IBC: It's not all about content
Editors' note: This is a post based on an article in this week's edition of Ariel, the 91Èȱ¬'s in-house weekly magazine, by Audio & Music Interactive's Head Of Distribution Technology John Ousby. It includes John's images from the , as blogged at .
Dr , the founder of , was unfortunately upstaged at his own keynote address at IBC in Amsterdam last Friday.
Google and YouTube are parasites. It's all about content; the rest is just railway lines.
This was the message given by ITV boss in his recorded interview included in the session. A few people in the audience started clapping until they realised they were outnumbered by the growing army of raised eyebrows.
Were we in a conference from the late '90s? Or did this have anything to do with the fact that ITV is later this month?
The idea that content can be easily separated from technology and distribution is plain wrong. One has been informing the other since the start of broadcasting, through the production technology available at the time or the way audiences find, share, discuss and consume it.
The biggest mistakes I have seen in the broadcast world are when interactivity is slapped on once the paint is dry in the production process or when a technology application is created without consideration of the audience it is intended for.
DAB slideshow showing some of the output from the olympics twitter feed from 5live
You can't spend much time at IBC without hearing about convergence and the ever-redrawn battle lines between content owners, broadcasters, internet service providers and telcos.
was around mobile TV [], which on the broadcast side hasn't had the most prestigious start after several years of hype, trials and struggling commercial services.
Mobile operators have struggled with small volumes of low quality video clips in walled gardens that are expensive to consume and unreliable in reception. With mobile services, context is everything - not just the web (or telly) bundled across to a smaller screen, but content which takes account of where and how it is consumed, and by whom.
You could draw the conclusion that video on the move just isn't as important as was thought. I believe it's just a question of when.
We are in a transitionary phase where we are just starting to see the possibilities for mobile video once it's made easy to consume and the pricing structure is relatively understood, as with the iPhone.
Let's just start to think about mobile video and audio, of which TV is a subset rather than a starting point - both broadcast- and internet-delivered video have a part to play in the future of mobile TV.
looks like anything else on display at IBC until you understand what it's doing. Live p2p video streaming based on the tribler infrastructure - a potential solution to iPlayer success... Great project involving 91Èȱ¬'s George Wright, and the among others. Of course, not just video can use this. Nice work.
Walking the halls at IBC proffered the usual mix of landmark moments, promising new technology and the next biggest, brightest display screens, some of which can be seen in .
It made me proud to see that the 91Èȱ¬ were involved in a lot of the best of show - DVB-T2 (next generation digital terrestrial TV delivery), Super High Vision (HD on steroids) and p2p-next (live peer to peer streaming), to name a few.
Let's just hope that next year can see a keynote fit for 2009 not 1999.
DVB-T2 - rotated constellations (256 QAM): 91Èȱ¬ has been leading the work of the DVB group in its next generation DVB-T work. DVB-T2 gives about 50% extra capacity than DVB-T and will be essential for Freeview HDTV services - currently planned for the end of 2009. More detail here:
dvb-t2
John Ousby is head of distribution technologies, Audio & Music Interactive.
Comment number 1.
At 17th Sep 2008, tommyneuf wrote:Grade of course is right, but he misses the point. There used to be five railways tracks, one lead to ITV - now there are millions, and they all go different ways.
Complain about this comment (Comment number 1)
Comment number 2.
At 17th Sep 2008, digitalradiotech wrote:"Live p2p video streaming based on the tribler infrastructure - a potential solution to iPlayer success"
Er, P2P-next is a P2P networking project for distributing *live* streams, whereas the iPlayer consists of on-demand streams, which by definition are not live, so P2P-next cannot carry iPlayer TV streams.
And P2P live streaming doesn't even save any bandwidth in comparison to unicast live streams anyway. Everybody that receives a live stream has to receive the data in that stream, so whether the user receives the data from 91Èȱ¬ servers or from other users on a P2P network doesn't reduce the amount of data a user needs to download.
P2P streaming offloads the 91Èȱ¬'s bandwidth onto users, but the problem is the amount of bandwidth the ISPs have to carry, so it won't improve that situation at all.
