The first trial of its new (and in the words of one YouTube spokesperson, “100-percent homegrown”) streaming infrastructure has been a disappointment, especially in light of all the data center capacity, transit links and engineering resources Google has to throw at the effort. Video quality was poor and quite choppy in our test of the service, and it seems we weren’t alone in this respect: VideoNuze, Gizmodo and Wired all mentioned similar issues with the video streaming.
Even more disappointing is that these issues sprang up while not that many people were watching. I found the video to be choppy with viewership hovering in the low hundreds, which hasn’t been unusual for this early test. TechCrunch counted less than 500 viewers online for a live interview with Tony Hawk, and many of the other live shows have garnered less than 1,000 viewers.
This goes to show it’s not easy to do live video, and even harder to do live video well. If YouTube plans to expand its live video capabilities further, it will need to drastically improve on its first real effort in the live video space.
No surprises there really considering it was an alpha test.
I’d say the biggest problem that YouTube needs to worry about — once they figure out the internal technical issues — will be the awful quality of the content itself. From what little I saw, it looked like public access amateur hour. If YouTube’s plan is to seep into the mainstream with their live broadcasting (which will undoubtedly happen within the next few years) not only will they have to “drastically improve” it’s technical infrastructure, but I’d say they’re also going to have to give their so-called “content partners” a lesson in how to put on a live show. There’s no way in hell they’ll be able to garner an audience larger than that of a public access channel if they’re not able to shed the high school AV club aesthetic of no-budget TV.