Livestation success mean lower user experience
- Image via Wikipedia
As reported by Matteo Berlucchi, Livestation has been forced to do an heart breaking decision. The recent war in Gaza has increased Livestation usage so much that it became essential that Livestation reduce the quality of it’s video to limit bandwidth consumption/cost.
The service was originally advertised as peer-2-peer but in actuallity the service is provided solely from corporate internet video access. The peer-2-peer functionality is not yet present in the client software.
Given that Joost moved away from it’s peer-2-peer solution it is hard to beleive that Livestation will get there successfully any time soon. It is much easier to provide the video streams from a controlled access and ensure acceptable user experience. With peer-2-peer it is much more difficult to offer realtime video streaming in a consistant manner due to the uncontroled nature of each participants internet connection.
Hopefully Livestation will get enough funding from advertisement to remain viable in the long run. It would be sad to see such a great service close due to it’s own success and lack of viable business model.
You will find below a copy of the original post:
Dear Livestation viewers,
in the past 2 weeks we have experienced an enormous growth in traffic because of the dramatic situation in Gaza. With a lot of people turning to the internet to find more information about the war, our traffic has consequently grown a lot and very quickly.
As the service is free and we still don’t make much money with advertising (we are 4 months old and if you like to advertise or sponsor a channel please do get in touch team@livestation.com) we had to slightly reduce the streaming quality. The change is visible but the watching and listening experience should be pretty much the same as it was before.
We apologise for this unplanned quality reduction but our aim is to make sure the highest possible number of people can access our partner channels from anywhere and at anytime.
Please do send us your feedback, tell your friends and keep watching.
![Reblog this post [with Zemanta]](http://img.zemanta.com/reblog_e.png?x-id=6c2b4d28-1e0b-4909-bc52-82a945d8fec9)
Bernard,
let me shed more light on this delicate topic.
You are right in saying that our P2P engine is not deployed at the moment. I thank you for the opportunity to explain some of the things going on in the background.
We acquired the original body of work from Microsoft Research in June 2006. We then spent almost two years developing it into a product. We deployed the P2P code in April 08 in the version we announced at NAB. We run that code base up to version 1.0.77 in the Windows version only because it was written in a programming language that only worked on MS Windows (C#). At the same time we started rewriting the P2P code in a language that could allow us to install on all platforms (C++).
Where are we now? Well, the new x-platform P2P code is almost done and it’s a complete rewrite of the code we run for a few months in 2008. We learnt a lot during those months. In particular, we learnt that our P2P has nothing to do with the P2P used for on demand services – think of waterskiing and windsurfing, some element, different sport! In a live environment, for example, peering works even with a handful of users as opposed to being a function of the number of peers. We can get very high peering efficiency in particular as we lower the video quality to around 500kbps. We can also support locality in the peering architecture so that people on the same ISP automatically peer with each other.
If you combine this with the fact that live streaming costs always more than on demand and that the problem faced by live is “concurrency” rather than “caching” you conclude that P2P is a must have rather than a nice to have if you are planning to become the destination for live streaming.
Unlike Joost and Hulu, our service requires a P2P desktop client (or a browser plug-in) in order to scale. Our plan is very simple: we want people to be able to watch on our site if the can’t or don’t want to install the player (for example at work). In this case we will offer a lower quality version of the streams with advertising. We will offer a plug-in to allow viewers to watch in the browser at higher quality thanks to the P2P technology as soon as feasible. Those who enjoy having a desktop app dedicated to their streaming/live TV services with Live Chat, Live Panel, Live Alerts, Always on Top, favourites, etc can carry on using what we have.
I think that the argument against downloads is wrongly put: of course no download is better than having to download but the point is purely around value: if the value you get from a download is sufficient to overcome the hurdle, than it’s no longer a problem. Take Skype, IM, etc.
Our sophisticated P2P will enable us to offer a free high quality service at very low cost to us, hence letting our viewers enjoy Livestation free of charge.
Thank Matteo for this enlightening post! I guess Livestation will certainly deploy P2P streaming in a near future given the current situation.
It is great to see that Livestation will move to a model similar to Hulu/Joost by allowing users to watch live content directly on your web site while maintaining a dedicated client for the ones who prefer this method.
Looking forward for this new client/web site.