IP technologies, including Web 2.0 and the emerging interactivity of the Web, can be combined with service provider APIs and hosted features to create new revenues and new partnerships for network operators. These tools from the Internet side can be combined with application and service tools, including Parlay and IMS on the operator side, and with SDP-hosted features overall, to create a new business model that eliminates the current risk of disintermediation that network operators face in an Internet/IP era.
- IP revolution creates "IP transformation" services revolution
- Open networks: Offering third-party access to network services and features
- Signaling-based services could be next major telecom profit source
- Podcast: IP transformation and new revenue-generating strategies
IP revolution creates "IP transformation" services revolution
IP convergence is a topic so old that some in the industry today were still in school when it was first raised. While it's only fair to say that everything is definitely not converging on IP (optics and Carrier Ethernet are examples), it is certainly true that services and networks have become very IP-centric.
Convergence, in short, has arrived. Given that, the term "IP transformation" has become the current "it" phrase, and it may seem like just an attempt to create a new buzz. It's not. Transformation is in some ways the inevitable consequence of convergence and, in others, the thing that drove it in the first place. Transformation has not arrived, but it's trying.
All the Tier 1 operators, and probably 80% or more of the Tier 2 and 3, have accepted the fact that the IP revolution at the technology or network level has created a revolution at the service and business level. A decade ago, consumer broadband was for all practical purposes a non-issue. The Internet was an emerging information tool but not a major service planning problem, and the over-the-top players that existed were just a bit beyond the garage-shop stage. Today, the Internet is the service framework of "first resort," the place where both users and enterprises expect to see their communications needs met. Everything else has to be framed around that expectation.
IP transformation as a business strategy
IP transformation is the term used by many operators to describe the changes they must make to their service and business planning processes to accommodate the Internet reality. The key to the process, in a technical sense, is creating a framework for service planning and deployment that matches the pace of changes in the current market -- a pace set by the Internet revolution.
Operators vary somewhat in their vision of how IP transformation can be achieved, but most programs have the following key elements:
- The notion of the network as a platform for applications and not just something that pushes bits from place to place. This implies that information technology tools like service-oriented architecture (SOA) and business process management (BPM) would be used to help bind networks and applications together, and that standards that support a linkage of hosted features and connection services (such as the Telemanagement Forum's Service Delivery Framework activity) are of special importance.
- Support for very fast and automated service activation and deployment, through a combination of creating services using those same IT tools and the use of service automation provided in OSS/BSS and service management systems.
- The creation of new provider "services" -- security services, identity services, location services, demographic and advertising services, extranet, unified communications, and more. These services, like custom calling today, would fatten provider bottom lines without requiring massive capital investment.
- The exposure of service components, including those in the third point above, through application program interfaces and other mechanisms, to allow for their use by partners and developers in creating services that are "on the top" and not "over the top" in terms of the service provider.
- The extension of the tools and elements of all of this into a new set of enterprise and consumer service offerings that focus on providing an experience on demand and not a connection. This is likely to include Software as a Service (SaaS) and managed services (managed LAN, managed security, managed intranet, managed extranet, etc.).
- The network is a platform, a base on which other services are built, which must therefore be aware of the higher-layer applications, optimized to secure their performance, and sustained as a barrier to market entry to competitors and a long-term differentiator to providers.
Operators today can articulate the goals of their IP transformation plans, and even the architecture principles on which they expect them to be based. They are less successful in framing how they will implement the plans because they are not confident their vendors can yet provide all of the tools.
Network devices implement "convergence," but operators show that "transformation" is more an IT process. This IT focus creates new issues and opportunities for both operators and vendors.
Benefits of transformation
A "transformation" mindset favors either equipment vendors with a strong position in OSS/BSS software and service delivery platforms (SDPs) or a division of project activity between the network layer and the IT layer. Since transformation has become a significant issue only in the last year, there has been a market rush to accommodate the new paradigm, and the results of this are not yet fully visible.
