Joi Ito's Web

Joi Ito's conversation with the living web.

As the thought of paying $3500 for a month of gprs sinks in and I think about the speech I'm going to give at MILIA to the carriers and content providers in the audience, I'm thinking more and more about how I think it might be a bad idea for the carriers to get into the content business.

I think that as broadband becomes a standard part of households, more and more people will fill up their iPods and mobile devices with all the content they need from their flat-fee low-cost pipe. Most content isn't THAT time sensitive. I don't see any reason to have to download content on-the-go over expensive gprs when devices can talk wifi or bluetooth and have enough storage to allow you to carry content around.

The main value that always-on provides is presence information, short messages and time sensitive stuff like news. I don't really see the need to have broadband to do that. I think the carriers should focus their energies on stuff like identity, payment systems, IM and presence and leave the content business up to people who know how to move large volumes of bits around at low cost. The problem with most telephone companies is that they have spent their whole lives worrying about quality of service, but moving large volumes of data around is not about quality of service. You can afford to drop a few bits if they're not time sensitive and it's a completely different game than the circuit business.

I realize that 3G networks are supposed to provide us with a cheaper way to provide mobile broadband, but I just can't imagine the cost of all of the roaming deals, the metering systems and the BigCo overhead ever being able to compete with the simplicity of the Internet and wifi. I am not convinced that there is a market for broadband mobile content.

This may seem obvious to Internet folks, but I think the mobile operators are seriously considering broadband content over mobile phone networks as "the next big thing".


in the middle run you are certainly right. and also it is not only true, that the carriers are moving into mobile broadband but also into producing and licensing content itself. that is most certainly thw riong way to go as i agree with what you said. but then again, how about p2p sharing of content via embedded 3g chips. how about sharing your playlists with your friends directly via your ipod or sharing photos via your digital camera. p2p behaviour and mesh networks will of cource impose a challenge in regard of trust systems, identity and presence. so that actually underlines your point on what the telcos should concentrate. i think their misperception at this point is, thinking they will have to fill up the bandwith with top-down imposed, paid content (like soccer streams, porn, etc.) in order to re-finance 3g. that won´t work but certainly the bandwith will be used up, once it´s there. so i guess bottom line is, they should concentrate on providing a good framework and attractive pricing schemes in order to foster usage. the road they´re traveling now, will lead some carriers straight intro bancrupcy.

Joi, what about commercials? While harddiscrecorders are beginning to flow on the TV market (they can leave out the commercials while they are recording), marketers have to find new ways to reach their receipients: I think mobile sreaming-commercials will grow with the bandwith you will be able to use. Greets, fred ;-)

Follow the storage performance/size/capacity/price trendline out 10 or 15 years, and we get to the hard-drive that John Gilmore's been describing: the size of a sugar-cube, costing $1-2, powered off of the occassional hard shake, and capacious enough to store every word ever uttered, every song ever sung, every painting ever painted, every movie ever shot, and every word ever written, at a resolution that can be maginfied into the microscale without distortion.

I agree that carriers have their heads up their asses, but I also think the promise of ubiquitous wireless broadband is just too compelling to pass up.

There's something approaching a facsimile of this in American cities thanks to the Starbucks hegemony and its resultant web of WiFi access, and frankly I can't get enough.

Just because we don't NEED it, doesn't mean it's not a good idea.

And don't forget that some of the things you mention in your list of what we should leave to the carriers (presence, IM...) may be significantly more compelling when we throw hi-res ubiquitous imaging devices (cameraphones!) into the mix.

John, but I think my point is questioning whether it will be the "carriers" who provide these networks and whether 3G is the right protocol. I want cheap, flat rate, wireless broadband. For that you want to keep it simple. I may be wrong. 3G may be cheap enough and may provide some mid-level value between the islands of wifi, but the mobile operators are going to really have to get their act in gear and figure out what the value proposition is... and it's NOT charging $3500 for reading my email in Europe. I do agree that photos may be a good use. I also agree with Felix that sharing is the key. I want real time more often when I'm posting to the Net than when I'm retrieving...

We all know that it's extremely difficult to get fixed-line broadband customers -- e.g. DSL subscribers -- to accept non flat-rate pricing models.
Mobile operators who move to IP-centric 3G backbones, OTOH, have the ability to leverage the unique *usage-based* billing relationship they have already established with their customers.

Applications of such usage-based billings could include on-line music shops à la Apple iTunes Music Store, if the per-bit costs on the 3G network can be brought low enough to make economically possible e.g. a US$0.99 charge per 3-minute 128Kbps AAC-encoded song. Note that this would mean charging US$0.99 for the transfer of about 3Mbytes of data, which would be orders of magnitude lower than what wireless providers charge on a GPRS infrastructure today.

Another possible, and possibly more realistic than a music download service, application would be for the operator to act as a payment intermediary between content providers -- e.g. newspaper web sites, popular blogs... -- and the mobile users who access these sites.
XML tags should indicate the content's public -- i.e. end-user visible -- list price in JPY, EUR, USD, GBP, CHF etc., as well as the channel the operator should use to pay the content provider -- e.g. PayPal e-mail address ;-) VISA card to be credited, IBAN & SWIFT BIC, ABA sorting&routing numbers...
If the desired payment channel is acceptable, the operator's network gateway/proxy communicates the chosen currency, the applicable taxes and commission it will withhold from the list price, and the operator's payment terms.
If the total amount withheld and payment terms are accepted by the content provider's automated business rules, the information transfer session can proceed. To prevent future disagreements the whole exchange leaves a non-repudiable audit trail on both sides, signed by each party's public certificates.

