Anyone wanting to try hosting their own HD or SD video portfolio in WebM can try http://www.vive.ly/ which is like parts of Dropbox, Zencoder, and Hulu in a blender, for original video files you want backed up, encoded, and published publicly or privately -- without the YouTube problems of giving up your copyrights or having competitor videos promoted alongside yours.
Drop your originals in a folder, we encode into SD and HD in H.264, Ogg, and WebM, and build you a mini-Hulu video site that works in Flash, Silverlight, or HTML5 across browsers. You can also download the encoded files for use elsewhere, embed them, or rebrand the video site with your own logo and domain name.
Use our Hacker News invite code hd4yc to sign up free, and let us know what you think of the WebM encodes. We encode all the files in parallel, and the video pages update as the versions become available, so you may have to wait a bit for the WebM encodes to finish.
If the Vively subscriber sets his settings to HTML 5 or "video for everyone" it will use that by default.
If you visit from iPad, you'll see video in HTML 5 regardless of the user's Flash setting.
Btw, if you post a vertical video shot on iPhone, that works too, it will be shown vertical.
We try to use the playback tech most likely to feel familiar and give good performance for the end user (viewer). A Vively subscriber can set to HTML 5, though, then preference is Ogg for Firefox, WebM for Chrome, H.264 for Safari.
Love it, I want to share your site as a quick tip on our show http://rowshow.com. On Thursday. Can you hook me up with a promo code to give out to viewers? Email in my profile.
The help for logged in users is TenderApp, and the signup welcome email provides several contact emails. You can reach me with my username on gmail. Will add contact info to about page, thanks for pointing out the oversight.
I assume you're either the founder or work for Vively--do you have a contact email or phone number? I have a few questions about this this service and how it might be useful for my company's customers.
As founder, I'd love to hear your thoughts, especially on making it more useful. My email's my username on gmail. I'm open tomorrow to call, just let me know contact info and time/timezone (I'm US/Eastern).
We prototyped that, and by pushing out the HD H264 file we encode to those services, it works, but we wanted to keep this as simple as possible at start. We're thinking about an "advanced settings" control panel to enable those kinds of features: "[x] I'm a pro, let me do more.".
"Currently, there are countless devices used to record videos and hundreds of different video file formats."
H.264 is by far the most common. I don't know of any video camera or DSLR that records to WebM or Theora.
"certain web browsers that you use to view video online only accept certain ‘codecs’ - or programs used to encode, transmit and playback video files - and others require plug-ins (converters) to integrate the video file with the browser."
"Certain web browsers"? Chrome and Firefox are the only ones that don't support H.264 natively. To view video on those browsers, it needs to be routed through a plug-in like Flash (standard on Chrome) or QuickTime. Browsers like Internet Explorer and Safari support the majority of all web video, right out of the box, no plug-ins needed. On mobile devices, the situation is even more obvious. Android is the only OS with support for WebM, and no mobile devices have hardware acceleration for WebM, draining your device's battery in no-time. However, all modern smartphones and tablets have support for H.264, most of them with hardware acceleration.
I'm all for an open source alternative to H.264. If it's at least as good and free to use, then I hope that it will become the standard; for the web, in desktop operating systems, for mobile devices, for video cameras and DSLRs. I'm just not convinced WebM is that alternative. It's unclear whether WebM is as good as H.264, and it's unclear whether it's free to use.
That's a very good point. IE9 has 3% usage share right now, but because it can't run on XP, IE9 won't replace all the IE8 installs out there. In the same vein, IE10 won't even run on Vista. That's why I doubt we'll see another single version of IE being the most dominant. On top of that, Chrome usage is growing rapidly. At the current rate, Chrome will be the most popular browser in 12 to 18 months.
Without plug-ins, Opera Mini doesn't play H.264, WebM, or any other video formats for that matter. The other Opera products have near zero market share. That's why I don't test for Opera, just like I don't test for IE5 and Netscape.
Opera Mini doesn't have plugins, afaik. It doesn't have an HTML parser; it's a thin client which renders pages sent to it in a binary format from Opera's proxy servers.
So just when most people started to finally standardize around one, we're going to introduce a new one that's already opposed by some of the most important players in the industry.
If you're just looking at browsers, then how about Microsoft, which still has a dominant browser market share.
If we're looking at desktop operating systems, Apple and Microsoft are the main suppliers. They provide built-in support for H.264, not WebM or Theora.
However, when looking at content providers, Apple is a very important player. The majority of online video content is sold through the iTunes Store. And rental places like Netflix and Hulu also rely on H.264.
It’s not really sensible to only look at browsers (and, presumably, only desktop browsers – Apple has more than 6% market share of mobile devices). Which formats do digital cameras use? Which formats is content sold in?
Flash is a delivery wrapper (and is often used to deliver h.264; I don't believe Adobe has announced any plans to support WebM) - the thing being discussed here is the format for the video itself...
