See also: IRC log
<trackbot> Date: 09 September 2009
Regret+ Jack
<scribe> Scribe: raphael
<scribe> scribenick: raphael
Informal discussion about the syntax of the Range unit syntax
Raphael: can we have multiple values in the Acept-Ranges ?
Yves: yes, comma separated
Michael: I would also suggest to
use the cue-values, like the accept header
... for example, I accept npt with 0.8 and smpte-30 with
0.2
Yves: makes less sense [I didn't get why]
<Yves> <<
<Yves> The response-header "Accept-Ranges" field allows the server to
<Yves> indicate its acceptance of range requests for a resource:
<Yves> >>
Michael: but then, what does it mean, they are equal ?
Yves: yes, the server decides
<Yves> +1
PROPOSED to accept the minutes of the 02 September 2009 telecon
<mhausenblas> +1
http://www.w3.org/2009/09/02-mediafrag-minutes.html
+1
Minutes accepted
Virtual F2F meeting: http://www.w3.org/2008/WebVideo/Fragments/wiki/FourthF2FAgenda
http://www.doodle.com/xbw9stfdmz3pr4nw
<nessy> +1
Unfortunately: for the meeting, Jack will not make it and Yves is at risk
Jack will be on irc
ALL: please, complete the agenda with background reading material to prepare the meeting
ACTION-95?
<trackbot> ACTION-95 -- Michael Hausenblas to review ALL UC with a mobile hat on and check whether these sufficiently cover the mobile usage -- due 2009-09-02 -- OPEN
<trackbot> http://www.w3.org/2008/WebVideo/Fragments/tracker/actions/95
-- continue
ACTION-101?
<trackbot> ACTION-101 -- Yves Lafon to write a UC describing the use of aspect ratio feature and thus motivating its usage -- due 2009-09-02 -- OPEN
<trackbot> http://www.w3.org/2008/WebVideo/Fragments/tracker/actions/101
close ACTION-101
<trackbot> ACTION-101 Write a UC describing the use of aspect ratio feature and thus motivating its usage closed
We received 3 reviews from MAWG
http://lists.w3.org/Archives/Public/public-media-annotation/2009Sep/0008.html
http://lists.w3.org/Archives/Public/public-media-annotation/2009Sep/0012.html
http://lists.w3.org/Archives/Public/public-media-annotation/2009Sep/0013.html
Werner to introduce other units such as smpte-50 and smpte-60
Nobody against ?
[silence]
<Yves> ok with this, (but we should avoid explosion of units)
<scribe> ACTION: raphael to address all comments and write a reply to MAWG [recorded in http://www.w3.org/2009/09/09-mediafrag-minutes.html#action01]
<trackbot> Sorry, couldn't find user - raphael
trackbot, status
<scribe> ACTION: Raphaël to address all comments and write a reply to MAWG [recorded in http://www.w3.org/2009/09/09-mediafrag-minutes.html#action02]
<trackbot> Could not create new action - please contact sysreq with the details of what happened.
Michael: should we have also a paragraph, directed to the MPEG-21 community why we are doing that?
<scribe> ACTION: Raphaël to draft a paragraph explaining upfront in the document why we are doing that ... justification for the MPEG community [recorded in http://www.w3.org/2009/09/09-mediafrag-minutes.html#action03]
<trackbot> Could not create new action - please contact sysreq with the details of what happened.
<scribe> ACTION: Troncy to draft a paragraph explaining upfront in the document why we are doing that ... justification for the MPEG community [recorded in http://www.w3.org/2009/09/09-mediafrag-minutes.html#action04]
<trackbot> Could not create new action - please contact sysreq with the details of what happened.
Good we have finally one review of our document from a WG
Yves: I will contact TAG once
more
... and tell them our current design choice regarding the use
of ? when transcoding is necessary
ACTION-103?
<trackbot> ACTION-103 -- Silvia Pfeiffer to write a paragraph in our 2 documents, most likely within the Terminology sections, explaining that when we say URI, we mean URI Ref -- due 2009-09-09 -- OPEN
<trackbot> http://www.w3.org/2008/WebVideo/Fragments/tracker/actions/103
close ACTION-103
<trackbot> ACTION-103 Write a paragraph in our 2 documents, most likely within the Terminology sections, explaining that when we say URI, we mean URI Ref closed
Silvia has proposed a paragraph
Raphael has slightly updated it, now present in the 2 documents
ACTION-49?
<trackbot> ACTION-49 -- Yves Lafon to draft the HTTP-Range syntax for different units (completing all the syntax for the two way handshake) -- due 2009-09-02 -- OPEN
<trackbot> http://www.w3.org/2008/WebVideo/Fragments/tracker/actions/49
close ACTION-49
<trackbot> ACTION-49 Draft the HTTP-Range syntax for different units (completing all the syntax for the two way handshake) closed
Yves: my proposal is to re-use
the same kind of syntax than for bytes
... for the Range header: Range: <timeformat> '='
<start time> - <end time>
... and for the Content-Range header
... Content-Range: <timeformat> ' ' <real start
time> '-' <real end time> '/' <total
duration>
... since we are not using '-' and '/' in our syntax, it is
safe
... there are a number of things to do ... register some units,
such as npt, smpte-30, etc.
