Some of James' Ideas

As cameras get better and better an interesting thing is happening - the creative exercise of generating an amazing image is increasingly being decoupled from the physical act of holding and operating the camera.

Used to be you had to get things like white balance, framing and the aperture setting right when you were shooting. Now we have RAW files, high resolutions and HDR modes so that you can select the colours, the frame and the aperture after the fact.

Chipmakers are already starting to hit a resolution wall (who cares how many megapixels you have? How many poster sized photos do we ever print out?). As a result they will have to look for other ways to add value. I think that this will be in things like automatic geotagging - so that the camera knows exactly where it is on the planet and which way it is pointing. Another direction will be in cameras that can take photos at a very high speed. Yet another will be cameras that can shoot at different focal lengths and calculate how far objects are from them.

Once these features are combined - we will have a device that just needs to be waved around at the subject. Intelligent software will then stitch the hundreds of images together into a 'cloud' of data - a mini virtual world. Photographers will be able to zoom around inside these virtual worlds at their leisure - pulling out of the cloud a photo that has the desired framing, colour, focal length etc.

I don't think this will render the skills of photography meaningless - but it will change their emphasis. Being able to get just the right aperture or focal length will not be as important. Finding the right subjects, getting the right light, being able to compose an attractive image - these things will continue to differentiate between real photographers and the rest of us.

Photography Blog

If it is true that broadcast networks are becoming just another set of pipes for content, then it follows that they could very well lose their traditional centrality in determining what content gets made. Advertisers will start to want to buy content directly or to make their own. Producers will want a way to tap advertising dollars.

In this environment it seems to me that there is an opportunity to create a marketplace for advertisers to deal directly with producers.Obviously these relationships can and will occur anyway. However just as there is a difference between having a garage sale and using Ebay, there would be an advantage to aggregating those who want to produce content and those who want to buy content.

The ideal would combine the auction/payment mechanics of ebay or adwords with advanced metadata customised to fit the content industry. Advertisers would be able to use this metadata to match up content with what they are trying to sell. Producers could look for opportunities to marry a project to a certain symbiotic brand.

As mentioned below, the system would not lock any party into any preset distribution platform or advertising technique. The producers of LOST could sell a watermark in the first episode to one company, top and tail to another, product placement to a third, then distribute it with the Sunday paper on DVD. The second episode could have a different mix of advertising and be an internet download. These decisions may be strategic blunders - the point is that the system itself would need to be flexible enough to let the market work out how best to proceed.

For the advertisers the advantages would be:

1. Cost - no need to pay huge premiums to the networks. International competition among producers who want to build a show/project for you.
2. Creativity and flexibility - talk directly to the producers about how the advertising should be incorporated. Alternately, use powerful search tools to wade through the thousands of offerings by demographic, popularity, production team, to find the right match for your product.
3. Data - know exactly who is watching your shows, how many times and what other things they like.

1. Funding - advertisers looking to get creative means more projects or more opportunity for support of something you're already doing.
2. Control - give up as much or as little control to advertisers as you want.
3. Access - at present funding is hard to get and controlled by a coterie of 'seasoned' producers. If everyone is on a level playing field the best idea and best team will win.

If my last two posts are on the money, it follows that some real opportunities exist in this space. For one thing, there should be an expansion of content production as advertisers wake up to the fact that they can make entire shows for the cost of 30sec ads. But that's just production - what excites me is the need for a marketplace. Traditionally delegates from channels have gone to XXXX and made bids for the rights to shows under YYYY. If the aforementioned changes continue, this sheltered world is about to be blown right open.

What will be needed is a marketplace where disparate companies can come together and bid for the rights to content. There needs to be a mechanism that brings producers together with advertisers. The idea is for advertisers to be able to make offers right alongside the networks both for shows and for 'pilots' or ideas for shows. So Myer could buy Housewives, Coke could bid on the pilot of the next show from the creator of the OC. The market would work both ways - Ferrari could put up its need for a car enthusiast show, and invite tenders/pilots from seasoned producers.

Such a system would need to embrace regional controls - not to create unwanted distortions but to increase efficiency - there is no point in a purely regional player like Myer buying eyeballs in the US. However, the marketplace would be agnostic about the underlying technology of distribution - if an advertiser wanted to distribute over the internet, DVD, flash drives, mobile phones, or even by buying time on a network, this would become a part of the negotiation.

