| » Home |
| » About Us |
| » Events |
| » Publications |
| » Sponsors |
| » Advisers |
| » Projects |
7. Connectivity, access and co-operation
7.1 Any-to-any connectivity (AAC)
7.2 Numbering and addressing
7.3 Access
7.1 Any-to-any connectivity (AAC)
7.1.1 Overall principles
• Any-to-any principles to govern all public communications services, including mobile and Internet
Source: Peter Gerrand
(Global) AAC means an obligation on all participating service providers to connect end-users of a designated service to any legitimate information source connected to the same worldwide network, be that another end-user (a person or device) or any provider of that designated service (whether a person or a device or a set of persons and/or devices).
AAC has been implicit in the design of public communications networks since the introduction of automated telephone networks; indeed the need for AAC, free of the commercial bias shown by a human telephone operator, is believed to have been the prime motivation for Almon B. Strowger’s invention of the first patented automatic telephone switch in 1891. Within 70 years, the AAC concept expanded from connectivity between all subscribers to the same local telephone network, to connectivity between all subscribers to the same automated national network, and thence to the goal of world-wide connectivity between the users of public telephone services on the international telecommunications network.
The social and economic benefits of the AAC principle have been so obvious to the general public, that a similar expectation is now widespread for new digital communication services, and should be incorporated in any future Communications Act: i.e. the AAC principle should be applied to all modern public communications services, including mobile services (such as Short Message Services) and Internet services (such as email and downloading from public websites).
The Object of Part XIC of the current Trade Practices Act includes ‘the objective of achieving any-to-any connectivity in relation to carriage services that involve communication between end users’. However the result of recent PhD research by Ross Kelso has been to discover that the incorporation of the AAC principle in the Australian 1997 telecommunications legislation occurred ‘with a diminished meaning and with less than satisfactory outcomes’ (Kelso 2007: 198). For example, the Telecommunications Act 1997 mentions ‘connectivity’ only once, and in the context of a ‘standard telephone service’ only (Kelso 2007: 215).
If the AAC obligation is not incorporated in a new Communications Act, the absence will provide scope for unscrupulous service providers to provide a technical barrier to other service providers, through insisting on higher charges to provide comprehensive AAC.
The invention of network numbering and addressing made it possible for network operators to apply AAC to the connection of telecommunications calls to users on other networks, users to whom they could not be reasonably expected to keep any information for commercial identification purposes. It is important to understand at this point that the AAC principle can only in practice be enforced with respect to the network numbers or addresses assigned to end users or other sources of information (e.g. Internet websites), and not to any desired end user person or business entity. To put this point more concretely by example: a public telephone network operator is obliged to make best efforts to connect a telephone call to the network termination point, e.g. the phone line, associated with the dialled telephone number; but cannot be expected to arrange for the user to be ready and able to answer the phone when it rings.
• Net neutrality: enable differential level of service provision and standards on an equal and transparent basis
Source: Vicki MacLeod, Principal, MacLeod Consulting
The difficulty with the current level of debate around net neutrality is that there is no clear definition of what it actually means, and where legitimate network management practices should be allowed. This proposal is about how to set the balance between the competing interests.
If net neutrality means no anti-competitive discrimination, everyone can agree with this aim, and in Australia we have a Trade Practices regime to deal with this.
However, where net neutrality has come to mean a shorthand for an absolute prohibition on price and service discrimination of any sort (where bandwidth hogs cannot be charged more than someone only sending a few emails) then this is clearly inefficient and would hamper the development of new business models and new services that consumers want and need.
The US’s Federal Communications Commission (FCC) has been at the forefront of dealing with the questions raised by net neutrality. In 2005, the FCC adopted an Internet Policy Statement and established four consumer-based principles:
1 Consumers are entitled to access the lawful Internet content their choice;
2 Consumers are entitled to run applications and use services of their choice, subject to the needs of law enforcement;
3 Consumers are entitled to connect their choice of legal devices that do not harm the network; and
4 Consumers are entitled to competition among network providers, application and service providers, and content providers.
