W3C

Media Fragments Working Group Teleconference

08 Oct 2008

Agenda

See also: IRC log

Attendees

Present
Erik, Raphael, Yves, Thierry, Davy, Silvia, Michael
Regrets
Jack
Chair
Erik
Scribe
Michael

Contents


<trackbot> Meeting: Media Fragments Working Group Teleconference

<trackbot> Date: 08 October 2008

<raphael> Scribe: Michael

1. ADMIN

Erik: Accept minutes?

<raphael> +1 for accepting the minutes

Yves will take care of the Wiki issue right after call

<raphael> RESOLVED: minutes from 2008/10/01 telecon accepted

2. ISSUES

-> http://www.w3.org/2008/WebVideo/Fragments/wiki/Issues Issues

Raphael: I've structured the issue on the Wiki, most seem to concern all UC
... temporal seem ok spatial is still open

Silvia: temporally connected fragments?

Raphael: Video UC

Silvia: multiple URIs is hard to deal with, though I see it is needed

Raphael: Thinking back of the interview, edited in two parts

<raphael> Silvia points out the complexity of multi-fragment URIs

<raphael> Given what Silvia suggests, I think this issue is related to the playlist UC

<nessy> xspf = playlist format

<raphael> The question boils to: can a multi-fragment URI be implemented as a playlist? is it the same thing?

Silvia: follow up on F2F

Erik: spatial fragment shape?

Raphael: Might be a matter of complexity, maybe in the second phase of the WG?

<raphael> Silvia suggests to look at the image maps specification

Raphael: will look into it

<scribe> ACTION: Raphael to look into image maps spec re spatial fragment shape [recorded in http://www.w3.org/2008/10/08-mediafrag-minutes.html#action01]

<trackbot> Sorry, couldn't find user - Raphael

Erik: containers and substreams

Silvia: From HTTP POV we're looking at resources
... in the media POV a container
... however re RTSP the format the container
... should focus on HTTP

Raphael: Wouldn't is make sense to list all container formats we want to deal with?

Davy: ack to support container formats

<raphael> Is the Matroska (mkv) format relevant for example?

<Yves> documenting what we investigated is good for other people to find out where we are coming from once we have a solution

<raphael> +1 for documentation

+1

sorry, for whom?

<scribe> ACTION: Davy and Erik to document container formats [recorded in http://www.w3.org/2008/10/08-mediafrag-minutes.html#action02]

<trackbot> Created ACTION-3 - And Erik to document container formats [on Davy Van Deursen - due 2008-10-15].

Erik: In-context / Out-of-context

<raphael> The wiki section is badly phrased, should be rephrased

<Yves> the difference between a fragment and a complete resource (which can be a part of something bigger), depends mostly on the way we identify the fragment

Erik: Representation and Implementation Issues
... URI Fragment / URI Query

<raphael> thanks Silvia, most appreaciated you took this action

<Yves> I would note that ';' is another separator that ca be useful here

<scribe> ACTION: Silvia to flesh out how far can we go with a hash URI regarding the interaction between the UA and the server [recorded in http://www.w3.org/2008/10/08-mediafrag-minutes.html#action03]

<trackbot> Created ACTION-4 - Flesh out how far can we go with a hash URI regarding the interaction between the UA and the server [on Silvia Pfeiffer - due 2008-10-15].

Erik: Fragment Name and Description

Raphael and Silvia agree on non-semantics of the labels

Erik: Status of a Media Fragment

<raphael> Problem with ? is that the query is a new resource, or at least no formal connexion between the fragment and the parent resource

<raphael> Yves, can you put on IRC the link of the draft you're talking about?

<Yves> http://www.mnot.net/drafts/draft-nottingham-http-link-header-01.txt

<nessy> http://www.ietf.org/rfc/rfc3986.txt

<raphael> Yves, would you like to take an action for checking the status of ';' in a URI?

<Yves> sure

<raphael> Silvia: ByteRanges has advantages (cachable, proxies, etc.) and I'm all for the '#' but 8 years ago, I found the limit of the hash URI

<scribe> ACTION: Yves to check the status of ';' in a URI [recorded in http://www.w3.org/2008/10/08-mediafrag-minutes.html#action04]

<trackbot> Created ACTION-5 - Check the status of ';' in a URI [on Yves Lafon - due 2008-10-15].

Erik: Protocols

<raphael> Yves notes that defining a range unit as as easy as defining a new header

<raphael> +1 for learning how RTSP fragments work

Sivia: my in-line action here is well covered

Erik: Server, Caches and Proxies

Silvia: would be nice to find a generic way
... however, seems not too realistic

<raphael> Silvia: would propose to cover HTTP (+ HTTPs, FTP)

3. USE CASES

Erik: going through UCs

Raphael: working on the Map and Multi Resolution UC
... zoomify providing input

4. STATE OF THE ART

-> http://www.w3.org/2008/WebVideo/Fragments/wiki/State_of_the_Art

5. AOB

Erik: put logistics and agenda for second F2F in Gent are online

<raphael> Logisitics page: http://www.w3.org/2008/WebVideo/Fragments/wiki/GhentLogistic

<raphael> IMPORTANT: book your hotel before Nov 8th for getting the discount !!!

<nessy> +q

Silvia: file upload enabled?

Summary of Action Items

[NEW] ACTION: Davy and Erik to document container formats [recorded in http://www.w3.org/2008/10/08-mediafrag-minutes.html#action02]
[NEW] ACTION: Raphael to look into image maps spec re spatial fragment shape [recorded in http://www.w3.org/2008/10/08-mediafrag-minutes.html#action01]
[NEW] ACTION: Silvia to flesh out how far can we go with a hash URI regarding the interaction between the UA and the server [recorded in http://www.w3.org/2008/10/08-mediafrag-minutes.html#action03]
[NEW] ACTION: Yves to check the status of ';' in a URI [recorded in http://www.w3.org/2008/10/08-mediafrag-minutes.html#action04]
 
[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.133 (CVS log)
$Date: 2008/10/08 13:04:53 $