| Flash Player is a virtual machine. Last time I opened a *.swf file in a hex editor, it was a compressed file (I think zlib). After you decompress the file, you have to parse the structures inside (code, data, images, sounds). After that you have to load them inside the virtual machine and emulate the code. I don't think there is a JIT inside (maybe I am wrong), so each instruction is decoded and emulated on the native CPU. Excluding flash byte code, there is also a native decoder for images, videos and audio. Flash player also has support for UDP, TCP and other things inside (which need cpu and ram to work).
If you open a page with 2 flash ads and a YouTube video embedded in it, you are in serious trouble. Each device has a different CPU, different memory bandwidth, different speed when writing on the internal memory.
My point of view: Flash Player is an excellent product. Adobe should make it open source, invest money in future versions and sell support services (RedHat tactics).
--- On Fri, 12/16/11, Bill Brutzman <bill.brutzman@scottynow.com> wrote:
From: Bill Brutzman <bill.brutzman@scottynow.com>
Subject: RE: [flexcoders] You are the product
To: flexcoders@yahoogroups.com
Date: Friday, December 16, 2011, 8:16 PM
My sense is that Adobe has realized that it close to impossible to port Flash to the staggering proliferation of tablets, smart phones, and other devices. Does anybody expect Flash to run on a Kindle or a Nook? In my little world of fantasy… I wish I knew how Flash worked… Perhaps a standards-based Flash lite could be cranked into HTML-6. --Bill From: flexcoders@yahoogroups.com [mailto:flexcoders@yahoogroups.com] On Behalf Of Kevin MacDonald
Sent: Friday, December 16, 2011 12:50 PM
To: flexcoders@yahoogroups.com
Subject: Re: [flexcoders] You are the product Good points. Thanks for responding. I'm not sure why you conflate me knocking Adobe for a lack of willingness to learn. I code on a daily basis in half a dozen languages for a small company struggling to reach profitability. Our client application is one piece of that. The 'learning' in this case is that some companies can be trusted more than others. Adobe puts forth a consistent marketing message to software developers: "Trust us! Follow us!", and they consistently fail to live up to that in order to sell us the next round of developer tools. Microsoft, while clearly capable of various brands of skulduggery, has consistently maintained a level of loyalty to their developers, and it has succeeded famously for them. Have you every noticed that 15 year old programs still run on Windows 7? I don't expect that from Adobe. But the heavy sell job on AIR followed by stepping at arms length from it irks me.
Kevin
2011/12/16 Csomák Gábor <csomakk@gmail.com> technology simply changes. i met a guy who was the lead engineer of commodore 64. do you think when he was on the top of his career, he stopped learning? this segment changes a lot. it is a lifelong learning. get used to it. html5 is not ready. even w3c says it'll be in 2014 (as i remember). and i think, it won't kill air. neither flash. of course it will depend on a lot of things, but the two technologies are good in different segments. you cannot do a prezi.com in html5, and you cannot do an entire webpage in flash. (login remembers will not work, etc...) the key is to know both, and know when to use what. On Wed, Dec 14, 2011 at 7:02 PM, Kevin MacDonald <kevinmacdonald@gmail.com> wrote: Hello developers, I have come to some unfortunate conclusions about how Adobe operates. I would be interested to get your opinions on the matter. Some years ago I helped build out a desktop application using Macromedia Director. It ran on both Mac and Windows, and was heavily backed by web services. In principle it was much like an Adobe AIR app might be today. After a few years Adobe bought Macromedia Director, with promises to the developer community that they would continue to support it. They came out with a few maintenance releases that were extremely buggy, enough so that we tried to roll back to the previous version. However, Adobe made sure there were some gotchas that made it painful to either stay on the current version or roll back. Shortly thereafter they killed Director altogether. An Adobe evangelist came to our office and sold us hard on moving to Adobe AIR, which we did. We completely re-wrote our application on that platform. Now, several years later, Adobe is very obviously moving away from AIR and towards HTML5, again with promises to their loyal developers to continue supporting it. Based on their history what I expect Adobe to do is kill AIR before too long. And you should have no doubts that they can make it very painful to remain on that platform. For example, AIR apps use whatever version of Adobe Reader is installed on the client machine. Adobe Reader updates happen independently of updates to the AIR run time. The latest update to Adobe Reader broke certain aspects of our client application, something that might directly hurt our business. What can you do when the HTMLLoader object no longer correctly displays a PDF? What I expected Adobe to do - and what the evangelist led me to believe - was that Adobe would evolve AIR and Flash Builder towards HTML5 over time, bringing all of us along with them. But they don't do that. They scorch the earth and start over. So, what's next? I suppose we will hear from Adobe before too long that we should run out, buy PhoneGap Builder 1.0, and once again chase their code-once-deploy-everywhere carrot. We are not the customer. We are the product. We are the means by which Adobe makes money for their shareholders, nothing more. I suppose in true jaded developer fashion this should come as no shock to me. But the truth is, it never feels nice to be a pawn in someone else's game.
|
0 comments:
Post a Comment