CCAMP Meeting Minutes
74th IETF Meeting
Chairs: Deborah Brungard & Lou Berger (replaces Adrian Farrel)
Scribe: Daniel King

-------------------------------------------------
First Session
Thursday: 9.00am - 11.30am (Morning Session I)
-------------------------------------------------

Agenda: http://www3.ietf.org/proceedings/09mar/agenda/ccamp.htm

 0. Administrivia I

    Slides: http://www3.ietf.org/proceedings/09mar/slides/ccamp-0.ppt

    - Adrian Farrel will shortly step down as WG chair. Lou Berger will
      become the next CCAMP WG chair.  The chairs thanked Adrian for
      his stewardship and service.
    - Adrian Farrel will become an AD, and is expected to be the AD for
      CCAMP.
    - Adrian was unexpectedly unable to attend the meeting.
    - New CCAMP WG Secretary Daniel King.
    - No requested agenda changes.


 1. WG status, RFCs, drafts, milestones, charter

    Slides: http://www3.ietf.org/proceedings/09mar/slides/ccamp-1.ppt

    RFCs
    - Several New RFC's.

    Drafts
    - One document in the editor's queue [draft-ietf-ccamp-pc-and-sc-reqs].
    - Two documents being processed by the IESG
      [draft-ietf-ccamp-gmpls-ason-routing-ospf &
      draft-ietf-ccamp-path-key-ero]
    - Several WG drafts are being prepared for WG last call. Expect to see
      these being discussed on the mailing list over the next few weeks.
    - Deborah Brungard requested if the authors intended to request WG last
      call for Ethernet Provider Requirements
      [draft-ietf-ccamp-ethernet-gmpls-provider-reqs]. No authors were
      available to comment.
    - Deborah Brungard requested the status of VCAT Support document
      [draft-ietf-ccamp-gmpls-vcat-lcas].

      Greg Bernstein: The authors believe the document is almost
      completed. The document does use a code point from the MLN
      [draft-ietf-ccamp-gmpls-mln-extensions] draft, so we are gated by this,
      to obtain our code point. We are queued behind the MLN extensions.

    - Deborah Brungard requested the status of OAM Requirements
      [draft-ietf-ccamp-gmpls-oam-requirements]. This draft has expired and
      interest has been minimal.

      Loa Andersson: I do not care if this draft has expired. I would request
      the WG chairs to send a pointer to the editors of the MPLS-TP OAM
      requirements document to make sure that no requirements exist in this
      [draft-ietf-ccamp-gmpls-oam-requirements] document that should be
      captured in the MPLS-TP OAM document.

      Deborah Brungard: Ok. Would anyone object to abandoning this document?
      No hands. [No one objected].

    - The BNF specification [draft-farrel-rtg-common-bnf] that Adrian Farrel
      authored has been submitted for WG last call. This document will be
      used as a normative reference for all CCAMP work.

      Ross Callon: Just a quick comment regarding the common BNF document
      [draft-farrel-rtg-common-bnf]. Adrian did manage to get an update
      submitted just before he became sick and the document was approved
      yesterday.

    Milestones

    - For August 2008 CCAMP was assigned the milestone to create the first
      draft of Control Plane Framework for MPLS-TP. A draft
      [draft-abfb-mpls-tp-control-plane-framework] exists and Lou Berger
      provided an update.

      Lou Berger: A draft exists and Luyuan [Luyuan Fang] discussed this
      draft in MPLS [MPLS WG] yesterday. Our intent is to provide an update
      and include some individual submissions and rewrite some incomplete
      sections. The document is not quite ready for WG status.

    - The WG had a milestone to submit OSPF-TE/GMPLS MIB documents. These
      documents will be available soon.
    - The WG had a milestone [October 2008] for an initial version of a
      draft defining RSVP-TE extensions for MPLS-TP. No document currently
      exists.
    - The WG has completed the ASON Routing solutions draft and has been
      presented for IESG review.
    - The milestone for  protocol extensions for GMPLS source-controlled and
      explicitly-routed Ethernet networks
      [draft-ietf-ccamp-gmpls-oam-requirements] will be abandoned.
    - The WG chairs will be updating the milestones and planning for new
      mile stones.

