[ppsp] 答复: 答复: Fw: Please review the updated Tracker protocol

Hi Cruz,

Thank you very much for the detail comments.

I will try to add some explanation in the draft. The remove of CONNECT,
DISCONNECT is in order to reduce communication rounds. At the very
beginning, we planed to define a protocol with clear definition of each
step. But, at last meeting, after discussing with many people, we think a
perfect protocol may not be necessary, and we fail to find a reason a peer
connect to tracker without FIND or JOIN. So we remove CONNECT, DISCONNECT.

I will revise the draft to make it be consistent. E.g. delete JOIN_CHUNK.

As for file sharing, I will remove it.

According to interim meeting at last night, STAT_QUERY should be removed for
now, and STAT_REPORT need more discussion. We need to discuss on the mail
list whether we would like Tracker to deal with ‘performance improvement’
kind of issue by STAT_REPORT.

Thank you again for your comments.

  _____  

Best Regards
Gu Yingjie

发件人: Rui Cruz [mailto:rui.cruz@ieee-pt.org] 代表 Rui Cruz
发送时间: 2011年9月28日 乐乐20:02
收件人: ppsp
抄送: Rui Cruz; Yingjie Gu(yingjie)
主题: Re: [ppsp] 答复: Fw: Please review the updated Tracker protocol

Hi Yingije,

In this new draft some messages have disappeared (CONNECT, DISCONNECT)
without explanation.

But for JOIN and STAT messages the descriptions in section 4.3 do not match
with their syntax in section 5, where it is stated in Table 1 that the valid
strings for requests are FIND, JOIN, LEAVE, KEEPALIVE, STAT_QUERY,
STAT_REPORT, and then in section 5.1.5.2 a JOIN_CHUNK message is described. 

Another concern is related to the statement 

"The PPSP tracker protocol can serve live media and VoD, as well as file
sharing applications."

The purpose of PPSP is for streaming in P2P. Including also "file sharing"
does not seem to be adequate.

I also do not see the purpose of the change in the description from the
messages formats to the XML schemas.

On the other hand, reversing the process of STAT_REPORT, as only a response
to a STAT_QUERY from tracker may be problematic.

Regards,

Rui Cruz

On 28/09/2011, at 12:34, Yingjie Gu(yingjie) wrote:

Thanks a lot, Jianyin,

Jinwei and I will review the encoding issues you mentioned after the interim
meeting.

  _____  

Best Regards
Gu Yingjie

发件人: 张剑寅 [mailto:zhangjianyin@chinamobile.com] 
发送时间: 2011年9月28日 乐乐15:08
收件人: Yingjie Gu(yingjie)
抄送: ppsp
主题: Re: Fw: [ppsp] Please review the updated Tracker protocol

Hi, Yingjie,

I have some comments on the XML expression as follows:

section 5.1.1

It's ok that one example is presented to show the format of the shared
mesage body. But it seems as the content emeded in the message body, having
no relation with the "Common Message XML Header", so my advice is to modify
the figure title.

section 5.1.2.1

The example for the peerlist representation is not the formal XML format. It
can be expressed as follows:

      <Peers>

          <Peer1>

              <PeerID> 48482345122344 </PeerID>

              <PeerAdd> 193.2.2.4 </PeerAdd>

          </Peer1>

          ... ...

      </Peers>

section 5.1.4.2

The example shown for the FIND response message is just XML schema, rather
than message instance. So maybe the figure title should be changed to "XML
schema expression for FIND Response message". The same problem exists in the
following figures from Figure 6 to Figure 11.

Welcome for more discussion.

  _____  

张剑寅(Jianyin Zhang) 

中国移动通信研究院/Research Institute of China Mobile

Address:宣武区宣武门西大街28号大成广场2门11层1107室（100053）

E-mail:  <mailto:zhangjianyin@chinamobile.com;marcozjy@gmail.com>
zhangjianyin@chinamobile.com;marcozjy@gmail.com

Mobile:+86-139-1002-2506

Fax：+86-10-63601087

2011-09-28

  _____  

发件人： Yingjie Gu(yingjie)

发送时间： 2011-09-27 09:39:58

收件人： ppsp@ietf.org

抄送：

主题： [ppsp] Please review the updated Tracker protocol

Hi all,

We have uploaded a new version of Tracker protocol. You can find it under
PPSP documents webpage, or
http://www.ietf.org/id/draft-gu-ppsp-tracker-protocol-05.txt

In this version, we make many changes according to the meeting minutes.

First of all, we optimize the message definiti on, make optional. We provide
more flexibility to implementation, which can help to decrease rounds of
message communication and increase efficiency. An example flow is provided,
together with example encoding.

Secondly, we update syntax. At last meeting, Wes raised that it was not
valid http-xml based encoding. We make large amount of revision to the
encoding to make it look better. Hope you can help us to improve it.

There will be a virtual interim meeting at this Wednesday. Please try to
review it before the meeting if convenient.

Thanks a lot.

  _____  

Best Regards
Gu Yingjie

_______________________________________________
ppsp mailing list
ppsp@ietf.org
https://www.ietf.org/mailman/listinfo/ppsp