The IP Development Network Blog

Joost is a hot topic. It deserves to be. There are many things about the service that aren't quite right, but underlying it all are three key ingredients for success: big-name backing, excellent PR and efficient, adaptable technology.

Of course they have issues. The project has spent less than a year in the public eye and there remain three other factors threatening the success of the venture: programming, devices and networks.

The programming is not right somehow - perhaps they need to consider adding "broadcast channels" because finding what you want from a static list seems to ensure the content gets stuck in time. The first time you go there, you find one or two things of interest, but when you go back you are left trawling somewhat. It makes the whole experience too predictable.

But, the backing is in place to fix these problems. The number of content producers is already impressive, even if their offering isn't. No-one is putting their best stuff up there yet - it's early days. Viacom own Dreamworks, who produce a large chunk of the most popular kids movies. If you have kids, what price an online archive of all Dreamworks films, on tap? Its not there, yet.

Before that happens, something needs to be done with the channel guide to give it a bit more life. It is surprising that there aren't more "social networking" buttons. If it were easy to rate content, you could be sure that there would be more turnover and better filtering of the good stuff. Make it less miss and miss.

This can all be done. A bit further down the line though, they are going to have to face up to the device question, which is to some degree, out of their hands. I don't know that Joost are betting on the PC being the dominant, always on device in the home, but if they are, then they may be swimming against the tide. Everything is a computer now, there are processors in everything. This will surely mean that specialisation wins over generalisation - TVs do video and HiFis do music. PCs do publishing. Multi-purpose devices are for mobile users who have a premium on weight.

But this is where the PR can help. Here, even the screw-ups are working for them. Colmmacc posted a link to a set of slides he had produced on the network architecture on my last post on that subject. I hope he didn't get into trouble for this because it turns out that hidden beneath the visible layer on the PDF was a whole host of info on Joost's 3 month strategy goals.

This leak, just serves to increase the buzz about potential partners and shows that the team knows what they are doing. An example is the combination of US Soccer and Real Madrid as launch partners, showing a clear appreciation of the David Beckham PR phenomenon that is about to hit the US. It has been suggested that this leak was a Machiavellian PR exercise, which is almost certainly not true, but it shows how highly regarded "The World's Hottest Start-up" is that people are willing to believe that this may be part of the plan.

I just hope that this doesn't cause the company to clam up, because a big part of what is making the PR work is the open approach of the team. Openness always makes you more vulnerable to leaks of "sensitive information", but it also allows you to get on with stuff and build partnerships. The reverse is also true - if you are afraid of giving the game away, you don't give partners enough to make them want to join you in your adventure.

The PR is the way that Joost can solve the device problem they face. Just as Skype became the must-have bolt on to cheap IP phones, it is quite possible that a Joost installation could become standard on the television itself, requiring just an ethernet or wireless uplink before it is ready to go. Of course, a set top box could do that job too in the interim period. At this point, the platform becomes a de-facto standard for online video distribution, with anyone and everyone as possible paying customers.

Joost: YouTube for Professionals.

I can't see Sky liking that very much, but it gives Virgin, the BBC, ITV, C4 and the other content owners and distributors a way around Sky's possible IPTV content / network monopoly.

Which brings me onto the network, which is growing on the back of version 0.10's For Friends offering. It seems that there is quite a bit more content added and a few tweaks here and there, but the programming remains as was. The main development is that there is a lot more advertising with discrete logo hyperlinks displayed throughout whatever you are watching.

The P2P hit rate that they seem to be getting is certainly improved over my earlier analysis, at least for the popular items. Level3 seems to be taking full responsibility now for the seed network, there is little or nothing now from the Infonet Luxembourg centre that there was a few weeks back.

On the "Most Popular" clip of dudes surfing in Hawaii, only 18% of the traffic came from Level3, and there was a broad distribution of 8 peers who between them delivered a further 74% of traffic. 66 other peers gave me the remaining 8% of bytes. When I did the same, while watching Rockie + Bullwinkle (not the most popular...), I found that there was only one peer hit, delivering 15% of bytes, but leaving 81% to the Level3 network. Only 4% of bytes came from the remaining 56 peers in that sample.

There is no crunch point yet, but if traffic does grow (still an if, given the programming issues), changes need to be made to the way in which Joost works with the networks to route the packets more efficiently. The huge amount of tromboning that currently takes place needs to be addressed by the networks as well as by Joost (this is a link to colmmacc's slides minus the "secrets").

Joost's card in this gambit is its efficient, adaptable technology. Used in the right way within a network, it offers an extremely powerful publishing platform offering the ISPs using it a potential differential advantage. That level of integration is very hard to do however (for the ISP more than for Joost, I suspect) for an ISP fighting a price war. Joost may have to wait a little before going down this road.

Alternatively, they could launch a guerrilla war against the networks. Of course they'll say that they won't, but it might not hurt the service providers to think that a possibility. Joost could entirely disintermediate the network from the video value chain.

But like I say, all this talk of global domination is a few years away, and it will not happen until Joost addresses the core programming and makes the service something you watch, rather than something you play with (as you do today). Toys get boring after a while.

Labels: IPTV advertising, Joost, next generation networks, video on demand

# posted by Jeremy Penston @ 5/25/2007 12:28:00 PM

Friday, 25 May 2007

Joost about on track