    Loa Andersson: I have been thinking about the control plane framework
    for MPLS-TP. I am not entirely sure which WG it should go to. The MPLS
    WG and CCAMP WG should agree on which WG picks up the work. We need to
    understand how to poll the work as it spans multiple WGs.

    Lou Berger: I figured that we would continue the past trend of polling
    multiple WGs, but I agree we do need to decide where to focus the
    document discussions.

    Igor Bryskin: I think the decision on where the work should be done
    depends on convincing people to throw pseudowires out of the framework
    and concentrate on MPLS-TP layer, or consider both MPLS-TP and
    pseudowire layer. I think it is crucial if we concentrate on just
    MPLS-TP layer. For companies its important because we want to support
    MPLS-TP and we have GMPLS, but completely different to support
    pseudowires which only router company can do.

    ???: First the MPLS-TP control plane framework should define how we
    control an MPLS-TP network. Then decide what the control plan should
    control. This may include PWE or not, also what else.

    Deborah Brungard the CCAMP co-chair: This discussion happened in the
    MPLS [WG] session. If you're interested in this discussion please be on
    the MPLS mailing list.

2. ITU-T and OIF progress report

    Slides: http://www.ietf.org/proceedings/09mar/slides/ccamp-2.ppt
    ITU Update
    Presented by Lyndon Ong

    Greg Bernstein: Is the updated version of the work plan available.

    Lyndon Ong: I think it's available from Adrian's website [CCAMP
    supplemental pages].

    Deborah Brungard: As Lyndon noted the Q11 has begun their work on
    G709 ODU0 and OTU4. If anyone is interested in starting this work in
    parallel to their work, that would be good.

    OIF Update
    Presented by Lyndon Ong

    - OIF documentation is also available on Adrian's website [CCAMP
      supplemental pages].
    - OIF would appreciate CCAMP comments on restoration
      specifications. This should happen after Q12 consider the architecture
      and then request feedback from CCAMP.

    Julien Meuric: What are the plans for Adrian's website [CCAMP 
    supplemental pages]?

    Lou Berger: The CCAMP supplemental pages will be moved and
    maintained by Daniel King.  The new URL will announced once the new
    site is available.

    Deborah Brungard: Regarding VCAT multi-session which is a connection
    over multiple-domains is not really a CCAMP issue. You may want to
    go back to the ITU to understand what is necessary to run over
    multi-domains.

    Lyndon Ong: We understand that.


3. ITU Q6/15 WSON Meeting Report

    Slides: http://www3.ietf.org/proceedings/09mar/slides/ccamp-3.ppt

    - Results of the discussion are noted in the meeting report:
      http://www.olddog.co.uk/download/WD6-38.doc

      Donald O'Connor: My understanding is that G.680 does not fully
      support scenario 3 in its present form. It needs additional
      enhancements. My concern is that the work remains [un]synchronized
      between the two groups. The work here [CCAMP] should not go
      forward into the IESG, until the G.680 is appropriately
      updated. We want to support ITU standard networks, not some
      non-standard OTN. Because we get into the same issues that we got
      into with MPLS-TP. Where we have some other alien network and how
      does it interconnect with standard optical networks. I'll assume
      we will continue liaisons and we won't be ships in the
      night. G.680 will need to be enhanced to support these scenarios.

      Deborah Brungard: Yes. Q6 did say they were planning to do
      that. Of course we will never go past what they are doing. This
      work will be aligned.

      Eve Varma: It was a really excellent meeting. A few eye openers,
      including, Case 3, if one excludes non-linearity's, what is the
      scale of the network you can deal with? Is this is something
      that's useful? Another aspect, if you could measure or use test
      equipment to make sure you were operating without non-linearity's
      was there economics that would counter that, was this something
      that would counter-balance. The last item was a scoping, it was
      understood that this was an intra-domain application and there was
      not a way of computing an end-to-end path.

