Open Video Conference notes (Friday)

[elz]

So the Open Video Conference was held last weekend at NYU. There seemed to be an interestingly eclectic group of people there: open source software developers, startup folks, media company folks, lawyers, academics, journalists, filmmakers, remix artists, educators, librarians, people interested in video and politics, etc. I don't have any real standard of comparison, but it seemed to me at least like there were a fair number of women there, and that women were pretty well represented among the speakers and panelists. There were also some interesting-sounding talks about open video in the developing world, which I sadly didn't get a chance to go to. Francesca Coppa and Naomi Novik were also there representing the OTW (woohoo!) - officially, I guess I was scouting out technologies we might be able to use to build a vidding section for the Archive of Our Own, but I'm a pretty big geek, so I would happily listen to people talk about codecs and copyright law and HTML 5 all day with no ulterior motives. :)

Day one/Friday: had some minor travel snafus, so I didn't get there until 11. Kicked around a little bit trying to figure out when talks were actually starting and ending and if there was actually a room where I could get a seat (not so much, as it turned out).

Metadata Roundtable I: Time-Based Metadata

Wound up missing the beginning and sitting on the floor behind people sitting in chairs, so my ability to see and hear what was going on was kind of limited. Still, I managed to glean a few interesting things from it. People are working on systems to tag/index videos by time codes and content so that it's easier to navigate within a video and to find videos that actually have the content you're looking for. There's a site called Metavid, for example, that's parsing CSPAN videos of the US Congress and putting them up online; they were getting 14-hour, undifferentiated chunks of video and have had to come up with ways to make something useful and searchable out of that. Likewise, the BBC, the Tate Museum and other educational institutions have all this video that could be much more useful and accessible to a wider audience if it were easier to search and navigate.

So the challenges seem to be:

1. How do you attach metadata to video from a technical standpoint



2. Where do you get the metadata (closed-captioning, crowdsourcing)



3. Who owns the metadata (big content companies may consider that information as proprietary as the video itself)



4. How do you verify that the metadata you have is accurate and what do you do with information that isn't accurate



5. How do you come up with standard ways of organizing this information so that it's easy to process video from different sources

From a fannish POV, this was not super-relevant: vids are generally short and don't contain speech and wouldn't really need that sort of indexing, and vidders don't tend to want to make their vids easily findable for the widest possible audience. Still, the possibilities sound really neat (especially in terms of educational video), and you can imagine a time when you could have digital copies of movies and tv shows where there's an easy, standardized way to jump to a specific scene or to search for a particular line of dialogue.

Lunch

Ate half my burrito before it lost structural integrity; spent most of the break trying to figure out what was wrong with my laptop. I was psyched to at least snag a bit of stone outcropping to sit on!

Independent Video Platforms

The seminar room was as cramped as it had been in the morning (now with bonus increased temperature and with the whole building smelling like burritos), but we at least had a patch of floor with a view of the screen. This was interesting - there were 7 or 8 people representing mostly social justice and independent filmmaking sites. There was some discussion about the difficulties of competing with YouTube and how to get more viewers and more attention, and how YouTube changed the landscape for independent video, but the most interesting comments to me were about the advantages of not hosting your content on YouTube or another large commercial site: greater freedom of expression, more control over your content, less danger of being turned over to your oppressive government, a greater ability to provide your content to people in developing countries who are often viewed as an undesirable audience by advertising-supported sites. The guy from Pad.ma (a site for independent filmmakers in India, which looked like it had some pretty cool UI features) talked about the challenges of providing video to Indian audiences who often have low-bandwidth internet connections.

So definitely some interesting stuff, although for the most part, the sites represented had issues that were amusingly antithetical to the issues we would/will have hosting vids on AO3: we're a lot less interested in attracting millions of viewers than we are in being able to afford to serve video to an active, existing community.

Emerging P2P Technologies

Before this, we took a break to talk a bit about vids and the AO3, and the different options we might have for providing the best possible service for people without maxing out our resources. And then this talk had perfect timing in offering a solution to exactly that problem: Riccardo Petrocco from P2P-Next was talking about his project, which aims to provide streaming video via bittorrent (and other similar systems) so that you can watch good-quality video in your browser as you download it without putting a tremendous load on the site that hosts the video. Thus an incredibly popular video, instead of killing your site, would just load faster the more people watched it. The software is being developed in conjunction with the BBC and other media companies but will be open source.

Obviously, if this works out as planned, it would be a tremendously useful tool for all kinds of sites. The downside, of course, is that it's not done yet. They do seem well on their way, though, and, IIRC, plan to have a release candidate by the end of this year. I'll definitely keep an eye out for that.

How to Make a Political Remix Video

From the title, I'd originally been expecting something more political in the non-fictional sense, but instead, Jonathan McIntosh of Rebellious Pixels showed his Buffy vs. Edward vid and talked about the process of making it. There were very few people in the auditorium who'd actually seen Twilight, which says something about demographics, but everyone seemed to enjoy it - I definitely did!

I was going to have more commentary here, but I think everyone on the internet had seen this by the time I got home on Saturday. *g* I was thinking this week about how much further it spread than the average vid does - it definitely offers a storyline that's appealing to a lot of people (ahem), but I wonder if it's also more accessible simply because it uses dialogue instead of music. (Harder to do in a different way, though, because you actually have to get the characters to talk to one another.) Vidders can do amazing things with the juxtaposition of music and lyrics with video, but it is, in a way, a language you have to learn how to parse.

Anyway, tangent. So that was day one! I actually went to more things on day two, but nothing that made me go "ooh, ooh, ooh!" quite the way the P2P-Next project did.

Comments

Hello my friends!

[(Anonymous)]

Hello my friends!