Comments on: Blackwave, says it has major storage improvement for video http://venturebeat.com/2007/12/07/blackwave-says-it-has-major-storage-improvement-for-video/ News About Tech, Money and Innovation Tue, 24 Nov 2009 05:47:06 +0000 http://wordpress.org/?v=2.8.5 hourly 1 By: Ramon http://venturebeat.com/2007/12/07/blackwave-says-it-has-major-storage-improvement-for-video/comment-page-1/#comment-669258 Ramon Fri, 07 Dec 2007 22:37:37 +0000 http://venturebeat.com/2007/12/07/blackwave-says-it-has-major-storage-improvement-for-video/#comment-669258 I think that's what's wrong about video compressing. It's not only about binary but a multiple sequenced scenes, a lot of things within the video are repeated data / patterns. I don't think they are in it for compressing the binary but actually finding multiples and then compress it. But like you said a server farm is more efficient due to cheap memory, I guess their approach to the market has to be the key to the solution. I think that’s what’s wrong about video compressing. It’s not only about binary but a multiple sequenced scenes, a lot of things within the video are repeated data / patterns. I don’t think they are in it for compressing the binary but actually finding multiples and then compress it.

But like you said a server farm is more efficient due to cheap memory, I guess their approach to the market has to be the key to the solution.

]]> By: Jason http://venturebeat.com/2007/12/07/blackwave-says-it-has-major-storage-improvement-for-video/comment-page-1/#comment-669244 Jason Fri, 07 Dec 2007 22:13:51 +0000 http://venturebeat.com/2007/12/07/blackwave-says-it-has-major-storage-improvement-for-video/#comment-669244 I'm highly skeptical of Blackwave's claims. Granted I haven't actually seen their product in operation, and my skepticism isn't likely to cause even a hiccup in their plans to take whatever they have created to market, but it just seems too fishy. First off, Binary data is always binary data, and many smart people have done a lot to improve the efficiency of storing binary data. Lightweight file systems have greatly reduced any gains that might be had by "improving" things. If they have some compression routine it isn't likely to do much for already well compressed video (just take any compressed video and try to ZIP it, RAR it, GZip it, etc and see how much improvement you get), and it would only increase CPU requirements when hosting (disk space is cheap, and getting cheaper). The real tell for me, is in the teaser element about scanning meta data for User Generated vs. Commercially Produced and tweaking the profile. The thing is you shouldn't need to do anything like this on an actual video hosting service. The context of how you received the video would tell you enough (and not require any algorithms), and popularity is a moving target, so a hit counter and play rate calculator would give you enough pieces to know if you need to implement a different hosting strategy. So if the algorithms are obvious and high performance hosting can be achieved by building a smart media server farm (which replicates popular content to decrease individual server load) out of Linux and commodity hardware, then all Blackwave could really be bringing to the table is a turn key server solution which would only be valuable until the open source community has a comparable solution (which they very well might by now already). Its these extra useless details that tend to suggest an investment opportunity that is more smoke and mirrors than actual substance. But if they have $16 million in the bank, then they should have enough money to prove all of their naysayers wrong (unless they aren't spending it wisely) I’m highly skeptical of Blackwave’s claims. Granted I haven’t actually seen their product in operation, and my skepticism isn’t likely to cause even a hiccup in their plans to take whatever they have created to market, but it just seems too fishy. First off, Binary data is always binary data, and many smart people have done a lot to improve the efficiency of storing binary data. Lightweight file systems have greatly reduced any gains that might be had by “improving” things. If they have some compression routine it isn’t likely to do much for already well compressed video (just take any compressed video and try to ZIP it, RAR it, GZip it, etc and see how much improvement you get), and it would only increase CPU requirements when hosting (disk space is cheap, and getting cheaper). The real tell for me, is in the teaser element about scanning meta data for User Generated vs. Commercially Produced and tweaking the profile. The thing is you shouldn’t need to do anything like this on an actual video hosting service. The context of how you received the video would tell you enough (and not require any algorithms), and popularity is a moving target, so a hit counter and play rate calculator would give you enough pieces to know if you need to implement a different hosting strategy. So if the algorithms are obvious and high performance hosting can be achieved by building a smart media server farm (which replicates popular content to decrease individual server load) out of Linux and commodity hardware, then all Blackwave could really be bringing to the table is a turn key server solution which would only be valuable until the open source community has a comparable solution (which they very well might by now already). Its these extra useless details that tend to suggest an investment opportunity that is more smoke and mirrors than actual substance. But if they have $16 million in the bank, then they should have enough money to prove all of their naysayers wrong (unless they aren’t spending it wisely)

]]>