The FCC has supported legitimate network management practices, but found against cable provider Comcast on 1 August 2008, on the basis that its practices were unduly discriminatory against one particular service.
The key policy driver is to ensure a balance between the needs of ISPs to manage their networks and their traffic, and the legitimate interests of competitors and consumers in ensuring that these management practices do not amount to anti-competitive discrimination.
The recently released draft report by the UK Government into the future of communications policy, Digital Britain, highlights this balance as follows:
Internet Service Providers can take action to manage the flow of data – the traffic – on their networks to retain levels of service to users or for other reasons. The concept of so-called ‘net neutrality’, requires those managing a network to refrain from taking action to manage traffic on that network. It also prevents giving to the delivery of any one service preference over the delivery of others. Net neutrality is sometimes cited by various parties in defence of internet freedom, innovation and consumer choice. …….Ofcom has in the past acknowledged the claims in the debate but have also acknowledged that ISPs might in future wish to offer guaranteed service levels to content providers in exchange for increased fees. In turn this could lead to differentiation of offers and promote investment in higher-speed access networks. Net neutrality regulation might prevent this sort of innovation…
…the Government has yet to see a case for legislation in favour of net neutrality. In consequence, unless Ofcom find network operators or ISPs to have Significant Market Power and justify intervention on competition grounds, traffic management will not be prevented.”
• End-to-End performance codes for public telecoms networks
Source: TransACT
…[C]ustomers value simplicity and reliability because the Internet is important to nearly every aspect of their daily life and this will become universally true with ubiquitous deployment of the NBN. Like other utilities, minimum standards are needed for the end to end performance of the service – or at least the elements of service that are controlled in Australia.
The retail ISP, the broadband access network provider, the backhaul provider, the interconnect/peering arrangements and the distant networks and servers all contribute to the overall performance experienced by end users. Today, unlike in the telephony world, there is no NBN product set agreed across each member of the value chain. TransACT suggests work could begin in industry associations (eg the Communications Alliance – formerly known as ACIF) to agree performance of public networks in Australia – as mandatory or voluntary codes – and so ensure consistency of experience for users of the NBN.
[Ref: TransACT, A blueprint for the 5th Utility, June 2008, viewed 31 July 2008, http://www.dbcde.gov.au/communications_for_business/funding_programs__and__support/request_for_submissions_on_regulatory_issues/submissions/TransACT.pdf]
• Any-to-Any rules to apply to machine-to-machine communications
Source: TransACT
Any-to-any provisions in the current legislation are taken to mean person-person communications. But, consideration should be given to whether it should include machine-machine and person-machine; both of which have meaning in the IP world.
[Ref: TransACT, A blueprint for the 5th Utility, June 2008, viewed 31 July 2008, http://www.dbcde.gov.au/communications_for_business/funding_programs__and__support/request_for_submissions_on_regulatory_issues/submissions/TransACT.pdf]
• Network neutrality requirements and practical implementation
Source: TransACT
TransACT, a carrier in South-East Australia, has proposed practical approaches and regulatory principles for addressing network neutrality once a national broadband network is operational. It says that network neutrality is of vital interest to those who deliver applications over networks independently of the network owner or ISP:
Extract ..
TransACT proposes to treat all services that use the network equally. Where services receive expedited delivery (guaranteed Quality of Services) it will be explicit in customer contracts; as will any shaping of traffic to ensure availability of service to all users. TransACT proposes to publish its policies and report monthly on interventions to network delivery (e.g. where traffic shaping has been applied to file sharing traffic).
The reason this debate is so important is that the ISP service is the general purpose connection for all services and, subject to adequate capacity, can displace most other specialised forms of connectivity. It is important for consumers as it will reduce the overall connectivity/communications bill over time. It is important for merchants as it will increasingly be the primary means of reaching consumers. And lastly, it is vitally important to the providers of network connectivity as it is becoming the primary form of connection and displacing other forms and associated revenues.
