

8.  Consideration (4.2) 'Reference validity'

  "All proposed services must define their impact on inter- and intra-
  document reference validity" [RFC3238].

  The OPES framework does not propose adaptation services.  However,
  OPES tracing requirements include identification of OPES
  intermediaries and services (for details, see "Notification"
  consideration sections in this document).  It is required that





Babir & Rousskov             Informational                     [Page 11]

RFC 3914          OPES Treatment of IAB Considerations      October 2004


  provided identification can be used to locate information about the
  OPES intermediaries, including the description of impact on reference
  validity [RFC3897].

9.  Consideration (4.3) 'Addressing extensions'

  "Any services that cannot be achieved while respecting the above two
  considerations may be reviewed as potential requirements for Internet
  application addressing architecture extensions, but must not be
  undertaken as ad hoc fixes" [RFC3238].

  OPES framework does not contain ad hoc fixes.  This document in
  combination with and other OPES documents should be sufficient to
  inform service creators of IAB considerations.  If a service does URI
  resolution or silently affects document reference validity, the
  authors are requested to review service impact on Internet
  application addressing architecture and work within IETF on potential
  extension requirements.  Such actions would be outside of the current
  OPES framework.

10.  Consideration (5.1) 'Privacy'

  "The overall OPES framework must provide for mechanisms for end users
  to determine the privacy policies of OPES intermediaries" [RFC3238].

  OPES tracing mechanisms allow end users to identify OPES
  intermediaries (for details, see "Notification" consideration
  sections in this document).  It is required that provided
  identification can be used to locate information about the OPES
  intermediaries, including their privacy policies.

  The term "privacy policy" is not defined in this context (by IAB or
  OPES working group).  OPES tracing mechanisms allow end users and
  content providers to identify an OPES system and/or intermediaries.
  It is believed that once an OPES system is identified, it would be
  possible to locate relevant information about that system, including
  information relevant to requesters perception of privacy policy or
  reference validity.

11.  Consideration 'Encryption'

  "If OPES is chartered, the OPES working group will also have to
  explicitly decide and document whether the OPES architecture must be
  compatible with the use of end-to-end encryption by one or more ends
  of an OPES-involved session.  If OPES was compatible with end-to-end
  encryption, this would effectively ensure that OPES boxes would be





Babir & Rousskov             Informational                     [Page 12]

RFC 3914          OPES Treatment of IAB Considerations      October 2004


  restricted to ones that are known, trusted, explicitly addressed at
  the IP layer, and authorized (by the provision of decryption keys) by
  at least one of the ends" [RFC3238].

  The above quoted requirement was not explicitly listed as on of the
  IAB considerations, but still needs to be addressed.  The context of
  the quote implies that the phrase "end-to-end encryption" refers to
  encryption along all links of the end-to-end path, with the OPES
  intermediaries as encrypting/decrypting participants or hops (e.g.,
  encryption between the provider and the OPES intermediaries, and
  between the OPES intermediaries and the client).

  Since OPES processors are regular hops on the application protocol
  path, OPES architecture allows for such encryption, provided the
  application protocol being adapted supports it.  Hop-by-hop
  encryption would do little good for the overall application message
  path protection if callout services have to receive unencrypted
  content.  To allow for complete link encryption coverage, OPES
  callout protocol (OCP) supports encryption of OCP connections between
  an OPES processor and a callout server via optional (negotiated)
  transport encryption mechanisms [I-D.ietf-opes-ocp-core].

  For example, TLS encryption [RFC2817] can be used among HTTP hops
  (some of which could be OPES processors) and between each OPES
  processor and a callout server.

12.  Security Considerations

  This document does not define any mechanisms that may be subject to
  security considerations.  This document scope is to address specific
  IAB considerations.  Security of OPES mechanisms are discussed in
  Security Considerations sections of the corresponding OPES framework
  documents.

  For example, OPES tracing mechanisms assist content providers and
  consumers in protecting content integrity and confidentiality by
  requiring OPES intermediaries to disclose their presence.  Security
  of the tracing mechanism is discussed in the Security Considerations
  section of [RFC3897].