In the last 24 hours, we've had 12 years of video uploaded to Justin.tv. 12.17 in particular. So more than double YouTube :-)
Think about it this way: if you have an average of 365 people uploading a video at any given time, you get 1 year of video per day. So YouTube probably has around 2,000 people uploading a video at any given time. (Compare to the number of channels on Justin.tv at any given time...)
That maths only adds up if they're uploading a live stream, but, correct me if I'm wrong, the majority of YouTube uploads aren't that. So if they had exactly 365 people uploading videos for exactly 24 hours, it could be much more or much less than one year of video, depending on how fast their connection allowed them to upload.
Yes, that's true. So YouTube probably has far fewer than 2k people uploading at any given time, since most people can upload faster than realtime (though some are probably slower than realtime).
But it wouldn't, on YouTube the average number of people uploading doesn't matter. One person, if his upload speed was fast enough, could upload a year of video in one hour, theoretically.
It will be interesting to see if there's a noticeable quality difference. Right now WebM is not comprable to H.264 except the baseline codec.
Its openness allows anyone to improve the format and its integrations, resulting in a better experience for you in the long-term - This line doesn't jive at all with my understanding of the VP8 codec. I thought implementations were left to devs, but the format was locked and didn't have room to be improved on? Either way the whole spec seems to be built around avoiding H.264 patent suite and "improving" on it is a minefield.
From comments DarkShikari gave (http://x264dev.multimedia.cx/archives/377), it looks like the format was frozen as soon as they released it, so that talk of "improving the format" is rather dubious.
"Right now WebM is not comparable to H.264 except the baseline codec."
It's comparable in the sense that AAC and MP3 are comparable. VP8 can look excellent, it's just that high profile h264 can look better at the same bitrate.
It's also extremely likely that youtube uses h264 baseline profile anyway, since they serve to mobile devices (unless they secretly transcode and store multiple h264 versions for every video).
Pretty much every video content provider I know does this. I work primarily on OTT Boxes (Roku, Boxee) and Connected TVs and all of those platforms support bitrate selection of video assets based on a speed tests. I'd assume most platforms doing progressive download of video over HTTP does the same.
The HTML5 trial has always been opt-in, but also features H.264 files. WebM has been the default for HTML5 since they introduced it, only falling back to H.264 if WebM is not available (or back to Flash if the video require adverts).
I think the big change is that they're now converting all uploads to WebM, previously they had to be above 720p OR uploaded in WebM to trigger this. Presumably because they've converted enough of the popular "classic" videos that they can now spare the encode time for all new files. Just another milestone on the journey towards having everything in WebM.
The summary is that new videos are being transcoded to WebM, whereas the rest is going to take a while. The fact that the transcoding is already partially underway and I haven't noticed any difference probably speaks well for how the transition has been going (admittedly, I generally keep my browsers very up-to-date). EDIT: Aaah, just noticed the opt-in part.
It's good they're doing this, but since I've switched to a browser with WebM support a few months ago, and opted-in to the trial, I've run across only a couple of videos not in WebM.
So far we’ve already transcoded videos that make up 99% of views on the site
or nearly 30% of all videos into WebM
There really isn't much of a long tail on YouTube, it seems. It's sort of worrisome that google could throw away seventy percent of the videos with the vast majority of users not caring.
The vast majority of video views (99%) would be unaffected but that doesn't imply that the vast majority of users wouldn't be since most users are going to view way more than one video.
I don't see any implication that YouTube/Google is throwing away anything. They stated that they chose to transcode a selection of videos that the vast majority of viewers view. They didn't imply that they weren't going to transcode the remaining videos, but that they feel the number of videos that they have transcoded is sufficient enough to declare WebM as being the default codec.
Really, that statement isn't any more worrisome than the unstated fact that Google could potentially delete every video on YouTube and every single email in Gmail and Google Apps on a whim.
That's the wrong way of looking at it. Most people upload stuff few people care about. Their is a large number of people who cater to a small community, and YouTube makes sharing videos easy for that community. A low number of views doesn't equate to low interaction, either.
Take a hobby of mine: miniature painting for $40k. I published battle report videos, painting update videos, etc. I didn't have a large number of views, but the views I got were from highly specific audience. So, I get 500 to 1000 views. That's clearly something that isn't high priority to the masses of YouTube, but for the community that has built up around sharing videos like that on YouTube, it's important.
From a content provider point of view, there is no point in creating WebM content. The H264 version works on all plateforms, why would they be interested in paying for transcoding time to access a market they have already...
It might sound great, but I do not see the industry following Google on that...
Drop your originals in a folder, we encode into SD and HD in H.264, Ogg, and WebM, and build you a mini-Hulu video site that works in Flash, Silverlight, or HTML5 across browsers. You can also download the encoded files for use elsewhere, embed them, or rebrand the video site with your own logo and domain name.
Use our Hacker News invite code hd4yc to sign up free, and let us know what you think of the WebM encodes. We encode all the files in parallel, and the video pages update as the versions become available, so you may have to wait a bit for the WebM encodes to finish.