On Sat, Sep 27, 2008 at 4:59 PM, Michael Snow wikipedia@verizon.net wrote:
To illustrate this with an example, maybe not the best but one that comes up often enough, consider video file formats. (Some of this is beyond my technical expertise, so please forgive any misstatements.) Adobe Flash has widespread adoption to the point of being near-universal. The company has also been moving to make it more open for people watching, distributing, and working on content in this environment. It's close to free, but I understand there are still some issues like patent "encumbrances" around Flash. Meanwhile, there are pure free software formats that do similar things but have pretty limited adoption.
Greg answered this much better than I possibly could, but I'll just chip in and say that more or less this entire paragraph is predicated on misconceptions. The biggest thing to keep in mind is that anything Flash (which is non-free) can do, Java (which is free and open-source) can also do, *and does*. Even if users' browsers don't natively support Ogg Theora, they can still view videos on Wikimedia without any extra setup using Cortado, which is packaged in the page. Go ahead and browse to any old video on Commons in your favorite browser. IE6 on Windows or whatever you like. You should be able to play it just fine:
http://commons.wikimedia.org/wiki/Image:22-digital-clock-screencast.ogg
So any suggestion that we compromise is a solution looking for a problem, unless there's something I'm missing. If the video playback on Wikimedia is worse than on YouTube, as Greg says, that's because video is YouTube's entire business, and it's an afterthought for us. With the appropriate manpower, our video playback could be polished up a lot.
In dealing with the limited adoption of certain free formats, some people have advocated a more evangelistic approach, if you will. Given the reach of Wikipedia in particular, it's suggested that our policy could push wider adoption of these formats. That may be, but the question is, how much is that push worth? What are the prospects for making those formats readable in the average reader's environment, and encouraging wider use as a standard?
Very high. I doubt that Firefox 3.1's support for Ogg Theora in <video> (you can try it out in nightly builds right now, it mostly works) would have come about as soon as it did if Wikipedia didn't serve video solely in Theora format. Opera is the other browser that has experimental Theora support in some builds, and look at this quote from an Opera developers' page:
"The Ogg Theora format is a promising candidate, which has been chosen by Wikipedia." http://dev.opera.com/articles/view/a-call-for-video-on-the-web-opera-vid/
Wikipedia's influence here is important. We're a top-ten website and we can help to push web standards in a positive direction.
Unfortunately there's some reluctance right now to supporting Theora natively and by default in some browsers, like Safari, due to the risk of submarine patents. But those browsers can still use Cortado just fine, so again, no users lost.
Does an uncompromising approach result in significant progress, or would we simply be marginalizing the impact of our work? And is it worth the "sacrifice" of the many people who would miss out on some of the knowledge we're sharing, because the free format isn't accessible to them? (That's also partly a problem of disseminating knowledge, of course.) If we adopt a compromise position as described earlier, how much do we lose in terms of promoting the freer formats?
There might hypothetically be some situation in which it would be advantageous for Wikimedia to support non-free file formats, but video is not one right now.
On Sun, Sep 28, 2008 at 6:37 AM, Henning Schlottmann h.schlottmann@gmx.net wrote:
Send a message to the public W3C-list, signed by you and Brion, that the Wikimedia Foundation wishes to hold a six-month discussion phase on which free, open video standard to use on MediaWiki and the Foundations wikis. State that you want all the browser guys involved in that discussion because you want this standard to be natively implemented in all consumer browsers within 18 months from now. Send it out as a press release as well.
You realize that this discussion has already been going on for a long time on the HTML5 discussion list, right? For a considerable period, the HTML5 standard explicitly recommended that browsers support Theora for the <video> tag. That was dropped when Apple and Nokia objected that they didn't want to support it for fear of submarine-patent lawsuits:
http://yro.slashdot.org/article.pl?sid=07/12/11/1339251
Everyone's aware that Wikimedia supports Theora, and it *has* made a difference. But video on Wikipedia is hardly so make-or-break that we could strongarm anyone into supporting our format of choice -- especially since their users can view it anyway as long as they have Java installed.