13.  Compliance

  This document may be perceived as a proof of OPES compliance with IAB
  implied recommendations.  However, this document does not introduce
  any compliance subjects.  Compliance of OPES implementations is
  defined in other OPES documents discussed above.





Babir & Rousskov             Informational                     [Page 13]

RFC 3914          OPES Treatment of IAB Considerations      October 2004


14.  References

14.1.  Normative References

  [RFC3238]                     Floyd, S. and L. Daigle, "IAB
                                Architectural and Policy Considerations
                                for Open Pluggable Edge Services", RFC
                                3238, January 2002.

  [RFC3752]                     Barbir, A., Burger, E., Chen, R.,
                                McHenry, S., Orman, H. and R. Penno,
                                "Open Pluggable Edge Services (OPES)
                                Use Cases and Deployment Scenarios",
                                RFC 3752, April 2004.

  [RFC3835]                     Barbir, A., Penno, R., Chen, R.,
                                Hofmann, M., and H. Orman, "An
                                Architecture for Open Pluggable Edge
                                Services (OPES)", RFC 3835, August
                                2004.

  [RFC3897]                     Barbir, A., "Open Pluggable Edge
                                Services (OPES) Entities and End Points
                                Communication", RFC 3897, September
                                2004.

14.2.  Informative References

  [RFC2227]                     Mogul, J. and P. Leach, "Simple
                                Hit-Metering and Usage-Limiting for
                                HTTP", RFC 2227, October 1997.

  [RFC2616]                     Fielding, R., Gettys, J., Mogul, J.,
                                Frystyk, H., Masinter, L., Leach, P.
                                and T. Berners-Lee, "Hypertext Transfer
                                Protocol -- HTTP/1.1", RFC 2616, June
                                1999.

  [RFC2817]                     Khare, R. and S. Lawrence, "Upgrading
                                to TLS Within HTTP/1.1", RFC 2817, May
                                2000.

  [I-D.ietf-opes-http]          Rousskov, A. and M. Stecher, "HTTP
                                adaptation with OPES", Work in
                                Progress, October 2003.






Babir & Rousskov             Informational                     [Page 14]

RFC 3914          OPES Treatment of IAB Considerations      October 2004


  [I-D.ietf-opes-ocp-core]      Rousskov, A., "OPES Callout Protocol
                                Core", Work in Progress, November 2003.

Authors' Addresses

  Abbie Barbir
  Nortel Networks
  3500 Carling Avenue
  Nepean, Ontario
  CA

  Phone: +1 613 763 5229
  EMail: [email protected]


  Alex Rousskov
  The Measurement Factory

  EMail: [email protected]
  URI:   http://www.measurement-factory.com/































Babir & Rousskov             Informational                     [Page 15]

RFC 3914          OPES Treatment of IAB Considerations      October 2004


Full Copyright Statement

  Copyright (C) The Internet Society (2004).

  This document is subject to the rights, licenses and restrictions
  contained in BCP 78, and except as set forth therein, the authors
  retain all their rights.

  This document and the information contained herein are provided on an
  "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/S HE
  REPRESENTS OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY AND THE
  INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS OR
  IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF
  THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED
  WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.

Intellectual Property

  The IETF takes no position regarding the validity or scope of any
  Intellectual Property Rights or other rights that might be claimed to
  pertain to the implementation or use of the technology described in
  this document or the extent to which any license under such rights
  might or might not be available; nor does it represent that it has
  made any independent effort to identify any such rights.  Information
  on the IETF's procedures with respect to rights in IETF Documents can
  be found in BCP 78 and BCP 79.

  Copies of IPR disclosures made to the IETF Secretariat and any
  assurances of licenses to be made available, or the result of an
  attempt made to obtain a general license or permission for the use of
  such proprietary rights by implementers or users of this
  specification can be obtained from the IETF on-line IPR repository at
  http://www.ietf.org/ipr.

  The IETF invites any interested party to bring to its attention any
  copyrights, patents or patent applications, or other proprietary
  rights that may cover technology that may be required to implement
  this standard.  Please address the information to the IETF at ietf-
  [email protected].

Acknowledgement

  Funding for the RFC Editor function is currently provided by the
  Internet Society.







Babir & Rousskov             Informational                     [Page 16]