Changeset 827

Show
Ignore:
Timestamp:
2005-02-03 09:40:05 (4 years ago)
Author:
silvia
Message:

Included the default time specification again because it's too painful to always specify.

Files:

Legend:

Unmodified
Added
Removed
Modified
Copied
Moved

  • standards/draft-pfeiffer-temporal-fragments-current.xml

    r662 r827  
    1 /* Magnus Westerlund: 
    2  - make it just for Annodex and suggest it for other formats, too 
    3  - not to have a default time spec 
    4  - not to bother with rtsp queries, just use rtsp fragments 
    5 */ 
    6  
    71<?xml version="1.0" encoding="UTF-8"?>  
    82<!DOCTYPE rfc SYSTEM "rfc2629.dtd"> 
    …… 
    210204      <t>URI query strings are specified after the special character 
    211205      "?".  A temporal query parameter defines one or more intervals 
    212       of time.  Time is given with respect to specific time 
    213       schemes. The available time schemes and their specifications 
    214       are described further down in this document. 
     206      of time.  Time is given with respect to specific time schemes. 
     207      The default time scheme is "npt" (normal play time), in which a  
     208      time point is specified in seconds through a floating point number 
     209      with an arbitrary temporal resolution. The available time schemes 
     210      and their specifications are described further down in this document. 
    215211      </t> 
    216212 
    …… 
    219215 
    220216      <artwork><![CDATA[ 
    221 time-parameter = "t" "=" timescheme ":" time-intervalspec 
     217time-parameter = "t" "=" [ timescheme ":" ] time-intervalspec 
    222218 
    223219time-intervalspec = time-interval ["," time-intervalspec] 
    …… 
    257253        </t> 
    258254 
    259         <t>http://www.foo.bar/video.anx?t=npt:15.2-18.7 --- video .anx 
    260         is transferred from 15.2 seconds into video.anx to 18.7 seconds. 
     255        <t>http://www.foo.bar/video.anx?t=15.2-18.7 --- video .anx 
     256        is transferred from 15.2 seconds into video.anx to 18.7 seconds; 
     257        the default time scheme "npt" is used. 
    261258        </t> 
    262259 
    …… 
    287284      temporal view onto a Web resource consisting of one or more 
    288285      intervals of time. Again, time is given with respect to specific 
    289       time schemes as defined below. 
     286      time schemes as defined below, "npt" being the default. 
    290287      </t> 
    291288 
    …… 
    295292 
    296293      <artwork><![CDATA[ 
    297 temporal-fragment = timescheme ":" time-intervalspec 
     294temporal-fragment = [ timescheme ":" ] time-intervalspec 
    298295        ]]></artwork> 
    299296       
    …… 
    324321        </t> 
    325322 
    326         <t>http://www.foo.bar/video.anx#npt:15.2-18.7 --- specifies a view 
    327         on the section between 15.2s and 18.7s of video.anx. 
     323        <t>http://www.foo.bar/video.anx#15.2-18.7 --- specifies a view 
     324        on the section between 15.2s and 18.7s of video.anx, with a 
     325        default time scheme of "npt". 
    328326        </t> 
    329327 
    …… 
    640638 
    641639          <t>Restricted the specification for Annodex and CMML 
    642           currently with a suggestion only to adopt it for other 
     640          currently with only a suggestion to adopt it for other 
    643641          formats. 
    644642          </t> 
    645643 
     644<!--  
    646645          <t>Deleted the default time scheme according to a recommendation 
    647646          by Magnus Westerlund, Audio/Video Transport Working Group at 
    648           IETF. 
     647          IETF. Put it back in because it's such a pain to always specify. 
    649648          </t> 
     649--> 
    650650 
    651651          <t>Replaced the word "timebase" with the word "basetime" as that 
    …… 
    654654          </t> 
    655655 
    656           <t>Added RTSP specification. 
     656          <t>Started a section on how to use timed URIs with RTP/RTSP. 
     657          Some initial feedback from the AVT Working Group at IETF seems 
     658          to indicate not to bother with rtsp queries and just to focus 
     659          on fragments. This may not be quite conformant to the URI RFC 
     660          because these RTP/RTSP fragments would be handled on the server 
     661          - yet this has not been resolved for any use of fragments over 
     662          RTP/RTSP. 
    657663          </t> 
    658664 
    …… 
    940946      <t>Many thanks also to the Audio/Video Transport working group 
    941947      at the IETF (avt@ietf.org), foremost to Magnus Westerlund, for 
     948      giving valuable feedback on how to improve the specification and 
    942949      picking up the discussion around how to use the scheme on 
    943950      RTP/RTSP. This is now not covered in this document, but should