The benefits for operators to leverage their position as payment collectors -- and get rewarded for the credit risk they shoulder -- are numerous, and include:

* the fact that it's better, and more scalable, to let the market itself -- i.e. the information suppliers and the mobile consumers -- direct in an emerging fashion the mobile data market's segmentation and possibly novel applications -- e.g. messaging, entertainment, information, file transfer, e-commerce, P2P... -- as well as its associated pricing. The operators should not presume to divine their customer needs' segmentation and the value and billings associated with it.

* the content providers get to monetize their content from a willing user base with whom they don't have any direct billing relationship

* for the mobile users, a particular content's pricing is communicated in a clear fashion, and the billing mechanism and relationship is pre-existing and thus simple, as well as better-known and more trusted than what an unknown Internet web site might offer.

I should probably include a whole draft RFC about the application-layer PKI- and XML-based interactions taking place between the content provider's server and the mobile operator's Internet gateway, and include discussions about tunneling GPRS packets from roaming networks via GPRS Routing Exchanges to make available some preferred pricing agreements an operator might establish with some popular content sites.
In some sense, the PKI-certified dynamic micro content download contracts expand to small web sites the financial flows which were hitherto limited only to operators signing formal roaming and revenue-sharing agreements between themselves. This comment field's margins are too narrow to contain the model's discussion ;-)

I think the idea behind flat fee is justified. However, with that said, I think the carriers will have to provide the content in order to pay for the network. It will be exactly like cable TV here in the states. Get a connection (cable line), pay monthly flat fee (only covers network costs), pay for certain channels (monthly subscription fee), pay more for premium content (pay per view). It makes sense for wireless carriers as well. But they need to figure out a way to provide a flat fee first. I don't pay more for my cable one month compared to another if I "use" the TV more. That is ridiculous. Sorry about your bill Joi. Good luck with that.

With their always-on connection would it not be possible for the 3G operators to offer a two tier payment system? When there is low network traffic they transfer larger, low priority, files at a low price. At other times, when immediate access is required for phone calls, instant messages or urgent files, you pay the full rate.

A few years back when NTT Docomo did their first FOMA demos, the news picture showed a busty bikini model holding one of the FOMA phones (pretty standard in Japan). My thought then and now is that the only marketable content for 3G phones would be porn. Unfortunately it seems the desingers of the phones forgot this because they failed to add a video out port on the phone. What salaryman would ever openly watch sukebe/ero video on his keitei during his 2 hour commute? If they had included a port for Glasstrons or some other model of over the eye video display, then customers might be willing to shell out the monthly costs to pay for video content on their 3G keitei.

Two things:

1. Chris_B asks "What salaryman would ever openly watch sukebe/ero video on his keitei during his 2 hour commute?" and I think it's the same guys who read sukebe/ero manga books in my face on the Den-en-toshi line.

2. Joi, your reply to my comment is right--I didn't really address your point. I do think the carriers are the right ones to provide ubiquitous bandwitdh--they have the infrastructure already (well, here at least). Blanketing Tokyo in WiFi is a nightmare proposition.

As for pricing, I think flat-rate is absolutely required, and thankfully AU bit the bullet and is forcing DoCoMo to do the same.

I'm not sure if the prices we're seeing are really a function of the cost of 3G or (myopic) opportunism on their part.

I think we need tiered flat-rate: different plans for different usage profiles so both the carriers and consumers stay whole.

But I wish the carriers would stop thinking they can be content providers or that they need to invent new content jo justify the bandwidth. Just give us a pipe to the internet, where there is lots of content already, and let's all build some tools to encourage people to create their own content each day with mobile devices.

If they're looking for other revenue streams, billing, identification, security, user content storage--these are all things that play to their strengths and enable the rest of the stuff to happen.

Yeah, I know I'm oversimplifying.

3G service is the final piece of the network that completes our dream of ubiquitous connectivity. A great thing! But until 3G service can hit the pricepoints of landline services it will be relegated to the network that fills the holes. The question that then needs to be asked is what content exists on the land line side that the 3G network can then support to make it globally available.

As Joi has mentioned above, that content will need to be something that is real time sensitive. To me that brings us back to why we all have cell phones... to communicate. Communication is the number 1 reason for having a 3G network. Fatter pipes allow us to communicate more affectively. By including presence and SIP (session inititation protocol) capabilities in the handset it allows for ubiquitous communications. I believe the greatest drive for content will be around voice and communications. As voice moves to the Internet (VoIP) you now have the capability to tie your aDSL, WiFi, and 3G networks together for something that really does need real time capabilities.

In my view, voice is just a form of content and will be the highest volume content driven over the 3G networks. Voice combined with things like presence, SIP and GPS will create unique and exciting voice services we haven't even dreamed of yet. As these new voice services evolve it will be interesting to see who ends up owning that "voice" minute.

What about WiMax. A 3G KIller?

4 TrackBacks

Listed below are links to blogs that reference this entry: Carriers should be context providers.

TrackBack URL for this entry:

The sound of market share from Oliver Thylmann's Blog
March 4, 2004 7:53 PM

The last three of Joi's post have hit the cord of wireless and warrant mentioning. I'll start from the top with some finishing comments. Sometimes you just read something fun, as this entry from Joi: When I was traveling in Read More

Joi Ito writes that carriers shouldn't be content providers: "I'm thinking more and more about how I think it might be a bad idea for the carriers to get into the content business [...] The main value that always-on provides is presence information, s... Read More

TITLE: WiFi NOT 3G will be the protocol for flat rate, cheap, ubiquitous wireless broadband? URL: IP: BLOG NAME: North American Bandwidth News DATE: 03/18/2004 06:35:35 AM Read More

Joi Ito wonders on the value that GPRS carriers can offer by providing content to users on the go. He notes that the pay-as-you-go route is best served for time- and presence-sensitive information services more so than the traditional corpus... Read More