A lot needs to happen for this to become a reality. For one thing, BitTorrent-type downloading needs to become much simpler. iTunes is showing how easy the interface should be for the end user - if it is anything harder than pressing 'subscribe' or 'download' it is too hard. The living room also needs to be colonised. It needs to be as easy to sit down in front of a Myer sponsored-internet downloaded show as it is to watch the footy. This is because the value proposition will not be as good for advertisers until all content is on an equal footing. Once these obstacles are overcome, piracy will cease to be as much of a concern (at least for advertiser supported content) because the shows will be free and much more convenient to access.

As technology provides new platforms for video distribution, advertisers should start to ask the same question - why does sending my data over these (broadcast) pipes cost so much more than these other (internet, DVD, etc) pipes? In a way, channels 7,9, and 10 are a lot like the mobile phone operators, in that their market position allows them to charge premiums (in this case for advertising) because up until now they have been the only game in town. New platforms will allow advertisers to think strategically about where they put their money:

$XXXXX = the Australian broadcasting rights to a season of a prime time show.
$XXXXXXX = how much the show brings in from advertising per season.
$XXXXXXX = Unilever advertising spend (per year).

New strategies may involve buying the rights to a show and distributing it yourself. Myer could buy the Australian rights to Desperate Housewives and distribute the show over the internet with just a Myer watermark and maybe a quick 'top and tail'. (See "Piracy is Good?" video) Johnson and Johnson (Clearasil) could buy Home and Away, chop it into segments and distribute it over mobile phones.

Alternately, advertisers may want to build their own show instead of an ad. Ferrari could build their own informational car show to appeal to enthusiasts.

These may be terrible ideas, but they illustrate the point that the market for content will go from being a simple linear chain in which channels buy shows and sell advertising space to a multi polar world in which different companies may invest in content for a multitude of different reasons and at varying price points.

It seems beyond obvious that many of the dynamics that have plagued the music industry in the transition from CDs to digital downloads are driving changes in the market for video. A lot has been made of technologies like BitTorrent which allow/require anyone to be a distributor as well as a consumer of content. At the very least, it is true that the technologies needed to produce content have become much cheaper. I tend to think video distribution still has some interface issues to overcome but certainly BitTorrent and H.264 video codecs show that the underlying technologies for cheap distribution exist. This is especially true if the distribution takes place 'in network' as service providers pay a lot more when data crosses over internet backbones.

As a result of these changes it is becoming rapidly apparent that advertising on broadcast television is overpriced. In a way, the situation is analogous to the price of an SMS message. SMS, which is just text, has less functionality than email and has got to be the most expensive form of data transfer on the planet.

SMS = $0.20
Email = $0.0001 (12GB = $70)

The premium probably stems from a lot of factors but at least one must be the small number of mobile phone carriers and networks. Within their walled garden, it made sense to charge as much as people would pay. However as users become more network savvy, it makes sense to ask the question - why does sending my data over these particular pipes cost so much more than these other pipes?

I was listening to an interview with Mark Zuckerberg where he mentioned the idea that Facebook was not creating anything so much as 'revealing' what he called the 'social graph' - a way of representing how everyone on the planet knows everyone else.

It seemed to me that there are a few other graphs which could also be mapped by a web service. Once mapped these graphs could yield value to advertisers and to end users. One such graph is the 'trust' graph - who trusts who to make recommendations (see post). Another is the geneological graph - who is related to who.

Facebook gives you the option to reveal how you know someone. In theory, this should map out everyone you are related to as well. However there are a few drawbacks - the person has to be alive, on facebook, and everyone in the process must take it seriously. At the moment, I think my brother and I "met randomly in a freak paragliding accident".

Genebook would address these problems. The service could be a facebook application, or an entirely new site. You would start by creating a page for everyone in your family, going back as far as possible. You could add media and 'tag' it so that any photos of your long dead great Aunt would appear on her individual page. Of course Genebook would need to be able to have a 'family tree' view as well as a page view.

Without having done any research into current offerings in this market, it feels safe to assume that this much functionality is par for the course. The advantage of Genebook would be in aggregating the memories and media of the entire living family. If my cousin was to start using the service it would first confirm if they were in fact my cousin, then give allow them to add media or create pages for anyone that they were related to. This way, the living members of the family could be adding the various photos, sound recordings, newspaper clippings and oral history of the clan to create a whole that would be much more than the sum of the parts.

Of course there would be clashes and difficulties. Genebook would need to give users the ability to keep a family member link, or piece of media private or exclusive to certain other family members. Discrepancies need not be a problem in that the system could present differrent versions of the same page to different viewers - thus it would be very easy to agree to disagree. If a synthesis was desired, it could be handled in a very wikipedia way with the ability to maintain a sideline discussion.

