The prior meeting minutes were approved without change.
<glennd> Reached out to IETF to provide a BOF (or Bar BOF) of GGIE on the work we are doing; gain more clarity on related IETF work; convey that we have some ideas that may be of interest to them; and to discuss items we have identified that the IETF might be positioned to work on.
<glennd > Areas of interest: RFC around content identifiers: Similar content IDs have been defined such as EIDR, Ad-ID, but GGIE’s concept of Cid is not limited to those two. Could also include e.g. private for content IDs. There is potential for work that builds on existing IDs based on some of GGIE’s Use Cases that were developed as part of the W3C Web and TV WG. Potential for a look-up system to store/retrieve information related to content based on identifiers.
<KilroyH> Elaboration of ultraviolet features as relate to the proposed GGIE BoF. http://www.myuv.com. UUIDs identify bit streams but don’t give an indication of relationship between works, metadata. Suggest doing a survey of content metadata that’s out there as a reference
<glennd> Haven’t heard back from IETF about whether or not this will be a formal BoF or a bar BoF or what.
<glennd> Slide deck is being prepared by Bill Rose as a showpiece for GGIE at a potential informal session during TPAC in Sapporo. Basic idea is columns are actors with description of what each actor does. Is this format OK? Changes, new format?
<?> Need to capture difference between static adaptive and dynamic adaptive systems
<KilroyH> Good format for the basic UC-1. Suggest adding colors to illustrate differences in build.
<glennd> Does this make sense for the audience?
<LeslieD> need to look at how it works for more complex UCs.
<glennd> UC-1b. Label columns with actors. Moving to UC-2.
<KilroyH> Not sure what this adds to UC-1b?
<glennd> Client is doing control and doing byte fetch. Not just simple start the flow.
<LeslieD> Format is working so far. Adding labels and colors to show how they change from the previous will help.
Action Item: Bill Rose to label and color code actors and highlight what’s new.
<NiloM> Lower right does not fit the actor/column paradigm.
<glennd> Moving to UC-3. Adds extraction of the Cid. 3 ways to extract it. Metadata, embedded A/V watermark, extract A/V fingerprint.
<BillR> This one didn’t fit into the previous format.
<glennd> Looks like we are evolving from metadata to watermark to fingerprint. Stack them instead of using arrow.
<glennd> Maybe we don’t make this a UC but a foundational idea from a UC. Get rid of the arrow in the background. Looks like the format is holding up with some edits.
Action Item: Bill Rose to change the slide to eliminate the confusion.
<KilroyH> Suggest keeping the UC slides simple, show difference between each, focus on user experience. Select assets, move on to trick play, not sure what UC-3 adds to user experience. Getting into network mechanics will get complex and cause confusion and comments on all of the ways it could be done.
<glennd> Some people will want to get down into the engineering details but the IG focus/scope is not how to do it but what we want to be able to do.
<KilroyH> Suggest a title for the presentation might be “here are the problems we are trying to solve”.
<glennd> We will give some of the background on what we are doing, scope, etc. Then move to the UC slides. Waiting to hear from the chairs. Scheduled to meet on Monday Oct 26th. There will be remote access capability.
<glennd> Going forward we will take the rest of the use cases and put them in this format; circulate to this group for review and comments. Oct 7th meeting will be a full 90 min, as we’ll walk through the whole deck as it is intended to be presented in Sapporo No call on October 21, 2015.
<glennd> Next call is Wed Oct 7 (last call before Sapporo TPAC). No call on Oct 21 or on Nov 4th.
<glennd> Meeting adjourned