re "submarine patents" - I get the impression that is an excuse left to last - its like invoking "national security" as a reason to not disclose something (that everyone knows is unrelated) - a catch all excuse, cop out etc... (I could be wrong, but it reaks of that - there are always patents around, on everything, its quite shocking).
On Jul 5, 3:24 pm, Reinier Zwitserloot <[email protected]> wrote: > Even if you have some sort of personal vendetta against the xiph crew > and want to encode just once, using the <video> tag (with fallback to > using a flash player to render your h.264 file) has plenty of > benefits, so any perceived or real shortcomings of Ogg Theora aren't > too relevant to danger that the <video> tag presents to flash. > > So, why would you double-encode: > > 1. (MAJOR): Because you're going to piss off firefox users. Firefox > has taken a stand and will not, as far as I understand, run <video> > tags unless there's an ogg theora source in there. > > 2. (ARGUABLE): Licence-free means its much easier to gain some > ubiquity; you're essentially future-proofing your content. Whether > this matters to you, or if its even a sound argument is a complex > issue that's being debated all over the interwebs and is probably not > too relevant for this forum. > > However, just reason #1 (no firefox) is more than enough to spend the > extra harddisk space. harddisk space is at about 50 bucks a terabyte > these days. Unless you're youtube or vimeo, I'm having a hard time > seeing how 'I wanna save space' is going to fly as an argument. Maybe > the encode CPU time, where, last time I checked, Ogg Theora is quite a > bit slower. A lot of the patents Ogg Theora has to swerve around > involve efficient encoding tricks. That's not relevant for static > content, though. > > Submarine patents: Yes, that's a problem, but note that Opera HAS gone > for it, and they are a company as well. If there are submarine patents > out there, they are expiring, and the longer you wait to come forward, > the less legal standing you have. It becomes kind of hard to honestly > claim in front of a judge that you had no clue all this stuff was > happening, and (IANAL!) there's some onus on the patent owner to > defend the patent when infractions are noticed. That's not too > relevant in a legal culture where tossing enough greenbacks on the > scales of Lady Justice will tip em, of course. > > quality-per-bit: Ogg Theora is very close. Google doesn't close their > body and html tags on their pages because browsers can handle it and > it saves them 8 bytes per transfer. Sounds ridiculous until you > realize the crazy amount of traffic google's servers have to handle, > so every bit counts for them. That's a _very_ niche argument for > everybody else, though. Bandwidth isn't exactly going to bankrupt you, > these days. It's not Ogg's fault: They can't use certain tricks > because there are some bullshit patents on common techniques that they > nevertheless avoid for lack of legal funds and a solid dedication to > playing it as safe as they possibly can. > > Long story short, though: Okay, then don't encode ogg. The <video> tag > is still going to put a serious dent in actual flash usage. Especially > on the fastest growing segments of the market, users will get annoyed > at lack of <video> tag based sources (netbooks and phones, which run > with non-windows OSes and have either no flash player or a very sucky > one, and Mac OS X, which as mentioned has a flash player that eats CPU > for no good reason). > > I'm really interested in how google is going to roll with this. They > played ball with Apple and released an API for them to get at the > underlying sources (that's what powers the iPhone YouTube app - > obviously not a flash player). This means there's a 640x480 non- > streaming h.264 copy (or that's the actual source of all youtube > videos, I haven't checked, and it doesn't matter) for ALL youtube > videos. Why NOT stick that in a <video> tag that falls back to a flash > player? Google goes out of their way to support HTML5 and promote the > vanilla web as the app platform of the future. I'd be confused if they > don't add <video> tags to youtube soon. > > Joe Data wrote: > > Sure, you could "bundle Ogg Theora and H.264", but what's the > > benefit? It seems that there are three reasons against using Ogg > > Theora (see the email announcement at > >http://lists.whatwg.org/htdig.cgi/whatwg-whatwg.org/2009-June/020620.... > > - no hardware video decoding support (cited by Apple) > > - "Ogg Theora's quality-per-bit is not yet suitable for the volume > > handled by YouTube" (cited by Google) > > - uncertainty over whether "submarine patents" will threaten Ogg > > (cited by Apple) > > > The last issue may or may not be real, since Apple has a vested > > interest in pushing iTunes and Quicktime. > > > Anyway, if you encode your video to H.264, you can display it with > > Flash on all computers and "natively" on the iPhone and don't worry > > about these issues above. So again, what's the benefit to add Ogg > > Theora? Sure, you can get the immaterial benefit of "we push open > > source video", but I don't think the added expense will be worth it > > for many businesses. > > > On Jul 4, 7:41 pm, Reinier Zwitserloot <[email protected]> wrote: > > > Not entirely. > > > > if you offer just flash, you create some annoyances for your users: > > > > - It won't work on the iPhone (major reason) > > > - On non-windows machines, it'll light up one CPU core, which means > > > notebook mac and linux users will burn through the battery. > > > - There's no useful right click context menu (e.g. no 'mute' in > > > there. There is <video> tags. > > > > So, what I'm about to describe is not just 'to be more standards > > > compatible', which is good, because 'just being more standards > > > compatible' never made anybody do anything. > > > > Here's what you do: > > > > You encode your video BOTH to Ogg Theora AND h.264 via the MP4 > > > container at 640x480 without streaming (so that its iPhone > > > compatible), and then: > > > > follow the instructions athttp://camendesign.com/code/video_for_everybody > > > > This gets you a nice fallback, where the <video> tag is used offering > > > both ogg and h.264, which covers Safari, Firefox 3.5+, Opera10, and > > > Safari iPhone, as well as flash as a fallback, which covers older > > > versions and IE. It then falls back further, to a download link. > > > > As its all nicely bundled up, the effort to do this is minimal, and > > > hosting your own video has always been quite an endeavour (you need to > > > figure out how to encode and all that - that's why so many people just > > > embed a youtube video!), so I doubt the technical difficulty of doing > > > this is going to stop people from adding video tag powered videos to > > > their websites. > > > > On Jul 4, 4:20 pm, Karsten Silz <[email protected]> wrote: > > > > > Hi, > > > > > Some people thought that the upcoming HTLM 5 with standard audio and > > > > video tags would spell the end of Flash (and Silverlight and JavaFX). > > > > I never thought it would because these plug-ins offer much more than > > > > just video and audio. > > > > > However, it seems now that there will be no standard audio and video > > > > codecs in HTML 5, which means that unless a de-facto standard emerges > > > > somewhere down the line, Flash with H.264 video will continue to > > > > deliver video to the browser masses. For more details, > > > > see:http://tech.slashdot.org/story/09/07/02/184251/Browser-Vendors-Force-... > > > > > In somewhat related news, XHTML 2 seems to have been canceled, making > > > > HTML 5 the only new HTML version going > > > > forward:http://tech.slashdot.org/story/09/07/03/1447237/XHTML-2-Cancelled > > > > > --- > > > > Karsten Silz --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "The Java Posse" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/javaposse?hl=en -~----------~----~----~----~------~----~------~--~---