The interaction between pricing, service model, and usage must be faced. Without incremental cost for usage, demand will exceed capacity – some end users will download vast collections of videos (more than any one person could watch),others will have permanent video feeds from favourite cities, beach houses etc whether watched or not. Charging is the first measure to manage demand. The second is to ‘shape’ the traffic according to the needs of the particular stream:
Few would argue with the proposition that suppliers of connectivity services should be able to charge for quality of service or be able to allow either A or B Party pricing (calling or called party pays in telephony terms). Allowing more charging options would be in the long term interests of the end user as many services on the Internet are appropriately supported by merchant, B Party or advertising fees and there is no reason why the same parties would not be prepared to fund some or all of the costs of the user’s connection.
As a practical approach to traffic shaping, TransACT proposes ‘Neutrality with Disclosure of Policies and Actions’ as mandatory for all providers of services over the NBN. Retailers would implement this at the retail layer while wholesalers would need to provide the option for implementing such an approach where it has infrastructure implications. Key features of this approach are:- All applications of a class (VoIP, video streaming, file sharing) are treated equally regardless of supplier (subject only to equivalent resource demands)
- Shaping policies are disclosed in service agreements
- Actions related to this policy are reported monthly.
[Ref: TransACT, A blueprint for the 5th Utility, June 2008, viewed 31 July 2008, http://www.dbcde.gov.au/communications_for_business/funding_programs__and__support/request_for_submissions_on_regulatory_issues/submissions/TransACT.pdf]
7.1.2 Fixed line voice services
• The essential call status indicators should be specified by the regulator when ‘declaring’ any public communication service Source: Peter Gerrand
Connectivity on the traditional Public Switched Telephone Network (PSTN) is made evident to end users by a set of audible call status indicators, transmitted to the end user’s telephone when setting up, receiving or clearing down telephone calls. The basic call status indicators began when the PSTN serviced fixed line (access) voice services only, and have continued to be used as national PSTN numbering plans since the 1980s have included increasing ranges of numbers for mobile access users. The basic call status indicators should be maintained as the underlying technology for public switched telephone services changes to IP-based switching, to meet community expectations. The essential call status indicators should be specified by the regulator when ‘declaring’ any public communication service, thereby implying an obligation on all participating service providers to provide these status indicators.
As a brief example, the following call status indicators should be declared essential for the public switched telephone service, to meet ongoing community expectations. During the call set-up phase, the calling party (A Party) should receive dial tone within a set time interval, indicating that the network is ready for the A Party to transmit the digits comprising the number of the called party (B Party), using a keypad or other user interface. Within a set time after the last digit has been sent, the A Party should receive one of the following status indicators: ringing tone, busy tone, network congestion tone or an explanatory ‘no progress’ recorded message from a relevant service provider. The provision of ringing tone to the A Party should closely coincide in time with the provision of a ringing signal to the B Party’s telephone or other terminating device.
The technical regulator (currently ACMA) should be authorised to develop and maintain schedules of call status indicators for declarable public communication services.
7.1.3 Mobile voice and SMS services
• Require any-to-any connections in mobile telephony
Source: Peter Gerrand
The concept of AAC needs to be extended to mobile telephone services in Australia, by enforcing a ‘roaming’ hand-over obligation on mobile carriers not providing radio coverage to a customer or to a called party in a given location, if the location is covered by another mobile carrier.
In other words, the failure of a mobile carrier to provide radio coverage of a location in Australia, either temporarily or over an extended period, should not provide any legal excuse from the AAC obligation, if that location is covered by another licensed public mobile carrier. This stipulation will require all mobile carriers to negotiate practical roaming agreements with other mobile carriers in Australia.
This principle applies equally to SMS services and mobile voice services.
7.1.4 IP addressed communication services (incl. VoIP and email)
• The principle of AAC should be imposed on ISPs following determination of the communication service
Source: Peter Gerrand
An increasing volume of voice communication services are being carried over IP-based infrastructure, whether the voice calls start or terminate on conventional telephone connections or on personal computers or similar devices.
