The future of flash, Google, Vp8, and Internet video (I)

See: http://x264dev.multimedia.cx /? P = 292

 

 

The future of flash, Google, Vp8, and Internet video

This will be a longer post than usual, because it will cover a lot of content.

The Internet is filled with complaints about flash for a long time, so it sounds like the whole Internet is shouting. However, of course, despite constant complaints, they are right: Flash has poor performance on platforms other than windows x86 and does not seem to be concerned by Adobe. However, compared to the repeat debate that continues to make people vomit, let's move our minds to try to find out what happened.

Flash is popular because of its functionality and flexibility.

In addition, before flash, the mainstream (network) video formats were Windows Media, real and quicktime: all of these are private and there is no free software encoder or decoder, and (except for Windows Media) require users to install bulky external applications, rather than just a plug-in. The reason for Flash's victory is obvious: it supports open multimedia formats, such as H.263 and MP3. Anyone can write files (FLV) in ultra-simple container formats ), it is lighter and more reliable than other options.

 

Therefore, Adobe (in fact Macromedia at the time) can obtain 98% of its initial users. With this, they began to become complacent. The opinions of any competitor were immediately ignored. With such a huge user base, how can someone compete with Adobe? This must be crazy. No one can do this. They made a big mistake in software development: I believe that only one contestant will do better in the competition. For x264, if we find a competitor doing better, we will immediately study and try to keep ourselves ahead. That's why x264 is the best video encoder in the world. But for Adobe, this (positive) attitude obviously disappears after it becomes a monopoly. This is the real danger of monopoly: it hinders development because the monopoly has no motivation to improve their products.

In short, they believe in themselves (they drank their own kool-aid ). But they are wrong on several key points.

The first error is that Linux and OS X are irrelevant.Linux is an operating system that is only used by a small number of end users, but these users account for a large part of software development and network developers around the world. The conclusion that Linux is not worth optimization is obtained only by using the number of users. Therefore, Adobe assigns a developer with only one independent developer responsible for the entire Linux platform. In terms of OS X, Apple computers have become more popular, especially over the years, developed by the same group. In addition, Apple is a huge company; Flash's poor performance on their platforms has aroused Apple's strong resistance. Therefore, Adobe makes itself and apple and developers an enemy.

The second error is the attack on free software.Almost all websites on the Internet use free software on their own servers, not just lamp and other things. Both YouTube, Facebook, Hulu and vimeo use FFMPEG and x264. H. the 264 encoder is so bad that it is much worse than FFMPEG's H.263 or theora; they actually assume that users will use x264 (instead of their own H. 264 encoder ). In actual server software, red5 is particularly popular in rtmp-based systems. However, despite this, Adobe has sent a legal ban to servers running rtmpdump (served a cease & desist order to servers hosting rtmpdump), claiming (absurd) that it (rtmpdump) it violates DMCA because it allows users to save video streams to their hard disk. Rtmpdump is not dead. Of course, this is only a program, but this attack has a sustained impact on developers around the world. This clearly shows that Adobe is not a friend of free software.

The third error is that the open flash implementation is not supported.The lack of a good open flash client is not really Adobe's fault; obviously, gnash's team is completely incompetent and no one else is interested. Cody Brocious spent a few days writing his own Flash Rendering code for a flash-to-iPhone app converter. He stopped development only because he planned to release it a few days ago, adobe released their own conversion programs. The Flash specification is open, and each decoder in Flash has an open software implementation: it seems that there is nothing to block a good open implementation. However, Adobe's mistake is not to do: They didn't push it because it is not important to them.

In contrast, let's look at the Free Software Implementation of moonlight and Silverlight. Microsoft has been actively working with the free software community to help moonlight. Think about how ridiculous it is. Microsoft, the root of Free Software, Linux supporter, has been actively supporting a lgpl free software project, not Adobe! The biggest problem is the monopoly caused by this: People think that flash is not safe, because only one implementation makes them at the mercy of adobe. In any situation, once a file format has multiple popular implementations, abuse by either party will become very difficult. Of course, this is what Adobe intends to do: they want to abuse power, which is why they do not support another implementation.

It is self-evident why flash is so disliked. It is far from the most insecure browser plug-in. According to secunia's research, Java has more vulnerabilities. Of course, it is neither the least reliable nor completely confidential. As mentioned above, this provision is open. However, due to the above three mistakes, Adobe has become the enemy of developers all over the world.

So what should we do now? Flash is useless, we hate flash, but how can we get rid of flash, at least in the aspect of internet video?

 

To be continued...