P2P-next is a bit of a waste of time in terms of its usefulness for streaming (other than for academic purposes), because well before the project will even finish multicast will be widely supported by the big ISPs, and it's impossible to invent a more efficient distribution technology for live Internet streams than multicast (multicast only requires one stream to be carried on any Internet link, so you obviously can't get more efficient than that).
Complain about this comment (Comment number 2)
Comment number 3.
At 17th Sep 2008, hackerjack wrote:Er, P2P-next is a P2P networking project for distributing *live* streams, whereas the iPlayer consists of on-demand streams, which by definition are not live, so P2P-next cannot carry iPlayer TV streams.
-------
The point is that it's a technology that could be applied with small modifications to iPlayer.
As I understand it as it stands P2P-next is designed for say 20 people to watch an identical stream, but instead of all 20 needing to get it direct from the server, each user is delayed a little (to keep in synch) while a little of the stream is bufferred and P2P shared.
Is it really that far a departure to enable this with overlapping start times ? You join an on-demand stream and immediately download sectiosn from others who have already watched it and are still connected to the stream. There is no need for concurrency. Admittedly this will require clients to cache much larger segments of the stream but in essence it could work.
Complain about this comment (Comment number 3)
Comment number 4.
At 18th Sep 2008, digitalradiotech wrote:"The point is that it's a technology that could be applied with small modifications to iPlayer."
My point was that you can't use P2P-next for the iPlayer TV streams at all, because they're on-demand, and P2P-next is for live streams.
"As I understand it as it stands P2P-next is designed for say 20 people to watch an identical stream,
Exactly - when people watch live streams they're watching identical streams to everyone else.
"but instead of all 20 needing to get it direct from the server, each user is delayed a little (to keep in synch) while a little of the stream is bufferred and P2P shared."
Users serve data to other users, so if X serves data to Y, there has to be a delay to allow data to travel from X to Y.
"Is it really that far a departure to enable this with overlapping start times ? You join an on-demand stream and immediately download sectiosn from others who have already watched it and are still connected to the stream. There is no need for concurrency."
But what's the benefit of doing that? It wouldn't save any bandwidth, because everyone has to download all the data that they end up watching no matter whether the data comes from 91Èȱ¬ servers or from other users. In fact, people would just end up downloading data for parts of programmes that they might not end up watching, so it would waste data.
Complain about this comment (Comment number 4)
Comment number 5.
At 18th Sep 2008, johnousby wrote:digitalradiotech - "Er, P2P-next is a P2P networking project for distributing *live* streams, whereas the iPlayer consists of on-demand streams"
er, i think you'll find that several tv channels and radio stations have live streams within the iplayer.
of course multicast is a potential solution for live streaming. P2P next was an interesting first public demonstration of another possible solution. still very early days, as you would expect of a first demonstration.
Complain about this comment (Comment number 5)
Comment number 6.
At 18th Sep 2008, digitalradiotech wrote:John, you said P2P-next is "a potential solution to iPlayer success". But it doesn't save any bandwidth for live streaming compared to using ordinary unicast streams, so what problem does it solve for the live streams?
Anyway, good to hear that the iPlayer is going to incorporate automatic bandwidth detection:
"Soon, the iPlayer will automatically detect a PC's broadband speed and serve up either higher or lower bitrate streams to ensure the smoothest viewing, Rose said."
Presumably you'll be using that as well so that you can provide the live radio streams at higher quality?
Complain about this comment (Comment number 6)
Comment number 7.
At 13th Jun 2009, U14033173 wrote:This comment was removed because the moderators found it broke the house rules. Explain.
Complain about this comment (Comment number 7)
Comment number 8.
At 9th Sep 2009, felicioo wrote:This comment was removed because the moderators found it broke the house rules. Explain.
Complain about this comment (Comment number 8)
Comment number 9.
At 13th Sep 2009, U14134634 wrote:This comment was removed because the moderators found it broke the house rules. Explain.
Complain about this comment (Comment number 9)