A simple test as to whether such a call takes place as part of a public communication service is to ascertain if the call has been directed to a number or address within a public network numbering or addressing plan.
If the call is directed to a national or international telephone number, the call will require connection via a public network service, even if the majority of the supporting network infrastructure is IP-based. In these cases, the requirements of AAC must be met by the participating service providers.
If the call is directed to an address within a private network, e.g. a Skype user name, then the service is a private communication service.
If the call is directed to an IP address, the test lies in whether the IP address corresponds to a public website (supporting public VoIP services, e.g. by automatically forwarding voice calls to an audio information source or to a public telephone number) or to a private IP address. In the first case, the address is by inference supporting a public communication service, and the AAC obligation applies. This case is dealt with at greater length in the following section.
Email Services
The principle of AAC for email should be imposed on Internet Service Providers, with the exception of email addresses on Internet addresses that are proscribed by the Australian Government, in which case a status indicator message should be returned to the originating party, explaining the reason for the failure to transmit the email message.
The use of email has become a staple communication service amongst most Australian businesses, governments and citizens, and therefore deserves to be supported and protected as a public communication service.
Email addresses are expressed in the form username@[domain name] and will only be accepted if the website addressed by [domain name] has an associated email server.
Internet browsing
The principle of AAC should be imposed on Internet Service Providers, with the exception of website addresses (URLs) that are proscribed by the Australian Government, in which case a status indicator message should be returned to the originating party, explaining the reason for the failure to service the browsing request
Internet browsing has also become a staple communication service for Australian businesses, governments and citizens; increasingly it is becoming the preferred medium for communication between government agencies and citizens. For this reason Internet browsing, whether as a one-way (non-interactive) or two-way (inter-active) communication service, should be supported and protected as a public communication service.
7.2 Numbering and addressing
7.2.1 Management of numbering and addressing
7.2.2 Management of number portability
7.3 Access
This heading covers two kinds of access end-user access and infrastructure access.
An example of the first kind is an end user accessing a set of telecom services using customer equipment such as a phone or computer connected to a telecoms network.
An example of the second type of access is a service provider accessing the network infrastructure of another service provider, so as to provide connectivity for a given communication service. Among the many infrastructute access issues are interworking between local and backbone networks, and peering and interworking between ISPs. In recent years, vast regulatory attention has been paid to interconnection of long-distance networks to a particular customer access network (CAN) within that CAN. It is the most problematic area for inter-carrier regulatory disputes over infrastructure access for broadband service delivery over the past decade.
For further explanations of the two kinds of access, and their significance, see Peter Gerrand.
• A telecoms access regime that encourages investment
Source: Jim Holmes, Director, Incyte Consulting
This proposal sets out a new access regime which would encourage investment in next generation networks, and, in particular, in high-capacity fibre distribution systems. Those are the foundations of next generation access.
The need for a new access regime
The world has turned since 1998, when the current telecommunications access regime was devised. It forms Part XIC of the Trade Practices Act 1998. The late 1990s marked the high point of service-based competition, and policies encouraging access and sharing of bottleneck facilities and wholesale services which it was uneconomic to duplicate. The key emphasis was on access to infrastructure, facilities and services that were largely in place, with substantial sunk investment.
The tools and mechanisms, including policy frameworks, for promoting fair and reasonable access on regulated terms and conditions are not the means of encouraging investment in substantial new platforms and distribution networks.
Some aspects of the current access regime are anathema for investment, sending all the wrong signals and adding regulatory uncertainty on top of the uncertainties associated with demand, funding and inter-modal competition. The regime is based on assumptions that no longer apply, and perhaps in some respects never applied. It assumes that:
- the long term interests of end-users (LTIE) standard provides certainty of application;
- commercially negotiated interconnection and access between enterprises, some of whom are vertically integrated, is likely or possible;
- commercially negotiated interconnection and access will usually deliver outcomes superior to regulation;
- voluntary access undertakings, approved by the regulator, will emerge to provide a useful and comprehensive range of wholesale offerings; and (partly repeating the second point)
- vertically integrated enterprises will seek to develop wholesale services for mainstream competitors (that is, those who seek to be more than convenient channels to market).