Steps such as the NEC acquisition of OSS/BSS vendor NetCracker are highly visible, and yet the role-transformation opportunity played in the deal has not been acknowledged. More subtly, activity in standards groups from the TM Forum to the GSMA and the ITU has shifted to a "higher-level" focus. It is likely that further movement, particularly in the form of M&A and partnering among vendors, will occur as transformation becomes a more visible driver.
When that will be depends on perspective. Any of the six key attributes of transformation outlined here can be easily identified as the trigger for some current market activity. Still, a Google search for the term "IP transformation" yields many system integrators among the top rankings (including Accenture and Pricewaterhouse), but only Alcatel-Lucent among key vendors. This is probably due to the fact that Alcatel-Lucent has a strong OSS/BSS and SDP position and is thus in a better position to leverage the transformation opportunity. Cisco clearly has designs on the space, and its Active Network Abstraction is a partnership with OSS/BSS vendors, including IBM Tivoli, which promises vendor-independent linking of network and operations management, a key element in securing the cost-effective operations that are a key to IP transformation.
All of the transformation strategies anticipate standards that are only now developing in bodies like the TM Forum and ITU. In addition, the concept of IP transformation is simply too broad to be implemented by a network operator in a single step, and the priority each places on the six items introduced earlier will change the range of vendors that might best address their needs at this early stage. As the market matures and operators advance their programs, a full spectrum of solutions will be expected and the market space will become very active indeed.
Open networks: Offering third-party access to network services and features
The term "open networks" is like many terms in telecom -- charged with emotion and hampered by multiple conflicting definitions. To some, the concept means "net neutrality" and a threat of further "disintermediation" by over-the-top players. Yet the most generally accepted definition of "open networks" among the operators themselves is "networks that provide third-party access to internal features and behaviors of services, not just to transport/connection resources."
Under this definition, not only are open networks not a threat, they're a path to revenue growth and perhaps the future of telecom. The question is how to get there.
The Internet is surely an "open network," but one that exposes only the ability to connect with others on a best-efforts basis. There are other things a network can offer, including some -- like location and presence services -- that exist today, and others -- like identity management and even demographic profiling for ad targeting -- that are yet to be provided in an organized way.
Some of these non-transport services or features are bound tightly to network operations and could not be provided by anyone other than the operator. Others could in theory be implemented by anyone but would probably offer a relatively low return on investment and involve a substantial deployment before they could be profitable. These are situations that network operators are used to dealing with, but they tax the financial resources of most other companies and nearly all startups. The mission of "open networks," then, is making these new incremental features and services available to third parties -- not necessarily free of charge, but available.
Two approaches to open networks
There are two popular paths to open networks today: the "open handset" approach and the "open API" approach. The former is obviously aimed primarily at mobile networks or networks where the user's appliance is made available through the network operator and not purchased retail and installed by the user. The latter is aimed at software developers and partners with development capability and is designed to create a means for adding new applications and innovations to telecom services. Given the long planning and capital cycles of the telecom industry, most executives feel that tapping third-party innovation is critical.
The open handset route
Open handset programs drive innovation from the outside in, making the assumption that a rich and innovative community of vendors and developers building handsets and handset applications will create a new set of applications for mobile broadband. This could help overcome the slow growth of non-voice, non-SMS services and could drive up mobile data traffic and usage, thus increasing revenue. It's likely that the former benefit is the primary driver for interest by mobile network operators. Moving services beyond voice is perhaps the No. 1 mobile priority.
The risk of an open handset program is the continuation of the "disintermediation" that operators feel has resulted from the Internet over-the-top model. If the handset is communicating with portals and other sites via the Internet, it is simply replicating the wireline access model, with all its risks. Still, it is clear that in the near term, at least, an open handset program is likely to speed adoption of mobile data services.
