≡ Menu

The IMS Gulag

In No Bundle of Joy, the Washington Post looks at consumer reaction to bundled offerings from carriers.  Tellingly, consumers are rejecting the bundled offerings.  Beyond a simple discount, most bundles force compromises few consumers really want to make.  Furthermore, by shopping you can get a better deal.   Here in Canada, for instance, I buy satellite TV from Bell, use VoIP long distance via PhoneGnome, buy cellular from Rogers, and DSL-based internet from Telecom Ottawa.   Bell’s dish gives me a better television selection than cable.  VoIP saves me more money than Bell.  Cellular features on Rogers, combined with the selection of phones, beats Bell hands down.  And my internet is $15/month cheaper than Bell or Rogers, and more reliable to boot.  I saved money and got the features I wanted.

The consumer actually does want an integrated experience.  Talk to anyone and they will tell you that it makes no sense that they have separate identities, separate mailboxes, separate applications of all kinds, depending on which network they are attached to, and which network identifier (ie phone number) they are logged in as.  Consumers want common applications, which is why GoogleTalk Blackberry and GoogleTalk PC, for example, are such a powerful pair. But do they want these applications all from one provider?  No.

As the Voice 2.0 world starts to mature, you will see these horizontal layers of functionality begin to appear.  Applications, and common services such as identity, and access will be layered on the networks, rather than being baked in.  This is the same transition that every other industry, save telecom, has undergone. It’s the reason that ethernet is successful as a networking standard, and that PC’s are built from components created by many industry players, rather than vertically integrated as mini-computers and mainframes were.

Voice 2.0 Transition

In the future, the physical networks are simply an afterthought.  It’s the applications, directories and access models that matter.  The IMS walled garden being touted by the incumbent telecom players allows them to control these.  It’s a centrally controlled state-run economy built around a bundle where choice is erased, and one size fits all.  Will customers tolerate this?  Or will they demand a more open infrastructure akin to today’s Internet?  I sincerely hope it’s the latter, because I honestly don’t want to buy my television from Rogers, or my cellular phone from Bell.

{ 13 comments… add one }

  • Vijay March 23, 2006, 7:49 pm

    So let me get this right, Voice 2.0 is the next generation network that the VoIP guys are chanting? I have travelled down the same train of thought and sometimes it leads back to the same thing, very close to the IMS architecture. I think we still have flexibility using the IMS architecture as long as we grant support for a federated identification system (apart from the interface to the HSS) and the PDF (Policy definitions) are liberal.

    It's my thought that IMS will come to the stage as a closed system and will remain as such till the ROI is made. After that, the infrastructure will have to be commoditized for innovative services to blossom and as such will be made more liberal.

  • Paul Jardine March 24, 2006, 3:24 am

    I don’t know if the Identity/Directory layer was in red to highlight it, but it seems to me to be a very important layer.
    I’ve argued for a long time that this is the layer telcos should be going after, by opening up access to verified identity through APIs. The customer needs to have the ultimate control of what (and to whom) information is released, which is what the telcos don’t grasp.
    They need to assist their customers to integrate all the services under a single identity (or multiple ids if they so choose). 5 years ago, they could have done it without much trouble, but now there are many different identity/directory solutions out there and there is not such a compelling reason to choose one particular provider over another.
    E.g. if you had your identity/directory information with one of your providers, who would it be? Rogers? Bell? PhoneGnome? or some idependent provider?

  • Kyle March 24, 2006, 5:50 am

    Interesting points you bring up.

    I predict in the future, though, that one will buy a network connectivity service, (i.e. “IP connectivity”, $$$) and end up purchasing “add-ons” from the same carrier again. With Bell’s rollout of fibre-to-the-node, the increased bandwidth will allow people to purchase their IP-TV solution, which will do for television what client/server side scripting did for the Internet. From there they’ll offer to add a VoIP package, and suddenly we’re back to bundling again. Similar can be said for Rogers.

    As for the service pricing, I would agree Rogers has (maybe “had”, haven’t checked in a while) more appealing residential consumer price-points. However, they don’t have (and I haven’t seen any news they will anytime soon) a very high speed network like EV-Do which will be increasingly in demand for Blackberry and other IP enabled PDAs. It will also really allow consumers to tap into the carrier-lucrative bandwidth services, such as streaming-tv to your cellphone, and fast [~2.3 MB/s] MP3 downloads. As for home Internet services, Telecom Ottawa’s DSL is more way expensive than the competition… $59.99+$10modem for the 5.0meg service?! Ouch.

    Great blog by the way.

  • Frank Miller March 24, 2006, 9:56 am

    I’m trying to figure out what everyone is so up in arms about with IMS? The cell phone systems are completely walled off from the Internet to date. Its only very recently that you can get general broadband data connections to your cell phone and then they will generally control its use by limiting the types of applications you can get onto the phone to the extent they can. So, now they’re talking about using SIP as a basis for evolving their networks. Did we really think that the architecture would not take into account the very real needs of these carriers to maintain the security and integrity of their networks? If you look at the specifics of IMS, the vast majority of the “extensions” over vanilla SIP phone services have to do with operational security, authentication, authorization, and integrity. People in the IETF SIP world have known for a long time that the security associated with the base protocols was lacking. Now the 3GPP people have gone off and done something about it and noone seems to like it.