      Deborah Brungard the CCAMP co-chair: Whatever ITU decides, we will
      do.

4. RWA

 a. Framework for RWA in WSON
    Presented by Greg Bernstein from Grotto Networking
    Slides: http://www3.ietf.org/proceedings/09mar/slides/ccamp-4.ppt
    Background reading: draft-ietf-ccamp-rwa-wson-framework-02.txt

    Malcolm Betts: Q12 is meeting on June 1st to June5th, and amongst
    other things they will discuss modeling terminology.

    Greg Bernstein: We can check on the CCAMP list and see if
    individuals can consider this as well.

    Lou Berger: Do you believe that an OEO switch can implement a WSON
    network, is it in scope of out of scope?

    Greg Bernstein: When you hit an OEO, it does look like the most
    basic wavelength converter.

    Lou Berger: Exactly. It seems that we have the converter pool, which
    looks a lot like an OEO. A full OEO switch could be a special case.

    Greg Bernstein: This takes us into the modeling issue I was
    discussing with Malcolm. I would like to model where I can say this
    wavelength converter can deal with this signal. We do not have this
    definition.

    Lou Berger: It seems that we should include this as a special case.

    Jonathan Sadler(???): The architecture probably does need to be 
    specified before the solution. The architecture does need to 
    include OEO, but there are cases when OOO and OEO are totally     
    different.

    Lou Berger: It gets really hard to tell between and OEO and regen.

    Jonathan Sadler(???)?: Certainly if you're talking about an 
    electrical regen.

    Donald O'Connor: I thought this project always included that
    [OEO]. If this is limited exclusively to transparent optical
    networks, because I raised this. I was told that we want to apply
    this to current ROADM networks.

    Young Lee: The Framework document mentions OEO in three places and
    is part of the scope.

    Lou Berger: You need to make it very clear in the document.

    Malcolm Betts: We need to define what OEO is in scope and how it is
    defined. Is it, can convert the wavelength but the signal format
    stays exactly the same. Or do I change the modulation scheme. We
    just need to make sure we have the attributes and model to describe
    them.

    Deborah Brungard: Yes. We discussed this with Q6. They mentioned
    they have object classes.

    Greg Bernstein: Object tributary classes would be very useful. We
    will throw this out on the list to kick around.

    Malcolm Betts: We should also allow the future possibility of all
    optical wavelength conversion in the encoding.

    Deborah Brungard: Yes. We had discussed this with Q6 agreed to work
    on that model and discuss with Q12.

 b. RWA Information Model for WSON
    Presented by Young Lee
    Slides: http://www3.ietf.org/proceedings/09mar/slides/ccamp-5.ppt
    Background reading: draft-ietf-ccamp-rwa-info-02.txt

    Lou Berger: In OSPF, there was a question [from Jonathan Sadler] on
    how you could reuse the semantics of the mode. Have you considered
    how the information model might be generally applicable?

    Young Lee: Such as?

    Jonathan Sadler: My question was about blocking
    characteristics. Even for traditional SONET ADMs. The point was this
    extension could be usable for technologies other than DWDM. To that
    end maybe we want to include an additional field that specifies
    which layered network its associated with.

    Young Lee: Currently we had no plan to support other
    technologies. Maybe we need to consider generic requirements in
    another document, as this document is for WSON.

    Lou Berger: Given that we are *C*CAMP and we are working on *G*MPLS. I
    think we want to examine if this is the first case, rather than the
    only case we want to support. Please take a look at the generic
    applicability of this work. This is what we were asked to do with
    the Ethernet documents, which started out as 1 document and ended up
    with 3, and there is already discussion on using the generic
    extensions for other technologies.

    Greg Bernstein: The chairs had previously asked us to break up the
    document into multiple documents. We can break off generic pieces,
    at the info model its WSON, any other documents would require help
    us would have to help us as we focused on WSON and we need
    assistance for say TDM.