One issue: the duration for some unit might not be easy to define
scribe: would welcome the opinion of Jack and Davy
Yves: I have proposed also
another solution, more flexible, but a bit more complex
... we could then mix units (bad?) but it solves the duration
problem
<Yves> issue is when you are requesting a fragment of a smpte-indexed video
<Yves> if the beginning of the file is not 0:0:0.0
<Yves> then you have an issue expressing duration
<Yves> (if you use the same unit)
Yves: I think that smpte information is embedded in the file
ACTION-69?
<trackbot> ACTION-69 -- Conrad Parker to draw a representation of the general structure of a media resource, for streamable formats (H/H' + K + D1 + D2 + D3) -- due 2009-04-24 -- OPEN
<trackbot> http://www.w3.org/2008/WebVideo/Fragments/tracker/actions/69
I will chase Conrad, we need this picture
ACTION-104?
<trackbot> ACTION-104 -- Yves Lafon to start a thread on the mailing list to summarize the state of the discussion regarding ? and # (? when transcoding happening, #for other cases) + use of URI template for ? -- due 2009-09-09 -- OPEN
<trackbot> http://www.w3.org/2008/WebVideo/Fragments/tracker/actions/104
close ACTION-104
<trackbot> ACTION-104 Start a thread on the mailing list to summarize the state of the discussion regarding ? and # (? when transcoding happening, #for other cases) + use of URI template for ? closed
A long thread has started
only Silvia and I have answered so far
http://lists.w3.org/Archives/Public/public-media-fragment/2009Sep/0016.html
Michael: having 2 ways for
expressing the same thing might make the standard too complex
to implement
... for me, a fragment is definitively a 'hash'
... needs to have another thought
Erik: same here, we need to discuss it with Davy
Raphael: I will put that upfront on the agenda, with background reading, so everybody must have an opinion
<nessy> We have to accept the definitions of URIs where a fragment has a specific meaning
I agree
<nessy> if some of the operations that we require on media documents do not fall under this definition, we cannot do it with fragments
<nessy> it is not our choice to redefine URI fragments and queries
Silvia: the problem is that the
specific meaning for fragments is loosely defined
currently
... and up to many interpretations
<nessy> not really
<mhausenblas> nessy, why not?
Raphael: thus the open question to the TAG group
Summary on protocol issue: http://blog.gingertech.net/2009/09/08/uri-fragments-vs-uri-queries-for-media-fragment-addressing/
<nessy> it clearly states that it is relates to a primary resource
<nessy> "The fragment identifier component of a URI allows indirect
<nessy> identification of a secondary resource by reference to a primary
<nessy> resource and additional identifying information."
<nessy> quoted from http://www.ietf.org/rfc/rfc3986.txt
The problem is that we DON'T know if when you do transcoding, this is part or not of the primary resource
<nessy> a transcoded resource is not identical to the primary resource
Michael: have hack corrib, http://ld2sd.deri.org/corrib/
Silvia: where do you have this statement written?
"a transcoded resource is not identical to the primary resource" <- needs a ref :-)
<nessy> it's in the nature of what a URI is
Michael: motivation, have a collaborative tool to directly generate the test cases in RDF ... instead of editing on wiki and then do a manual conversion
<nessy> different transcoded representations have to be different resources
Silvia: what you say is appealing, but needs to be defined somewhere
Raphael: and in particular, where to define the border?
<nessy> so, if the primary resource is addressed by http://example.com/xxx , how do you address the transcoded resource?
is the black and white version of an image a different resource and not part of the primary version?
<nessy> I can only think of two ways of doing it: http://example.com/yyy or http://example.com/xxx?transcode
<nessy> yes
<nessy> even if the server creates the black and white version on the fly, it still requires a different URI to tell the server to create it
<mhausenblas> see http://ld2sd.deri.org/corrib/mftc.rdf
<nessy> we know that even trying to identify the same resource in different representations for different languages hasn't worked and people generally use index.en.html and index.de.html etc
Raphael: silvia, one could argue
this is a different representation of the same resource, just
degraded, that could be serve with the same URI
... Silvia, not in the SW world, and honnestly, content
negotiation based on languages is very widely deployed on the
web nowadays
<scribe> ACTION: Michael to add the missing test cases in corrib [recorded in http://www.w3.org/2009/09/09-mediafrag-minutes.html#action05]
<trackbot> Created ACTION-108 - Add the missing test cases in corrib [on Michael Hausenblas - due 2009-09-16].
Michael: but please, all, you should give it a try
ACTION-93?
<trackbot> ACTION-93 -- Michael Hausenblas to revisit the TC and see which are effected by the temporal-optional-comma-decision -- due 2009-07-29 -- OPEN
<trackbot> http://www.w3.org/2008/WebVideo/Fragments/tracker/actions/93
-- continue
<nessy> Accept-Language is part of http - there is no content negotiation for transcoding defined in http, so it can only work through query parameters with the current web
Michael: I will add line number
but not explicitely number for TC by next week
... and I will complete my actions
ACTION-82?
<trackbot> ACTION-82 -- Michael Hausenblas to flesh out TC vocabulary re ISSUE-9 -- due 2009-07-31 -- OPEN
<trackbot> http://www.w3.org/2008/WebVideo/Fragments/tracker/actions/82
-- continue (still)
Michael: but it is easier now that Corrib is in place
close ACTION-82
<trackbot> ACTION-82 Flesh out TC vocabulary re ISSUE-9 closed
we don't need it anymore
but the iSSUE remains open
Erik: Davy is wroking on his action and it will be ready for next week
[silence]
no telecon next week but the virtual f2f
I think i know why I cannot have anymore AP, maybe because i have changed affiliation