Re: [ppsp] WG item adoption confirmation

Hi all,
      For the transport protocol, which is beyond the current scope of PPSP, as suggested in the charter. However when we talk about the transport protocol used in practice for P2P streaming applications, UDP has been seen the most commonly used protocol now, with the transition from TCP to UDP both for data transport. And many applications(e.g., ppstream, pplive) even change from TCP to use UDP for signaling transport(draft-zhang-ppsp-protocol-comparison-measurement-00). 
       The rational behind this is that firstly, streaming applications *don't*care much of packet loss and secondly, p2p streaming tracker and peer query mechanism ensures there are *enough* active peers to exchange data, so a peer doesn't care *much* if one request is successfully transmitted or not. This is proven in wired network. But when we consider a converged environment, we may need more investigation on whether UDP is *enough* for transport.
      For the encoding issue, since we polled and seems "text" is acceptable by most guys and there are some uncertainty on "binary", I would suggest (individually) to add one section to analyze the strengh and drawback of both encodings in the protocol draft. 
      Regarding the WG item adoption, I would like to see that there is rough consensus on the questions recently raised and discussed in the mailing list before the adoption.

BR
Yunfei

zhangyunfei
2011-04-12

发件人： Martin Stiemerling
发送时间： 2011-04-08 22:25:40
收件人： ppsp@ietf.org
抄送： 
主题： Re: [ppsp] WG item adoption confirmation

[speaking  as  individual  -  not  as  PPSP  co-chair]

Hi  there,

Here  is  my  incomplete  review  of  draft-gu-ppsp-tracker-protocol-03  and  my  opinion  of  whether  it  is  ready  to  become  WG  item:

-  Why  are  there  still  2  encodings  in  the  draft?  Isn't  it  time  to  conclude  on  one  encoding?
-  Section  1:  "the  main  part  is  the  abstract  description  of  the  operations...".  This  means  that  this  is  actually  not  a  draft  about  the  tracker  protocol?
-  Section  1:  "for  both  a  bittorrent  style  offline  and  real-time  streaming  protocol".  Why  is  it  so?  We  are  in  PPSP,  so  we  should  work  on  something  for  streaming,  isn't  it?
-  Why  is  there  the  notion  of  battery  level  in  the  status  messages?
-  Section  9.1.3:  What  is  the  issue  with  fragmentation  in  here?  
-  What  is  the  transport  protocol  where  the  tracker  protocol  should  run  over?

The  proposed  methods  look  reasonable,  but  the  overall  draft  organization  still  suffers  from  what  Section  1  hints  to  that  it  is  solely  an  abstract  description  of  the  operations.  

I'm  **not**  in  favor  of  getting  draft-gu-ppsp-tracker-protocol-03  to  be  a  WG  item,  for  these  reasons:
-  document  quality  is  not  good  enough  in  my  opinion
-  it  is  unclear  what  the  transport  protocol  is.  there  is  a  hint  to  UDP,  which  is  not  a  good  choice  to  be  used  in  this  particular  case
-  there  is  not  yet  a  real  protocol  described  in  the  draft,  but  only  the  skeletons  of  two  protocols  (binary  and  HTTP).  

I  would  suggest  (still  speaking  as  individual)  to  first  make  some  important  decisions,  e.g.,  encoding,  fix  the  document,  etc  and  **afterwards**  make  a  new  call  for  WG  adoption.

Thanks,

   Martin

martin.stiemerling@neclab.eu

NEC  Laboratories  Europe  -  Network  Research  Division
NEC  Europe  Limited  |  Registered  Office:  NEC  House,  1  Victoria  Road,  London  W3  6BL  |  Registered  in  England  2832014  

>  -----Original  Message-----
>  From:  zhangyunfei  [mailto:zhangyunfei@chinamobile.com]
>  Sent:  Tuesday,  March  29,  2011  10:22  AM
>  To:  Cullen  Jennings;  ppsp@ietf.org
>  Cc:  Martin  Stiemerling
>  Subject:  WG  item  adoption  confirmation
>  
>  Hi  all,
>        As  discussed  in  yesterday's  meeting,  we  will  likely  adopt  draft-gu-
>  ppsp-tracker-protocol  as  a  new  WG  item.  Please  post  on  the  mailing  list
>  if  you  have  any  objections  on  this  before  Apr.  10th.
>          If  there  are  no  objections  by  Apr.  10th  2011,  the  draft  above  will
>  be  accepted  as  WG  document  fulfilling  the  “tracker  protocol”
>  deliverable.Thanks.
>  
>  BR
>  Yunfei
>  
>  
>  
>  ________________________________
>  
>  zhangyunfei
>  2011-03-29
>  ________________________________
>  
>  发件人：  Cullen  Jennings
>  发送时间：  2011-03-28  17:27:23
>  收件人：  ppsp@ietf.org
>  抄送：
>  主题：  [ppsp]  Notes  from  PPSP  meeting  IETF80
>  
>  
>  A    few    notes    I    took    from    the    meeting
>  
>  First,    thank    you    to    Christian    Schmidt    for    taking    minutes    and
>  Martin    Stiemerling    for    jabber    scribing.
>  
>  We    need    to    prioritize    the    use    cases    and    decide    what    we    will
>  work    on    first.
>  
>  draft-ietf-ppsp-problem-statement    is    getting    close    to    WGLC
>  
>  We    will    likely    adopt    draft-gu-ppsp-tracker-protocol    as    a    WG
>  item
>  
>  Were    about    60    people    in    room
>  
>  
>  
>  
>  _______________________________________________
>  ppsp    mailing    list
>  ppsp@ietf.org
>  https://www.ietf.org/mailman/listinfo/ppsp
_______________________________________________
ppsp  mailing  list
ppsp@ietf.org
https://www.ietf.org/mailman/listinfo/ppsp