The open API route
The open API (application program interface) programs are more complex and potentially more valuable to the operator. An API is a way for third-party software to access network features directly, creating a new kind of service logic that shifts between pieces hosted by the network operator and pieces created "on top" by partners and developers. This approach is already in use on a modest scale in a number of countries worldwide, and standards groups like the GSMA (GSM Association, an industry group of international mobile operators) are trying to standardize or at least collect open API programs under the umbrella of their "Third-Party Access" initiative.
The challenge for API programs has been steering the course between creating a robust and solid interface that can be supported operationally and that ensures user rights and network stability, and something that's easy for Web developers to use. So far, most of the APIs have been based on a voice-standards set called PARLAY, and most expose simple calling capabilities and perhaps location services. While these are useful, they're not likely to create a revenue revolution for operators. The GSMA approach seems more flexible in tackling the way operators might expose new service features like advanced presence services that can cross between cell phones, wireline and even computer or demographic services to support advertising.
It is likely that the best approach, the one that will ultimately prevail in the market, will be a mixture of handset and API capabilities. An open handset architecture, linked with a powerful set of exposed service features via convenient APIs, could be used by third-party developers, phone manufacturers, Web developers, portal vendors, content players, and any number of partners to create applications. Even at a nominal charge per use, these could significantly build revenues for operators, justifying further network investment.
Creating an effective open network for the future
The general risk in open networks lies in what is often called "operationalization" -- making the network support process scale to the scope of the planned services and ensuring that problems can be identified and corrected even when they cross the boundary between partners and the network. The offering of componentized service features for syndication is one focus of the Telemanagement Forum's Service Delivery Framework team, and the standards here should make it possible to integrate open network access programs with OSS/BSS and customer care processes effectively. In fact, it is possible that elements of the OSS/BSS process -- including customer care, billing and journaling, and surveillance and operations control -- could all become "features" exposed by the open access programs themselves.
Open network programs are an essential element in the maturing of the telecom industry in the face of voice commoditization. Although the optimum shape of these programs is still in doubt, almost every major network operator is aggressively committed to supporting them, and standards bodies are working on the operational details. In the long run, these programs may be more important to the evolution of network services than more traditional "closed-network" architectures like IMS. In that case, the success of IMS may depend on its ability to work within an open network framework.
Signaling-based services could be next major telecom profit source
Service providers have always believed that they are in the "service" business, as the name suggests. Look deeper, though, and infrastructure investment for the providers has typically been focused on connection and transport -- on bits. It's logical, but it may be that the traditional notion of bits as the network's optimum product is obsolete.
Technology improvements have driven down the cost of a bit, and the realities of marketing broadband services to consumers have driven down the price of bits as well. One result is a decline in revenue per bit -- now reportedly more than 50% per year. Control of operations costs has been another focus as operators struggle to accommodate the changes in their markets. The most important changes are likely to come by looking more closely at the bits that have been the traditional source of provider revenues. Should we think fat bits, thin bits, or maybe just no bits?
The concept of fat versus thin bits refers to the notion of differentiated network services versus undifferentiated services. Some network operators believe that quality of service, meaning premium bit-handling, offers a way to increase margins or "fatten" bits, but many (perhaps most) fear that the best-effort notion of Internet services will prove good enough for most applications. That leads to the "no bits" conclusion.
Making no-bit profits is traditionally seen as a focus on content, Software as a Service (SaaS), or something delivered over the network. This essentially admits to the notion that bits may become a loss leader, and profit has to come from something bits deliver. There's another possibility, though. There are two things inside a network, and bits are only one of them. Signaling is the other.
Making signaling the revenue generator
Signaling is the exchange of information about communications requests and communications users that facilitates connection and information exchange. SS7 is the signaling system of the PSTN; SIP is the signaling of traditional VoIP and IMS. Signaling controls services and relationships over networks, and while voice services based on signaling (custom calling features) have declined in price, they're holding price and margins better than services that are purely bit-based.
