April 1, 2008
Cable OSS Roundtable
Operations Support Systems Experts Speak their Minds
By Jonathan Tombes, Editor
Some of the questions are classic or evergreen. What's the right mix between custom-built and off-the-shelf technology? How does the cable-style back office stack up against that of the telcos? Others topics are more recent, for instance, related to cable operators' launch of Internet protocol (IP)-based business services and the so-called "mash-up" of applications.
As we did last year in February 2007 with regard to IP Multimedia Subsystem (IMS), we've rounded up a collection of experts on operations support systems (OSSs) to share their answers to those and other timely questions. An entire session at this year's SCTE Conference on Emerging Technologies (ET), moderated by Charter Communications Vice President IP Engineering and Development Matt Bell, focused on "Rapid Application Delivery and Service Quality Management."
The presentation by Rich Woundy, Comcast's SVP for software and applications in the office of the CTO, provided a sense of how IP-based services has complicated service support considerably. (Let's leave aside switched digital video and targeted advertising for the moment.) In his schematic of network interaction on the control plane, Woundy showed six categories leading into data reduction and analysis and the enterprise service bus, namely: IP detail record (IPDR), simple network and management protocol (SNMP), intrusion detection system (IDS), Syslog, Radius and Diameter. The days when SNMP sufficed to track network elements are long gone. Read on to hear more about what these OSS specialists think about the current landscape. -CT
Brian Cappellani, CTO, Sigma Systems
Your ET paper talked about OSS at the edge. What is "the edge" from an OSS perspective?
By "the edge," I mean closer to the invocation of the service - the edge of the network. The intelligence about the subscriber and control of access to information traditionally associated with OSS is moving from the "back office" to be closer to the applications requesting it.
This is being driven by a couple of factors: the more "on-demand" nature of the types of applications and services being deployed on the network and the potential for new business models for interaction with third-party applications.
Is it safe to anticipate a rise in third-party applications over cable's various platforms (OCAP, IMS/Packecable 2.0, etc.)?
I think that service providers see the large array of applications being developed that serve the "over-the-top" market, and the innovation that has driven, and realize that there is potential value they can add to the experience of users of these applications by using their existing enablers, including subscriber information.
OCAP is designed to foster that type of third-party application development community by exposing enabling capability within the confines of the set-top. I think there is a larger opportunity for operators to monetize the assets that they have through revenue streams outside of just charging subscribers - whether that is through support for advertising-based models or policy-controlled access to enablers such as presence, location, QoS, etc.
Would you include business services in that mix? If so, what adjustments to the industry's control plane are necessary?
Yes, I would include business services in the mix. In fact, the ability for businesses to create their own tailored applications using exposed capabilities like click-to-call or location enablers can be a powerful customer retention tool. If as a service provider, your customer has their application layer connected to your infrastructure, it is a powerful disincentive for that customer to move to a different service provider, even for a lower price. One might argue that the business market has the critical mass of developers that could sustain a development community around these APIs (application programming interfaces). This situation is already occurring in the wireless industry.
To facilitate this, you need a strong policy infrastructure that goes beyond the current network/resource-based definition of policy to look at subscriber and partner-based entitlements.
Are there any standards developments that you find noteworthy and relevant to the cable operator community?
Around the notion of exposing subscriber information, I think the work being done in the SCTE 130 standard for advanced advertising is particularly appropriate, especially Part 6, the Subscriber Interface System.
From a broader perspective of defining the way this environment might function, I would point to the work being done by the Open Mobile Alliance in their OMA Service Environment and by the TeleManagement Forum in their Service Delivery Framework group.
Bill Otto, CTO, TierOne OSS
Process automation seems to be one of your core competencies. What sort of software or abstraction layer technologies do you favor in that area?
Process automation in the OSS provisioning space for traditional services requires that a number of tasks be defined that can execute sequentially or in parallel. These tasks may be manual or automated through interfaces with systems that automatically perform various actions required. This is typically implemented using workflow technology that allows processes and tasks to be easily configured and supports the necessary interface mechanisms. The spread of SOA (service oriented architecture) technologies and interfaces is reducing the complexities of implementing automated processes.
With the adoption of IP and the move toward self-provisioning and real-time service configuration, the focus will move from process to transactional for the delivery of services. We see SOA as an important enabler in providing the abstraction required for this transition.
One of your mantras is expertise in both third-party OSS vendors and telco environments. Could you elaborate? Is work in this area necessarily consultative and customer specific (vs. off the shelf)?
Current OSS products are very configurable, able to support a variety of specific processes, network devices and configurations and data requirements. When deploying these OSS products as part of a solution for a service provider, it is necessary to understand the business goals of the service provided, the processes that are used and how the system(s) will support the processes, and the network technologies and services that are being used in order to provide the best solution for the customer. It is also very important to have knowledge of the specific OSS to determine how it can be utilized as part of the solution.
One of your announced customers is Rogers Business Solutions. Would you consider them cable or telco (or both or neither)? Can you describe your work with Rogers further?
Rogers Business Solutions is now part of Rogers Cable, so from an organizational point of view, they are a cable entity. The services that they offer are generally the types of services that are associated with more traditional telcos. TierOne's work with Rogers (Cable and Wireless) has primarily to do with inventory management, network discovery, and Remedy ARS applications, as well as some work related to fault management.
What are some of the specific challenges facing any provider offering solutions for the business services market?
Specific challenges in the business services market include:
• Cost pressures - being able to deliver services profitably at a competitive cost
• Efficient network operation - a key component of the cost of delivering services
• Quality of service - delivering the high quality, reliable service required to maintain customer loyalty
• Speed of service delivery - customer services need to be installed and operational in a short period of time
Is it your sense that cable operators have OSS requirements that are distinct from those of the traditional telcos? Or is the line sufficiently blurred these days?