It is a measure of the failure of this regime that there has only been one approved voluntary access undertaking by FOXTEL in relation to set-top boxes. There are many disapproved undertakings. It is also a measure of failure that most major operators, particularly Telstra, appear not to wish to use this facility again.
The regime was doomed in the first place. It provides a testament to wishful thinking about the way opposing commercial interests might intersect. The legacy that has emerged is unremitting and gloomy:
- no certainty
- protracted regulatory processes
- substantially increased arbitrations and delay (the cost of which reflects in permanent economic loss through deferred investment)
- substantial regulatory gaming
The essence of a new regime
The new access regime needs to move from voluntary access undertakings to mandatory ones. The difference would be that the regulator would bear the onus of justifying disapproval or non-registration, based on clear criteria.
The most important change in the law will be to move the balance from ex ante regulation to ex post regulation for the protection of competition. In other words, regulators will be looking at how access arrangements have actually worked in the marketplace; rather than assessing the likely future effect of proposed access arrangements. This will be the process for making decisions:
1. Determination of relevant markets for telecoms services
2. Designation of markets where there is satisfaction of threshold criteria favouring ex ante regulation against the potential harm from market power (or market dominance). The threshold criteria, which need to be considered together, are:
(a) whether the relevant market under consideration is subject to high and non-transitory entry barriers
(b) whether the relevant market under consideration is now or will tend over time to be subject to competitive market forces sufficient to protect the interests of subscribers; and
(c) whether ex post competition controls alone, in the absence of ex-ante regulation in the same relevant market, would likely be sufficient to address concerns related to market dominance, taking into account the particular characteristics of the relevant market under consideration.
3. If the market is one where there is one (or even more than one – though there is argument about the circumstances where that can possibly be the case) dominant operator, then the regulator may impose ex ante, asymmetrical obligations that are the minimum necessary to address the risk involved and yet be effective. One such remedy could be the need for a Reference Interconnection Offer (RIO). The Offer need be the minimum necessary, and terms that are unacceptable will be determined by the regulator and approved following public consultation.
This process reflects best practice. It is expeditious. It avoids delay, uncertainty and confusion, and is far more supportive of investment than the present regime.
Why is this important for the longer term, and not just an improvement on the present?
Some, including the author of this proposal, still have abiding faith in the development of effective and sustainable competition in many parts of the telecommunications industry, including network services. It is a different faith from that which informed the 1998 laws. The 1998 legislation seemed to envisage a comprehensive array of ex ante regulatory arrangements, possibly built on the labour of the parties themselves to a large extent. That was to be aided and abetted by a willing regulator (that is, one that wants to regulate).
The approach in this proposal is consistent with a different regime – one with minimal ex ante provisions but with a coercive approach. Voluntary arrangements are for commercial dealings, not for regulation. Over time, the threshold of what is minimal in all the circumstances may change. It may not wither on the vine; conceivably it could grow and reflect the increased complexities of a converged market and convergent underpinning technologies.
• Termination Charges: Bill and Keep
Source: Jim Holmes, Director, Incyte Consulting
Termination charges are the compensation paid to an access provider for terminating a call made from a service directly connected to another network operator to a service directly connected to its network.
The access provider is required to terminate all voice calls made to services directly connected to its network. This is the any-to-any connectivity principle. Without that principle network service competition would be largely untenable, since larger network operators would have strong incentives to deny interconnection to smaller or new network operators, or would extract above-cost charges for doing so.
The issues of termination charges also arise in the following circumstances in many countries, including Australia:
• Payment for interconnection is by the network operator of the calling party. The called party is not required to pay to receive calls. This allocation of cost reflects an implicit allocation of value, and the assumption is that the value associated with the call is all with the calling party.
