The IP Development Network Blog

Am I looking in the wrong place? I wrote earlier this week that although telco platforms might make 3rd party services richer, there is a problem because every feature that a telco might enable is already being copied by application platforms that run over the top. On reflection, I wonder if what telcos really bring to the party is not actually richer functionality but is in fact cheaper and/or safer functionality.

The Telco Platform
The initial piece was a lead up to the Telco 2.0 conference on 16th - 18th October in London. Telco 2.0 is based around the principle of the telco as a platform and although the idea was well ahead of its time when it was conceived a few years back, much of that advantage has been lost as internet applications have evolved. Meanwhile, telcos have done what telcos do best - delayed.

As much as I agree with the ideals of connectivity being supported by a sustainable economic model, I can be practical verging on cynical sometimes. I have asked myself on numerous occasions what exactly telco platforms bring to the party? Why would an application developer want to wade through treacle to integrate their applications with one, let alone several hundred slightly different telco platforms?

Facebook didn't need to.

I just don't see any functionality gains that make it worthwhile for an application developer to go through this pain - go on, prove me wrong (you know you want to). The only reason for working with the telco would be to get around obstacles that might be in the way.

What might these obstacles be?
These obstacles fall into two camps: economic and ethical/legal. Economic obstacles are simple to understand - the application is too expensive (across the entire delivery chain) when delivered over an open network. What I have termed ethical/legal is much more complex but essentially can be boiled down to what is right or wrong and the obligation of the delivery chain to minimise the latter.

Economic Obstacles
IPTV has economic obstacles. It needs to work inside telco platforms because the cost of internet distribution is forcing telcos to impede the open network model through traffic shaping. A painful death is waiting for the telco that doesn't manage traffic or reduce the cost of delivery by integrating the application into its platform.

Such integration is a win/win - although for the application provider it might also look like blackmail. The telco wins because it can eliminate its transit, core network and big chunks of backhaul cost making the delivery cost effective and the service performance excellent. The IPTV application provider wins because it can deliver its service; should they try without working inside the telco platform, they will be on the last train to nowhere. The alternative is MAD.

But perhaps proper IPTV (as distinct from internet video) is a special economic case because the file sizes are so far in excess of what other applications generate? That might be the only example where there is an economic imperative for the telco to build obstacles to traffic delivery from the open network and where the threat of losing customers to other telcos using the open model can be ignored.

Ethical/Legal Obstacles
The internet's biggest strength (its openness) is also its biggest weakness. Whether we as industry insiders like it or not, there is a dark side to our creation that allows people to do unspeakable things because they can do so without being traced.

For many years, telcos have fine tuned a position which says that they are not the police and that they have no responsibility for the traffic carried on their network. For hosting companies with servers that store such data, this position has always been sailing close to the wind, but in most jurisdictions the principle that the ISPs responsibility ends with notice and takedown has been accepted. Again, this position is based on the principle that the bits and bytes are the customer's responsibility in the same way as the road network and the vehicle rental company are not responsible for truck bombs.

The problem is that once an application platform is provided that facilitates the building of networks whose purpose is evil, or where activities are only legal at a certain age limit, a far greater degree of responsibility is imposed on the platform provider. The obligation becomes proactive - instead of waiting for notification by the authorities before executing a takedown, the platform provider's responsibility also includes to police the use of its service.

I would liken running a platform to running a hardware store. In this case the owners have a responsibility to ensure that the sale of explosive chemicals is closely monitored and that their customers are of a certain age that legally allows them to buy certain products. If they do not make the necessary efforts to prevent such things, then they can be held liable if someone uses their products to build a truck bomb.

A real problem
Legitimate businesses like MySpace and Facebook face the biggest threats because their platforms are being used (by a minority) for evil and where another (much larger) set of content might be deemed to be inappropriate for minors. Om has written recently about how Facebook faces significant difficulties with law enforcement because their platform is being used in such a way. Let's be clear, it is not Facebook committing unspeakable crimes - they are not the ones doing the deeds - but they are held partially accountable because it is happening on their platform.

There is little policing and no audit trail for a platform sitting on the internet. The identity of the miscreant is hidden by the layers between the platform and the IP address allocation and there is no horizontal record of the user's activity. This can usually be patched together after the event, but this is shutting the door after the horse has bolted isn't it? Like having no coppers, just lawyers...

A real solution?
There is a liability because the application is providing a platform for the activity. I wonder if this is not a key area where the telco platform has a role to play? Without the telco's cooperation, the application provider has no clue whether John Smith really is who he says he is or whether he is a convicted paedophile using an assumed name.

Such a person may be allowed to walk the street, but may not be allowed within 250 yards of a school. Put that into an internet context he may be allowed to use the internet but he may not be allowed to use MySpace. But how on earth does MySpace know this? It may be that it needs to talk to the telco platform before allowing the user in.

Similarly, the telco platform could conceivably take responsibility for providing the digital proof of age card for the much less serious but much more common problem of under-age access to adult content. Parents can then be much more certain that their technically adept children are not going to be able to override their parental controls quite so easily.

What's in it for me?
What's in it for the application provider? This one is pretty simple - they get proof that they have made the necessary checks on their customers' right to use their service. With this should come immunity from liability when laws are broken, which should make sure that such a solution is popular among legitimate businesses.

What's in it for the telco? This might also be simple - if applications can remove legal liabilities by these checks, they may well quickly conclude that they can only work with providers who allow them to make such checks. If a telco doesn't provide the data, the customer is locked out.

Telcos are pretty good at record keeping - they have had to be throughout history because CDRs are a big part of their business. They know where a connection originates and who is using that connection. They bill the customer and already credit check - adding a proof of age and checking whether the user is a registered sex offender is not a big leap.

Privacy need not be an issue
What about privacy? It should be noted that such a model has a clear Chinese wall between the telco who has the data and the application who has the user. The only data that needs to be passed is two yes/no flags saying whether the user is allowed to access a given service. No names or otherwise need change hands unless the application also needs to positively identify, for example in a banking environment.

Summary
In the end this is just a couple of examples. But, it deals with real issues - a commercial one and one that every right minded person wants solved - and issues that can only be solved with links between various platforms.

There are many ways which I have not considered where the telco platform with its knowledge of exactly who and where the user is (as opposed to who and where they say they are) can be used to benefit the application and the industry as a whole. Until now I have been looking for new features or better features and I have ignored the much more basic elements - perhaps I have been looking in the wrong place? It may well be the boring stuff, the stuff that makes using an application cheaper and safer, where the telco platform is needed?

Labels: platforms, Telco 2.0

# posted by Jeremy Penston @ 10/05/2007 10:56:00 AM

Friday, 5 October 2007

Cheaper, Safer - not Richer