    Igor Bryskin: If you use a model that has a shared a traffic
    engineering database I do not see why you need to add this.

    Jonathan Sadler: The reason I suggested we add a field that
    discusses technology is that you have link identifiers. You are
    referencing numbered IP addresses. You have different layers on
    different interfaces.

    Igor Bryskin: GMPLS has a way to say which link identifiers belong
    to which layer.

    Lou Berger: That discussion is really good. Where you make the
    change, I will leave it the authors to propose where it should be
    split out.

    Greg Bernstein: We will put some ideas out on the list.

 c. RWA Encoding for WSON
    Presented by Young Lee
    Slides: http://www3.ietf.org/proceedings/09mar/slides/ccamp-6.ppt
    Background reading: draft-ietf-ccamp-rwa-wson-encode-01.txt

    - No comments

 d. GMPLS G.694 Labels for WSON
    Presented by Diego Caviglia
    Slides: http://www3.ietf.org/proceedings/09mar/slides/ccamp-7.ppt
    Background reading: draft-ietf-ccamp-gmpls-g-694-lambda-labels-03.txt

    Deborah Brungard: Is this ready?

    Lou Berger: Is it likely that there will be any format changes from
    the ITU?

    Diego Caviglia: I supposed this format will be stable.

    Donald O'Connor: My understanding is they are still working on it
    [G.697].

    Lou Berger: If we are tracking what they are doing and there may be
    changes then we should not issue the document. If its stabilized
    then we can move forward. Maybe it's worth doing a last call from
    our side but not progressing beyond that until we have a liaison to
    see what they are doing.

    Greg Bernstein: We were working back and forth with our format and
    their format. They are going to put our encoding in their format.

    Deborah Brungard: That's why I was a little surprised to see that
    you [Diego Caviglia] requested working group last call. I think we
    need to wait.

    Julien Meuric: Another cycle with the ITU is fine. I do not think
    it's a huge risk to go to last call.

    Deborah Brungard: There is no real urgency. We will liaison this to
    Q6. Maybe we can last call in Stockholm, but if not definitely by
    the end of the year.

 e. OSPF in Support of RWA in WSON
    Presented by Greg Bernstein on behalf of Fatai Zhang
    Slides: http://www3.ietf.org/proceedings/09mar/slides/ccamp-8.ppt
    Background reading: draft-zhang-ccamp-rwa-wson-routing-ospf-00.txt

    Lyndon Ong: I guess the node info reflects constraints getting in
    and out of the node. What if the constraints do not apply or there
    are no constraints. What would advertise?

    Greg Bernstein: That's a good thing to put into the encoding
    details. The absence of it, we would probably advertise default
    connectivity. It's a good thing to document in the encoding or info
    document. Lack of any specific information [would default to
    complete connectivity?].

    Lou Berger: I look forward to seeing generic items being identified.

    Julien Meuric: I believe that generalizing stuff makes sense. Is it
    too early to document these items. Should we not document this in
    later OSPF or ISIS documents and keep this generic?

    Lou Berger: I think we will give the authors a chance to make a
    proposal and not dictate an answer. In terms of specifics, sure, but
    I'll defer to authors.

    Deborah Brungard: I think we wait before we make this a WG
    document. Let's request what constraints would be required for other
    technologies, TDM, etc. Feed them to Greg.

 f. Link Bundle in Wavelength Switched Optical Networks
    Presented by David [surname missing] on behalf of Xihua Fu
    Slides: http://www3.ietf.org/proceedings/09mar/slides/ccamp-9.ppt
    Background Reading: draft-xihua-ccamp-wson-link-bundle-01.txt

    Igor Bryskin: This is a complicated problem. What do you do if you
    have two components and a red lambda on both of them, or one
    amplifier on one and not the other?

    David ???: You cannot do bundle in those case. We need to consider
    the balance.

    Igor Bryskin: We only advertise for path computation. No point in
    advertising something unless it's not needed for path computation.