The most general definition of signaling is "a mechanism for exchanging intelligence and controlling behavior at a distance." In networking, the term is most often used to describe the interaction between a network user and the network that provides service access. The challenge for making signaling the basis of profitable services is in some ways the challenge of making networking's narrower definition more like the broader one.
The SS7 and SIP relationship in voice may be an example of both the opportunity of next-generation network signaling services as a value base for service providers and an example of what's wrong. SIP signaling isn't particularly sophisticated or flexible; it's very much like voice signaling in the PSTN. Some believe that a more flexible approach should be taken.
XML, SOAP, Web services and related technologies are one new way that service providers are looking at signaling. One reason is that most of the big Web players, from Amazon to Google to Yahoo, base their "Web services" on XML signaling between user and Web portal. It's proved to be a powerful approach, and the Parlay Group has defined some Web services for popular PSTN Advanced Intelligent Network (AIN) custom calling features and services. For services beyond the traditional ones, it's less clear where signaling standards based on XML might come from. It's even less clear whether any standards for non-voice-related services would get meaningful support from the Internet community.
Signaling centralization and cost questions
Another question with advanced signaling is that of centralization and cost. IMS, as many have noted, is a centralized architecture, even though many of the components are distributed. SIP relies on servers in most of its applications, and network operators are increasingly interested in distributed signaling applications. What are they? Examples are Skype and BitTorrent; both are forms of peer-to-peer.
Nearly all P2P applications are forms of what is called a distributed hashing tree (DHT), and when pictures of the structure are drawn, people cringe at the thought of the number of hops or nodes that information might pass through. P2P doesn't take actual traffic through these nodes, though; it only distributes signaling. The request for something (a call) passes through a structure to reach its destination, but once it does, the actual IP addresses are exchanged, so data connections are made directly.
P2P is a very attractive concept as a framework for signaling in a next-generation network for two reasons:
- It can be used to transfer any kind of information, from a simple filename to a massive piece of content. Obviously, it can be used to transport XML.
- It is a network based on facilities hosted by participants, with minimal (sometimes no) centralized infrastructure. This reduces both capital and operations costs.
Network operators are already experimenting with advanced signaling in their own developer programs, often called "Third-Party Access" referring to the GSMA program for adding Web support to telecom services with that name. The GSMA program is also a contender for the source of a standardized system for advanced signaling using XML, but the work is still in its very early stages.
If the service providers get the details right, signaling-based services could be a major profit source. The network operator's most powerful asset, after all, is its knowledge of the users of the network, and this knowledge is essential to create effective, valuable services. Bits without signaling offer no real value, and adding value to networks may be easier by adding it to signaling than to transport.
Podcast: IP transformation and new revenue-generating strategies
You must have Adobe Flash Player 7 or above to view this content.See http://www.adobe.com/products/flashplayer to download now.
Download for later:
IP transformation and new revenue-generating strategies
• Internet Explorer: Right Click > Save Target As
• Firefox: Right Click > Save Link As
Service provider APIs and hosted features can create new revenues and partnerships for network operators by combining IP technologies, including Web 2.0 and the emerging interactivity of the web. Application and service tools including Parlay and IMS on the operator side can be combined with Internet tools to develop new business models that could eliminate the risk of disintermediation that network operators are witnessing in the Internet era. In this podcast, Site Editor Kate Gerwig and CIMI Corp. President Tom Nolle discuss the IP transformation and its relationship to the network as an application platform.
About the author: Tom Nolle is president of CIMI Corporation, a strategic consulting firm specializing in telecommunications and data communications since 1982. He is a member of the IEEE, ACM, Telemanagement Forum, and IPsphere Forum, and he is the publisher of Netwatcher, a journal in advanced telecommunications strategy issues. Tom is actively involved in LAN, MAN and WAN issues for both enterprises and service providers and also provides technical consultation to equipment vendors on standards, markets and emerging technologies. Check out his SearchTelecom networking blog Uncommon Wisdom.
This was first published in July 2009