Apple - Apple TV - Guided Tour - Large

There are now lots of online services that are attempting to replace the video store. The services are reportedly plagued by problems in quality, high bandwidth requirements and a lack of the special features now found on most DVDs. To me the biggest drawbacks are in convenience - long form feature films call for a living room and a big screen - locking the content to a certain computer is a recipe for obscurity.

Apple now has AppleTV (take2) which aims to allow users to download directly from the living room with a remote. The downloads are progressive which should help with quality and cut down on any latency issues. The film can be taken along and viewed on an iPod or iPhone.

These all seem like steps in the right direction in terms of convenience. The problem is in the DRM. The user has 30 days to start the film, but once started, only 24 hours to finish watching. After this time, the film is locked. This is applying a 90s mentality to a 00s service. Netflix has pioneered a new way of renting films in which you retain (say) 3 films for as long as you want. When you want a new film, you have to send one of the old ones back.

It seems to me that this would be the best way to rent movies online. It would enable users to set their own viewing habits. The beauty of the Fairplay model of DRM for iTunes music purchases was that it provided enough freedom that most users did not even know the system was locked. Movie rentals should be the same.

The fact that Net Neutrality has become an issue is itself quite interesting. It seems like it could be the first real attempt of the technology companies to enlist public support to scare off prejudicial changes to their regulatory environment. I'm not sure what I'm basing that on, or whether it is right but a hint that I'm on the right track is that no one is talking about solutions. At some point someone will need to build more capacity -  hence the intractable debate - but something can be done in the meantime to make the internet more efficient.

One of the main issues is 'network shaping'. Using this process, an ISP or teleco (often the same entity) would be able to detect certain types of network traffic (bittorrent, youtube video etc) and slow those bits down while giving priority to other bits which carry their own proprietary content. Some reports suggest this is already happening to varying degrees.

The rationale for such behaviour is that there is not enough room on internet backbones for everything. New services are very bandwidth hungry and are chewing up all the space - so it makes sense to build some intelligence into the network so that it can discriminate between bits and everyone still gets their email.

The problem with this, is that it goes against one of the founding principles of the internet - the idea that the network should be dumb, and that innovation should happen at the edges. One of the reasons that there has been an explosion of new businesses and services on the internet is due to the fact that anyone can build a website, plug it in and it will work - there is no need to ask permission from anyone. Contrast this to the dearth of new services on mobile phone networks where a new service would need to be enabled by a central authority.

So, assuming this bandwidth crisis is real - what can be done?

1. Internet pricing plans should be changed to reflect the realities of how expensive bits are to get to the user. Stop calling plans 'unlimited' when they are not. If it costs more at certain times of the day, have that reflected in the pricing. If it costs more to send big chunks of information very quickly, allow the pricing to discriminate between this type of use and slower transfers.

2. Give users more control over how they use their own bandwidth. A lot of the services that the telcos are complaining about do not need to be sent at peak usage times in big fast chunks. An RSS feed that is drawing down a video podcast every day could easily be programmed to hit the server at a specific 'off peak' time or to download the file over the course of 10 hours rather than 10 minutes. The end result to the user is the same - they get the file in time to go to work the next day - but the changes to the network if this were adopted en-masse would be huge.

This is some amazing photography - I would love to see a behind the scenes video of how he lights these scenes and whether there is a lot of post production work involved...

Dave Hill Photo

I think:

There are some really innovative elements. In particular the fact that it is hooked up to a cellular network so you always have access to download a book but that you DON'T pay a subscription model for the cellular connection. In a way it points to the proliferation of great new services and ideas that could eventuate if the network providers were forced to go 'open' and allow any device to run on their networks.

It is ugly. People have praised the great interface, but to me it looks ridiculously clunky and inelegant. The idea is that you conserve power by making the big screen non-interactive, but that is such a step backward in UI. The keyboard in particular is an odd choice - it takes up so much space in something that is meant to be a 'book' replacement. "Touch" would be a much better fit - slide your finger across the screen to 'flip' to the next page.

The premise of being able to carry all your books with you at once is cool. For a minority of the market it will be cool enough to require them to carry a dedicated device. Possibly a bigger opportunity lies in increasing the screen size, resolution etc of an iPhone type device to the point where e-books become easily readable. That way you're not carrying an extra device, it comes with built in access to the internet (and can thus purchase new books on the fly). The drawback may be that it needs to be charged more often, but you're used to charging your phone anyway.