    Young Lee: What are the requirements for bundling in WSON networks?
    Why do we need to do this?  David ???: You cannot do bundle in those
    case. We need to consider the balance, if you have a fixed network.

    Mustapha ???: I agree with Lee, we only have 40-80 channels. Best to
    try to have diverse path, better than putting all channels on one
    path

    David ???: Making view simple is always good.

    Mustapha ???: Yes, but why put all your channels on one path. You
    need to have a survivable network.

    Malcolm Betts: I thought this draft was about providing information
    on links, not how I use that information to compute paths. Second
    question, would it be possible to combine two drafts to provide
    information and links and nodes on those links. I am not sure if
    that is possible. It's worth considering.

    Lou Berger: If the author wants to continue the document we need to
    discuss the scaling benefits.

    Loa Andersson: I have not read this draft. But there was a draft
    that came up in the routing WG and MPLS WG (CTG) that discussed
    bundling.

    Lou Berger: We do not understand the requirements for CTG yet, best
    to wait for those.

5. Impairments

 a. Framework for WSON Impairments
    Presented by Greg Bernstein
    Slides: http://www3.ietf.org/proceedings/09mar/slides/ccamp-10.ppt
    Background reading: draft-bernstein-ccamp-wson-impairments-03.txt

    Lou Berger: Will this document be revised and modified while we go
    along.

    Greg Bernstein: Yes.

    Donald O'Connor: The companies could submit these drafts as working
    documents to ITU. Or do a liaison and attach the drafts for comment.

    Malcolm Betts: Greg this was not captured in the report. Is there a
    usual size of networks around which we do not have to worry about
    impairments? Where do you need to start worrying about
    non-linearity's? The second thing is where we have networks of
    non-linear impairments. Let me quote from the report "there was
    agreement between members of the IETF CCAMP working group and Q6,
    that current state of the technology does not provide the model to
    calculate viable paths in this scenario. A technical breakthrough
    would be required to achieve this". This puts things in context. Is
    there some useful sized networks where we only have to consider
    linear impairments?

    Greg Bernstein: We would really like input on this, including
    definitions, both informal and formal.

    Lou Berger: How many people have read this document? [>6 hands]. How
    many people think this is a reasonable starting point? [Similar
    number of hands] How many people do not? [no hands] Everyone who
    read the document agrees it's a good starting point. We will take
    this to the list.

    Donald O'Connor: I do not think it should be a WG until document
    until the results of Q6 meeting.

    Lou Berger: Let's look at next version and you send us a mail when
    you think it's ready for WG status. One minor comment. It would be
    great if you change the boiler plate so it's WSON specific or make
    it truly a "Framework for Networks with Impairments".

 b. Information Model for WSON Impairments
    Presented by Greg Bernstein
    Slides: http://www3.ietf.org/proceedings/09mar/slides/ccamp-11.ppt
    Background reading: draft-bernstein-wson-impairment-info-01.txt

    Donald O'Connor: I have two questions. I think there should be a
    parameter where it shows how it was acquired and why. PMD example,
    directly measured or estimated.

    Greg Bernstein: Q6 is looking to define a list like that.

    Deborah Brungard: Perhaps we can discuss this more on the mailing
    list. Why do we need these?

    Donald O'Connor: The PCE needs to apply some design rules. There
    will be some paths that are acceptable; the path computation is a
    function of the info, algorithm and design rules. It is
    important. If a PCE it needs some level of confidence if this is
    measured or estimated. If these are advertised then there should be
    some qualification with it. Is it measure or is it an
    estimation. This needs to be vocalized within Q6. My second question
    is the additional transmitter and receiver characteristics. Can you
    give us some examples?

    Greg Bernstein: G.680 does not have them [transmitter info]. If you
    want to find out about a transmitter, you have to look at an SDH
    document or a G.709 document, or G959.1 document. Are you going to
    point to all those documents? Some of the ITU parameters come in
    sets. The other point is the accuracy. That would be a good
    question. Other cases they might say you have to compute both cases,
    including the maximum.