OSS requirements for cable operators and traditional telcos are becoming much less distinct. Both are delivering the same services (voice, video, data) and are converging on similar core network technologies (e.g. VoIP, Ethernet, MPLS, IP), so the differences are primarily in the access/delivery mechanisms (HFC vs. copper DSL or PON). So while there may be differences in how services are delivered requiring specialized systems or implementations, there are many common requirements in network planning and capacity management, effective network operation, and providing high quality, reliable services that result in increasing commonality in OSS requirements.
David Jacobs, CTO and Co-Founder, JacobsRimell
You have suggested that there's a strategic imperative for cable operators to move from being "producer efficient" to "customer effective." Do you think the mindset is changing?
Being customer effective is essentially saying, "Are our systems tuned to the needs of the customer?" vs. "to our own use." In terms of seeing a change in the mindset, at the top of the organizations the mindset has definitely changed; however, as you move further down through the organization, this way of thinking diminishes as they have to work the new mindset thinking into the existing or legacy infrastructure and operational processes and procedures.
We still believe cable operators are more advanced with this way of thinking, as it's inherently in their DNA from their TV background; however, their Achilles heel has traditionally been their customer care remit - and this is where major changes have clearly been implemented. Senior management has recognized that there are major costs (both direct and indirect) associated with poor customer care, for example, and are looking at how to weave customer effectiveness into the organization as a whole and not just at the customer support level. A good example here is that "producer efficient" is generally associated with a single product offering point of view, whereas "customer effective" is more aligned with converged services delivery where cable operators are now far more efficient at delivering aggregated services.
How would you measure such an actual shift taking place in operational strategy?
At the end of the day, the only measure that counts is churn and services adoption rates. Clearly the cable industry has had a great run at this lately.
Third-party applications play an important role in your view of the future. Do you see these riding OCAP/tru2way and PacketCable 2.0/IMS infrastructure? Or more so coming over the top?
Absolutely, these third-party applications become the operators' "mashed" or blended products and services and will rely on enabling technologies such as OCAP and PacketCable 2.0 in order to allow the operators to more transparently implement "external" services. For example, IMS in PacketCable 2.0 will enable a common signaling environment, enabling more reliable services delivery; OCAP enables applications to be more openly (widely) deployed rather than having to be set-top box environment-specific.
Can you further clarify the question of who owns the apparent no-man's land of cross-functional challenges?
Ultimately, product managers are having to work in a broader cross-silo mode as the boundary between services is blurring as they move to being more open and subscriber-centric. However, we still believe this has a long way to go with most operators due to the legacy siloed architectures (including those of revenue recognition and cross charging) that are established.
The barriers to improving this cross-functional co-operation, though, are typically due to lack of reporting and visibility for revenue and cost allocation.
Is database management expertise lacking among cable operators? That seems to be a large part of what it's going to take to manage the increasingly complicated matrix of subscribers, multiple applications, billing and provisioning.
We don't think this issue purely relates to database management - technology is not the fundamental problem here. This is an organizational and strategic issue related to actually being able to focus on customer needs, integrate multiple suppliers, and relate these to a strategic plan. The key issue here, then, of course, is being able to effectively relay these concepts to the technology and operational groups with sufficient leadership in order to effect what is required, which will be a company-wide data strategy.
Bob Hobbs, VP of Marketing/Business Development, iBBS
Where would you say iBBS stands on a spectrum, with off-the-shelf software on one end and custom-built solutions on the other?
IBBS is neither an "off the shelf" product nor a "custom built" solution. We provide our technology in software as a service model where we leverage the Web, our know-how, and scale economies to put leading-edge solutions within the reach of all broadband providers so that they can better compete in their respective markets.
How has the OSS/BSS arena shifted for cable operators over the past two years?
During the last two years, cable operators have begun to significantly scale their data business and have launched voice services - in many cases, the services were launched without a significant investment in the back-office systems supporting those new services - and thus each service was a silo unto itself. Many of the inherent advantages of triple-play bundles - at the application level - cannot be fully realized without having the support systems fully integrated as well. Additionally, end user segments are stratifying - providing an opportunity to increase revenue, reduce costs, enhance the experience, in a more customized way. But without the appropriate back-office systems, operators cannot take advantage of these new marketing opportunities. Finally, looking across the enterprise, the lack of the appropriate back-office capability limits the ability of the enterprise to manage its efficiency and learn from its information systems. This could be a longer discussion.
What is happening ... currently there is a degree of integration occurring primarily between the business layer and the services layer, especially on the voice and data sides. There also is some lower level integration across services silos - for example, data service management integrating with voice service management.
Do telcos have a back-office advantage over cable?
Yes, without question. Not because they are more innovative, but because they have been investing in it as a mission critical asset for many years. There are mulitple network layer and business layer interconnects between peering companies, there are significant taxation requirements on calls, the systems are by design usage based, and network availability (99.99 percent) was not an option. There are also several regulatory requirements on telcos that have forced better OSS/BSS capabilities relative to cable.
Is there still a lot of no-man's land (between marketing, engineering, billing) in the "cable IT" world?
Yes. You can probably derive some of my thoughts on this from the other answers. The business layer is not fully linked to the services layer, to the network layer, to the element layer, so the quality of data generated by networks and the correlation to customers is not available. And this issue is present almost independently in each of the triple play services, so not only do we need better vertical integration, but also better horizontal integration in order to derive competitive advantage from the OSS/BSS.