    I guess my point is, IMS is necessary for what its being used for. Sure they’ve added some non-standard headers and they want to use IPSec which forces you to use IPv6 but these things in and of themselves do not change the fact that I can still go use Vonage or whatever for phone service if I want. Where this will come into play is when they start to use it to integrate VoIP phones with cell phones and landlines and all the rest. But remember, everything we’re using now except for VoIP phones is already “walled off.” IMS doesn’t really change that. What it does is provide a mechanism by which these cowboy Internet softphones can be brought into the fold and interoperate operationally with the other types of terminals in the horizontal fashion thats in the figure that Alec posted. Not only is this probably required to make sure that everybody plays together nicely in the presence of the very clever and determined warez boys, in the end, it will likely increase the overall quality and reliability of the “cowboy” Internet UAs that are out there today.

  • Adam Uzelac March 24, 2006, 12:13 pm

    I view the IMS/bundling of services association in a different light. If the subscriber’s experience of a “thing of interest” or “application of interest” is dependant of the quality of the medium, security of identity, etc – then the IMS frame-work offers solutions that a provider of that application should definitely consider. It’s the “assumed” aspects of the real-time communications, like there will be enough bandwidth, or the communication is secure, that cause the experiences to be less than acceptable.

    IMS solves the “silo-offer” approach that was a relict of the circuit-switched world. IMS has the potential to deliver on the promises of VoIP/Voice 1.0 in the “why the soft-switch”? Argument of 5+ years ago. Alec – It not obvious to me that a Walled Garden is imminent with IMS, as the market-place will not allow that to happen, as you pont out in your original post.

    Adam “voiploser” Uzelac

  • Alec March 25, 2006, 12:52 pm

    Vijay, I don't see the incentive for the operator to open up an IMS system once it has come to market. Do you?

  • Alec March 25, 2006, 1:08 pm


    Thanks for the tip. I've been a happy subscriber to Telecom Ottawa's 3 mbit service (34.95/month) for some time. Didn't realize they had a 5 mbit service. The biggest reason I switched from Sympatico was that they email servers at the time were completely unreliable. Some times it would be two or three days before I could access mail. Also, I owned my own DSL model, so the $10/month rental didn't make sense.

    Rogers doesn't have EVDO, true. And if you can get EVDO coverage (here in Ottawa all you can get is 1xRT) it's great. I am personally excited by UMTS.

    Choice is really the nub of the issue for me.

  • Alec March 25, 2006, 1:13 pm

    Frank, absolutely agree with what you're saying. I simply want a more open network. I want choice — of applications, (why can't I choose which voice mail service provider I want?), terminals and so on. If an IMS carrier came to market offering that, then I'd be happy. I don't believe they will do it though.

  • Alec March 25, 2006, 4:55 pm

    Paul, red was chosen simply to contrast. No other reason. But yes, I agree completely that one of the most important opportunities is the directory. And I agree, and have thought for a long time, that the directory should be open. Identity is a foundational element of any communications system.

  • Alec March 25, 2006, 5:16 pm

    Adam — it’s not obvious that a walled garden is imminent, but perhaps you remember the story of the fox and scorpion, and the river crossing? A fox meets a scorpion on the river bank. The scorpion asks to ride across the river on the fox’s head, and the fox refuses. “You’ll sting me, and I’ll die”, says the fox. The scorpion points out that if he were to do that, they would both die. Eventually the fox consents and they start out. Part way across the river the scorpion does sting the fox, and as the fox is about to drown, he asks they scorpion why he did it. The scorpion answers “it’s my nature”.

  • Vijay March 29, 2006, 5:01 am

    Hi Alec,

    Just when cruising through this post realized that you had responded to my comment. Someone really out to work on something to keep track of comments… but anywho..

    The answer to your question is that, there is no definite answer. But given that eventually the value of the network would drain and they will be seeking for new services, there is no way to replenish the need without opening up the network. Also, just the fact that so many closed systems, including the source of solaris opening up, just gives hope.. and just that, for the future of the IMS network. I guess, in one word, its just that – Hope.

    In another perspective, if IMS doesnt open up, it would be way too easy for P2P SIP based networks to beat the living lights out of the sturdy IMS network. So they wouldnt have much choices for soon.

    As I said, the IMS architecture is splendid. The way the policies get defined is what makes all the difference. Maybe they will go easy on it, and thats one part of the system which is not yet carved on stone anyhow.

  • Patrick Mullen March 31, 2006, 11:17 am

    Hi Alec,

    Some of our guys are working in some research labs for IMS offerings (and at least in the US, they are still working on it, no matter what you read from vendors) and the conversations I have had with people that probably know what they are talking about is that IMS will actually make it easier for 3rd parties to create things that can be plugged into the pipe carriers distribution footprint.

    As for the bundles being attractive, just because certain providers don’t offer attractive bundles doesn’t mean consumers don’t want them provided from the same carrier, maybe just not that specific carrier. The bundles now might be tied together, but they are not really integrated. IMS helps with actually bringing services together. It also allows agility and improves the time it takes to bring a new service to the consumer.

    IMS isn’t going to hurt VoIP and the pureplay companies at all, but the fact that voice is trending to zero and the only value in voice is to cement the customer relationship for other services (data/video) is what will really hurt the pure plays.

    If a carrier can offer you voice/video/data/wireless with good service at a overall lower cost, then of course consumers want all of the services. That is how the bundles will win, lower overall cost. Just because some companies today are not doing it right doesn’t mean that it won’t happen. I have read that the Verizon Quad Play with FTTH will be around $100. That is hard to beat.

  • Alec March 31, 2006, 3:50 pm

    I think you nailed it Patrick when you said that the fact that voice is trending to zero will hurt the pureplay VoIP guys, not IMS. From my POV as an application provider, what I want is to have access to the voice network. What that means long term is that the network operator has to open up. Otherwise, they’re easy meat for someone who has an open network.

Leave a Comment