    Eve Varma: I sympathize that the path computation is only as good as
    the data that fed it. During Q6 discussions not all the parameters
    are even measurable. Some are from datasheets, some might have
    issues with accuracy and technology. There is a wider range of
    variation. Some are static, was there value in distribution the
    info. Not sure how practical it is.

    Richard Graveman: A comment on security of the information. It
    mentions improper disclosure; I'd also include improper modification
    as well.

    Greg Bernstein: Yes.

    Igor Bryskin: When we compute paths we compute on the network and it
    was thoroughly engineered. Until now there is not a need to provide
    this information. I do understand there could be multiple cases
    where some paths can be met and so cannot be met. I would then ask
    the optical engineers who plan this network.

    Greg Bernstein: We are here to transport these things instead of
    manually adding them. It may be spec sheet info.

    Malcolm Betts: Following up from Eve. Sometimes measurements require
    long term measurement, or you're looking at spec sheet. We need to
    characterize these things.

    Greg Bernstein: Yes, I agree.

    Donald O'Connor: Getting back to the same topic. I was getting back
    to the fiber link. It could be dynamic, it could be static. I think
    there should be a parameter where it shows how and why. PMD for
    example, if it's advertised is it estimated from datasheet or
    directly measured.

    Lou Berger: If you're identifying additional parameters that you
    would like to see carried. It would be good if your co-authors would
    bring this into Q6.

    Greg Bernstein: Yes.

 c. WSON Impairment Encoding
    Presented by Greg Bernstein
    Slides: http://www3.ietf.org/proceedings/09mar/slides/ccamp-12.ppt
    Background reading: draft-bernstein-wson-impairment-encode-00.txt

    - No comments

 d. Requirements of Impairment Compensation Control in WSON
    Presented by Shoichiro Seno
    Slides: http://www3.ietf.org/proceedings/09mar/slides/ccamp-13.ppt
    Background reading: draft-seno-ccamp-wson-impairment-compensate-cntl-00.txt

    Donald O'Connor: I think you should take this to the ITU first. ITU
    needs to make standards for this type of DCM (Q6). The appropriate
    path is for ITU to standardize this form, then use this group to
    develop appropriate control plane standards. Currently most of this
    is outside the WSON scope.

    Deborah Brungard: I was going to note that to [Q6]. Also Q9, we need
    to know what equipment we are controlling.

    Donald O'Connor: I do not think Q9 gets down to this area. I'd start
    with Q6.

    Greg Bernstein: There are two things I liked about this draft. It
    broke up the issue of what are you measuring and for what purposes
    and identified the problems and protocols that could be used. We did
    hear from Q6, with regards to measurement.

    Deborah Brungard: Greg, could you put some of this work into the
    Framework document.

    Greg Bernstein: It depends if we have enough interest in
    measurement. It may be good. I did not have enough info for the
    measurement. Other folks have brought it up. Right now we are
    focusing on impairment aware. I am scoping the framework document.

    Donald O'Connor: On the measurement aspects, again, I'd allude to
    measurement details beyond what the parameters are, versus estimated
    or measured.

    Julien Meuric: I agree that it is a little early to discuss this but
    I believe it may be interesting for the future.

    Eve Varma: I'd echo Don's thoughts I would not support any draft
    that supports and specifies measurements and parameters, this is the
    domain of Q6.

    Deborah Brungard: You, [Shoichiro Seno] will need to work with Greg
    and potential have a different framework document. We have to be
    very careful that we are not doing the measurements.

    Greg Bernstein: It Q6 mentioned that would get a list of
    measurements out to us [on there to-do list].

    Shoichiro Seno: I will work with Greg to further clarify what is in
    the framework.


-------------------------------------------------
Second Session
Thursday: 3.10pm - 4.10pm (Afternoon Session II)
-------------------------------------------------

6. Administrivia II
   Slides: http://www.ietf.org/proceedings/09mar/slides/ccamp-14.ppt
    - No Agenda changes
    - Adrian Farrel was ill and unable to attend.

