39 ms·

> Not that long ago, Mozilla tried to challenge H.264 on the Web. Ahh, hubris. They thought they could force the industry's hand with their market share... whi

• cpeterso 11y ago
  I agree that H.264 was not a fight that could be won, considering existing H.264 content and mobile devices' hardware support. But I'm sure the situation seemed more hopeful when 40% of the browser market (30% Firefox plus 10% Chrome, at the time) would drop H.264, especially when Google also owned YouTube and On2 Technologies (VP8). Dodging the problem by leaving the codec support to the OS would fragment cross-platform compatibility if websites stuck with a Windows-only codec. Today, Firefox does use H.264, AAC, and MP3 decoders from the OS, but AFAIU doesn't expose any other OS codecs to web content. Do you think the current situation with EME DRM is "H.264 all over again"? At least now, we can sandbox the DRM bits and dump NPAPI plugins. EME and CENC commoditize DRM, separating content encoding and DRM. Content providers don't need to reencode their video library to add support for new DRM systems.
  • pdkl95 11y ago
    > fragment cross-platform compatibility No, it wouldn't have, in the long run. H.264 was chosen by the hardware manufacturers (phones, video cameras, etc), and the software world would have fallen in line. As they did. Besides, any temporary problems could have been handled by a fix provided by not-Mozilla, just like how DVD support was enabled in many distros (by downloading libdvdcss from France). > drop H.264 Believed the hardware industry would reverse years of designs based around H.264 is magical thinking. Software follows what the hardware supports (you are not going to get everybody to re-encode their cat videos). Anybody that believed that the browser market could trump the mobile phone market is showing they don't understand how de facto standards work. When Mozilla was arguing against H.264 5 years ago, they were already several years too late. Screaming at the tide to turn back doesn't actually work, even if your intentions are good. > VP8 ...was future technology. Once it was actually released by Google with the WebM container it became a relevant option (which was a good idea to support in addition to H.264 that was already very commonly used. The nonsense before Google released VP8 was merely wishful thinking. Much like how solar power advocates like to include "future improvments" in their claims, claims that H.264 was not needed usually suggested using VP8 while it was still vaporware. > Today, Firefox does use H.264, AAC, and MP3 decoders from the OS Which proves it is possible, and could have been done 5 years earlier to kill off Flash. > but AFAIU doesn't expose any other OS codecs to web content. They shouldn't, in general. Adding another list that can be probed for browser identification wold be a terrible idea. > Do you think the current situation with EME DRM is "H.264 all over again"? I think Mozilla is responsible for EME being completely accepted. Before announcing EME support, it was still possible to argue that EME was not supported by all browsers, an admittedly small reason for websites to not use EME. With Mozilla abandoning their mission (specifically principles #2, #6, and #7) to chase market share and become a Chrome clone, a major battle in the War On General Purpose Computing has been lost. If Mozilla cared about the future of an open internet in the slightest, they wouldn't support any form of DRM, with no exceptions. THAT is the place a principled stand should have been taken, not video codecs. Instead, Mozilla has chosen to let others pick the battlefield. The claim was that this was necessary to safe their market share. Well, that didn't work[1]. Now that Mozilla has given up the fight against DRM, the problem will spread. The ebook industry already wants EME-like support. [1] https://upload.wikimedia.org/wikipedia/commons/8/86/Usage_share_of_web_browsers_%28Source_StatCounter%29.svg https://upload.wikimedia.org/wikipedia/commons/8/86/Usage_sh...
    • cpeterso 11y ago
      Thanks for the information. I cede to most of your points, especially on codecs. :) A few more questions: > Which proves it is possible, and could have been done 5 years earlier to kill off Flash. How would HTML5 video, using OS codecs, kill Flash unless it also had DRM support? Flash has had DRM since 2007 (Flash Player 9). > I think Mozilla is responsible for EME being completely accepted. Before announcing EME support, it was still possible to argue that EME was not supported by all browsers, an admittedly small reason for websites to not use EME. Chrome shipped EME in 2013. Safari and IE shipped EME in 2014. EME content was already live and supported by ~60% of browsers when Mozilla began implementing EME in mid-2014. Perhaps Firefox could continue to lean on Flash and Silverlight to play DRM'd video.. until content providers decide they don't want to continue supporting NPAPI for Firefox's small market share. Flash is insecure and Silverlight is on the path to EOL. > The claim was that this was necessary to safe their market share. Well, that didn't work Whether EME has any effect on Firefox market share is yet to be seen. It just went live this week. Chrome or Edge users are unlikely to switch to Firefox because they want DRM video to stop working.
      • pdkl95 11y ago
        This is the problem with letting the opponent pick the battlefield. If you've decided that "playing DRM video" is a necessary feature, you are choosing to perpetually play catch-up. "The only winning move is not to play." The media industries are very used to being the middlemen that dictate how their industry works. As middlemen, they have the power to control both sides (publishers and consumers). As monopolists, they get to abuse the market by tying/bundling the things people want to the things they want to push ("if you want to watch $POPULAR_SHOW, you need to submit to using our DRM"). They also get to play various financial games, though that is less important re: video. These problems are not technical in nature, and cannot be solved by writing software. Chrome and Edge are colluding to require some type of DRM (that they or their friends control)? They should face the Sherman and Clayton Antitrust Acts. Unfortunately, we have let institutional corruption take over the parts of government that should be enforcing those laws. Choosing freedom instead of submitting to the copyright cartels requires sacrifice. Mozilla could help this with stuff like an educational campaign, but there will always be casualties when fighting entrenched powers. Many people chose convenience instead of investing in their future, so now we are in an uphill battle. It's difficult to convince people that they should choose a healthy browser instead of the copyright cartel's junk-food, and I don't know of any great solutions. What I do know is that giving up the fight and granting those middlemen even more power is only going to make future battles for free software and an open internet even harder. Do you want to tackle is problem now? Or do you want let power accumulate even more and fight a harder problem in the future?
• monocasa 11y ago
  Historically, leaving something with as large of an attack surface as video decoding up to the OSs to update hasn't really worked well for browser security.