Re: [MPTF] ADR_Min_Control and maxBandwidth

Hi Jan,
To your points:
 #1:  This is interesting as well, but I'm not entirely sure of the use cases.  I'd still keep the maxBandwidth for adaptive streaming separate though.  I might
 #2:  adaptive streaming is a matrix of multiple bitrate levels, and how much temporal data is downloaded from each.  max Buffer size is focused on the temporal aspect of that matrix and doesn't eliminate the need for maxBandwidth.  Also enforcing an actual bandwidth limiter by the UA could be counterproductive.  There is a benefit to having download burst rates spike above the video bitrate playing.

Jason Lewis
Sr. Software Engineer, Mobile and Connected Devices
Disney  Technology   Solutions and Services
925 Fourth Avenue, Suite 1600 | Seattle, WA 98104
*: 206.664.4683  | M:  360.747.7773 | *: jason.lewis@disney.com<mailto:jason.lewis@disney.com> | Skype: jasonlewismedia | LinkedIn: jasonlewismedia<http://www.linkedin.com/in/jasonlewismedia>

From: "Mays, David" <David_Mays@Comcast.com<mailto:David_Mays@Comcast.com>>
Date: Tue, 3 Jan 2012 09:06:57 -0800
To: Jan Lindquist <jan.lindquist@ericsson.com<mailto:jan.lindquist@ericsson.com>>, "public-web-and-tv@w3.org<mailto:public-web-and-tv@w3.org>" <public-web-and-tv@w3.org<mailto:public-web-and-tv@w3.org>>
Subject: Re: [MPTF] ADR_Min_Control and maxBandwidth

Jan,

On your point #1:

Your idea of limiting bandwidth usage by application rather than by streaming is interesting. I think it's difficult without a way to clearly define the boundaries that comprise a complete application.

Defining that as the application developer might not be hard, but conveying that notion to an application consumer in an understandable way seems very difficult to me.

David Mays| sr. software architect | 15.217 | one comcast center | philadelphia, pa. 19103 | 215.286.3395 w | 215.847.9631 m
----------------------------------------------------------------------------------------------------------------------------------------

From: Jan Lindquist <jan.lindquist@ericsson.com<mailto:jan.lindquist@ericsson.com>>
Date: Tue, 3 Jan 2012 08:49:16 +0100
To: "public-web-and-tv@w3.org<mailto:public-web-and-tv@w3.org>" <public-web-and-tv@w3.org<mailto:public-web-and-tv@w3.org>>
Subject: [MPTF] ADR_Min_Control and maxBandwidth
Resent-From: <public-web-and-tv@w3.org<mailto:public-web-and-tv@w3.org>>
Resent-Date: Tue, 3 Jan 2012 07:49:57 +0000

Hello,

This e-mail is simply a question to the TF as to how the maxBandwidth is used. There was a question during the phone conference which adds an interesting dimension to max bandwidth, if the property is strictly only for adaptive streaming or not. I did not remember hearing a clear response and if the MPTF is agreeable I would like to explore the support of maximum bandwidth and how it affects the UA as a whole.

1) max bandwidth for application not only adaptive streaming
If I intereprete the original use case (CT1) end-user to limit their overall bandwidth usage I assume this means not only adapative streaming but the whole webpage or application. Can we make the parameter more generic so it is tied to the application in question and not only adaptive streaming. This would seem like a tremendous value to be able to manage the web traffic from a webpage.

2) max bandwidth or max buffer size
If we are attempting to restrict the speed for adaptive streaming maybe we should be discussing max buffer size. The usage of term bandwidth has a more broader significance

3) Callback on max bandwidth
Another question of conflict of requests between different applications even for the same end-user. If one application can request the max bandwidth it is not sure that the indicated bandwidth is actually available. It would be very useful for a developer on the web page to get information as to the success to set the maximum bandwidth. Can a callback be introduced to indicate what is the actual max bandwidth for that web page (application)? If the UA receives requests from different applications then it can indicate what is actually max bandwidth. Note I am not suggesting to try to prioritize requests or any negotiation just reporting.

An example of the last suggestion. So if the UA only has 20MBTS available. If first application (A) request for 15MBTS the callback returns 15MBTS. When the second application (B) requests 10MBTS callback will return 5MBTS. The application (B) can provide a pop-up indicating rendering is less than optimal performance. This is also useful information for application developer or provider. Once (A) releases the max bandwidth (B) gets a callback with MBTS.

Regards,
JanL