7. OAM Configuration Framework for GMPLS
    Presented by Attila Takacs
    Slides: http://www.ietf.org/proceedings/09mar/slides/ccamp-15.ppt
    Background reading: draft-ietf-ccamp-oam-configuration-fwk-01.txt

    Lou Berger: Is the monitoring status flag redundant?

    Attila Takacs: No, not really as the flag is used to indicate when
    monitoring traffic should be inserted into the LSP.

8. GMPLS Recovery Extension for Reversion
    Presented by Attila Takacs
    Slides: http://www.ietf.org/proceedings/09mar/slides/ccamp-16.ppt
    Background reading: draft-takacs-ccamp-revertive-ps-03.txt

    Deborah Brungard: How many have read the draft. [>4 hands] Anyone
    opposed? [no hands raised] Ok, let's take it to the list.

9. GMPLS Control of Ethernet PBB-TE
    Presented by Don Fedyk
    Slides: http://www.ietf.org/proceedings/09mar/slides/ccamp-17.ppt
    Background reading: draft-ietf-ccamp-gmpls-ethernet-pbb-te-02.txt

    Igor Bryskin: You added I-SID. How many can you signal?

    Don Fedyk: One or more.

    Igor Bryskin: It's something I wanted to clarify.

    Don Fedyk: Service Instance Identifier that belongs to the client.

    Deborah Brungard: How many have read the draft. [>6 hands raised]
    Anyone opposed? [no hands raised] Ok, let's take it to the list.

10. GMPLS Switch Programming
    To be presented by Adrian Farrel for Kohei Shiomoto
    Slides: http://www.ietf.org/proceedings/09mar/slides/ccamp-18.ppt
    Background reading: draft-shiomoto-ccamp-switch-programming-00.txt

    - Adrian Farrel was ill and unable to attend.

    Chairs: This is the initial version of the draft.  It is targeted as
    an information document and does not modify any exitsing
    procedures.  The intent of the document seems to be to clarify when
    data can should be expected to be received.  It seems like a
    reasonable start at this.

11. RSVP-TE Extensions to GMPLS Calls
    To be presented by Adrian Farrel for Fatai Zhang
    Slides: http://www.ietf.org/proceedings/09mar/slides/ccamp-19.ppt
    Background reading: draft-zhang-ccamp-gmpls-call-extensions-00.txt

    - Presented by Young Lee as Adrian was unavailable and Fatai Zhang
      was unable to make the trip.

    Lou Berger: One comment to bring back to the authors. Consider
    simplifying the document by using the existing ERO/RRO
    objects. Using existing semantics might make it simpler.

12. Problems Observed With RSVP Recovery Signaling
    Presented by Nic Neate
    Slides: http://www.ietf.org/proceedings/09mar/slides/ccamp-20.ppt
    Background reading: draft-rhodes-rsvp-recovery-signaling-01.txt

    Dimitri Papadimitriou: There have been certain exchanges with Adrian
    and I regarding how the ID is used and it's open to
    interpretation. It's a segment you are protecting.

    Lou Berger: I believe he [Nic Neate] is rejecting Adrian's comments.

    Nicholas Neate: The text does not reflect this.

    Dimitri Papadimitriou: We have tried to explain how this works.  It
    is a question for the group on how we respond to individuals who
    repeatedly ask the same question and don't accept the answer.

    Lou Berger: [Removes WG Co-chair hat] the original comment was it
    was open to misinterpretation and we can see that at least one
    reader is not clear.  I propose that we (the original recovery RFC
    authors) put forward an informational document on how this works.
    If this still doesn't answer the questions, then the group can then
    decide how to proceed.

    Deborah Brungard: This is a good recommendation and is appropriate
    given that you still have your Technical Adviser hat until you
    formally become WG Co-chair (after this meeting).

    Nic Neate: My point is I want to make sure it's clear and we all
    know how this works.