• It is assumed that the call is caused by the actions of the calling party and consequently, on the principle of cost causation; that it is the calling party who should be responsible for paying for those costs. This assumption is open to considerable challenge, given the means that are open to a called party to control disclosure of his/her number, ability to manage calls with the functions such as caller ID display associated with modern terminal equipment, and the ability to control the duration of a call once answered.
Problems associated with termination charges
There are a number of problems associated with termination charging regimes, including:
• Continuing regulatory intervention: The history of inter-operator commercial negotiation over termination charges is not inspiring. Operators seldom agree on the rates that should apply. Invariably they seek to obtain advantage over each other in terms of paying less than requested for termination on the networks of others, and seeking above-cost (leaving aside the meaning of the term) outcomes for themselves. The incentives for operators to behave in this way are very strong. Termination charges involve the ability to transfer costs and inefficiencies to a competitor, and to determine a significant element of the competitor’s costs. It is therefore small wonder that regulators are invited to intervene to arbitrate on these matters or the set interconnection charges directly. The opportunity and incentives for gaming in the arbitration and charge-setting processes are also significant. The result is that interconnection disputes and related resolution processes involve substantial uncertainty over considerable periods of time, and consume significant resources from operators and regulators alike.
• Cost standard: Regulators are bound to nominate the criteria on which they will determine interconnection charges. The current best practice is to do this on the basis of costs. Cost-based interconnection charging is set out in Section 2.2 of the Reference Paper attached to the WTO Telecommunications Basic Services Agreement to which most countries have acceded3. The cost standard is therefore authoritative. This in itself is not a problem. The specific cost standard that has been adopted – long run incremental costs of an efficient operator using forward looking technologies – is more troublesome. This standard is particularly appropriate to situations where a party has a choice between building a facility and buying access to a facility owned by another. The LRIC (long run incremental cost) standard is particularly useful in determining the fulcrum point at which a potential user of an access service is indifferent. However, unlike situations involving access to essential faculties (those which it is not economic to duplicate) there is no possibility of building in the case of call termination. The called party is on a specific network and calling him/her involves termination by the operator of that specific network. Building another network is not an option. The LRIC cost standard, so useful in other contexts, is inappropriate in this one.
• Misdirection of commercial focus: Operators spend considerable time, effort and focus on feeding from each other – on attempting to win interconnection disputes. They have every incentive to do so. An interconnection win has double benefit: in harming a competitor through imposing greater costs than could have resulted; and in obtaining greater revenue than might have been. Ideally the creative energies and resources of operators should be directed to providing service to their retail and voluntary wholesale customers.
• Retail price impacts: High termination charges underpin high retail charges, and are typically factored into retail charges. This is the case even though termination charges are only paid on the balance of interconnected traffic between networks. The economic benefit of improved allocative efficiency through greater consumer surplus is deferred and therefore lost.
Solution
The whole procedure associated with mock-negotiation and inefficient-resolution of interconnection matters needs to be removed. The matters need to be remitted to regulators to determine with a very clear remit to reduce termination charges as far as they might be reduced consistent with the allocation of value of calls between the parties involved. Invariably this will result in either no termination charges (Bill and Keep, or BAK) or near-BAK.
Near-BAK has some advantages over BAK, including:
• Provides for some compensation where the traffic is imbalanced
• Avoids the ‘hot potato’ issue of early handover in situations where this maximises the transmission costs of one operator at the expense of the other
• Reduces (although does not eliminate) the critical issue of location and number of points of interconnection between networks
There are some disadvantages relative to pure BAK, including:
• The need for monthly or periodic billing, and the possibility of billing disputes on a regular basis
• Setting the near-zero figure that is sufficient to provide a disincentive for hot potato traffic handover behaviour.
3World Trade Organisation, Telecommunications Services Reference Paper, 24 April 1996, see: http://www.wto.org/english/tratop_e/serv_e/telecom_e/tel23_e.htm
« Back
Network Insight sponsors include:
