Tuesday, November 30, 2010

Software through Iteration

Steven FraserGreetings from the Australian National University where Steven Fraser, Head of Cisco Research is speaking on "Best Practices: Tales of Adoption and Agility through Iteration". He is starting back in the early days of software engineering with the first NATO Software Engineering Conference and Fred Brooks.

The seminar is taking place in the Computer Science and IT Common Room. This is a curious room with two doors: one marked CS Staff Room and the other CSIRO Staff Room, but leading into the same shared room.

Steven discussed ways to encourage the adoption if best practice in software development. Curiously, given that he was talking at a unviersity, he did not mention the role of education. He then went on to discuss CISCO's work with universities on research. CISCO has $5M for "gifts" to researchers, typically for small projects with a couple of students (4% currently goes to Austrlaian projects). CISCO issues RFP-Based Research Proposals. I noticed there is an RFP for "Green" including "Business Sustainability in the Context of Energy Efficiency: organizational and business aspects of energy conservation and sustainability".

Software "Best" Practices: Tales of Adoption and Agility through Iteration

Steven Fraser (Cisco Research Center)


DATE: 2010-11-30
TIME: 16:00:00 - 17:30:00
LOCATION: CSIT Common Room, Level 2, CSIT Building
CONTACT: Elisa.Baniassad@anu.edu.au

Software "best" practices have been increasingly embraced through a mix of enthusiast passion and practical need. Agile practices no longer have the luxury of "choosing" small non-mission-critical projects with co-located teams as first suggested by early practitioners. This talk will outline a variety of "best" practices and processes to share observations on approaches and adoption.

Steven will also address "Collaboration Opportunities with Cisco"

Note that "afternoon tea" (inc. wine, beer and nibbles) will be provided.

Steven Fraser is the Director of the Cisco Research Center (http://www.cisco.com/research) in San Jose California with responsibilities for developing university research collaborations and facilitating technology transfer between researchers and Cisco Business Units. Previously, Steven was a member of Qualcomm's Learning Center in San Diego, California enabling technical learning and development in software engineering best practices. Steven held a variety of technology roles at Bell-Northern Research and Nortel including Process Architect, Senior Manager (Global External Research), and Design Process Advisor. In 1994, he was a Visiting Scientist at the Software Engineering Institute (SEI) at Carnegie Mellon University (CMU) collaborating on the development of team-based domain analysis (software reuse) techniques. Fraser was the XP2006 General Chair, the Corporate Support Chair for OOPSLA'07 and OOPSLA'08, Tutorial Chair for both XP2008 and ICSE 2009 and co-Publicity Chair for XP2010. With a doctorate in Electrical Engineering from McGill University in Montreal, Fraser is a member of the ACM, a senior member of the IEEE, and a frequent impresario of conference panels and special events. (For Steven's publications, see http://www.informatik.uni-trier.de/%7Eley/db/indices/a-tree/f/Fraser:Steven.html)

Teaching Skills for Open Government

After the excellent seminar "Using Creative Commons in the Public Sector" in Parliament House last week, I am rethinking the design of my Electronic Data Management Course at the ANU. My attempts to define skills for an got bogged down in the details of the UK Government Knowledge and Information Management Professional Skills Framework, SFIA, and the ASA/RMAA Statement of Knowledge for Recordkeeping Professionals. Instead I perhaps can look at what the Australian Government is aiming to achieve and then teach what is needed to achieve that.

On 16 July 2010, Lindsay Tanner, Minister for Finance and Deregulation, issued a Declaration of Open Government. This identified three key principles:
  • Informing: strengthening citizen’s rights of access to information, establishing a pro-disclosure culture across Australian Government agencies including through online innovation, and making government information more accessible and usable;
  • Engaging: collaborating with citizens on policy and service delivery to enhance the processes of government and improve the outcomes sought; and
  • Participating: making government more consultative and participative.
Changes to Freedom of Information and the creation of the Office of the Australian Information Commissioner came into force on 1 November 2010.

The Intellectual Property Principles for Australian Government agencies" were changed by the Attorney General's Department, 1 October 2010:

11. (a) Agencies should encourage public use and easy access to material that has been published for the purpose of :

  • informing and advising the public of government policy and activities;
  • providing information that will enable the public and organisations to understand their own obligations and responsibilities to Government;
  • enabling the public and organisations to understand their entitlements to government assistance;
  • facilitating access to government services; or
  • complying with public accountability requirements.

11.(b) Consistent with the need for free and open re-use and adaptation, public sector information should be licensed by agencies under the Creative Commons BY standard as the default.

Government Information Licensing Framework (GILF) adopted by the Queensland Government:
The Government Information Licensing Framework (GILF) makes it easy for people who use public sector information (PSI) to understand the rights of use associated with the material.

GILF comprises:

What cloud computing really means

Ian FosterIan Foster (Argonne National Laboratory) will speak on "What the cloud *really* means for science" at the CSITO ICT Centre, Australian National University, 2pm 9 December 2010:Link

LinkWhat the cloud *really* means for science

Ian Foster (Argonne Labs)


DATE: 2010-12-09
TIME: 14:00:00 - 15:00:00
LOCATION: CSIT Seminar Room, N101, CSIT Building, North Road, Acton, ACT
CONTACT: tom.rowlands@csiro.au

We've all heard about how on-demand computing and storage will transform scientific practice. But by focusing on resources alone, we're missing the real benefit of the large- scale outsourcing and consequent economies of scale that cloud is about. The biggest IT challenge facing science today is not volume but complexity. Sure, terabytes demand new storage and computing solutions. But they're cheap. It is establishing and operating the processes required to collect, manage, analyze, share, archive, etc., that data that is taking all of our time and killing creativity. And that's where outsourcing can be transformative. An entrepreneur can run a small business from a coffee shop, outsourcing essentially every business function to a software-as-a- service provider--accounting, payroll, customer relationship management, the works. Why can't a young researcher run a research lab from a coffee shop? For that to happen, we need to make it easy for providers to develop "apps" that encapsulate useful capabilities and for researchers to discover, customize, and apply these "apps" in their work. The effect, I will argue, will be a dramatic acceleration of discovery.
Ian Foster is Director of the Computation Institute, a joint institute of the University of Chicago and Argonne National Laboratory. He is also an Argonne Senior Scientist and Distinguished Fellow, Chan Soon-Shiong Scholar and the Arthur Holly Compton Distinguished Service Professor of Computer Science.

Ian received a BSc (Hons I) degree from the University of Canterbury, New Zealand, and a PhD from Imperial College, United Kingdom, both in computer science. His research deals with distributed, parallel, and data-intensive computing technologies, and innovative applications of those technologies to scientific problems in such domains as climate change and biomedicine. Methods and software developed under his leadership underpin many large national and international cyberinfrastructures.

Dr. Foster is a fellow of the American Association for the Advancement of Science, the Association for Computing Machinery, and the British Computer Society. His awards include the Global Information Infrastructure (GII) Next Generation award, the British Computer Society's Lovelace Medal, R&D Magazine's Innovator of the Year, and an honorary doctorate from the University of Canterbury, New Zealand. He was a co-founder of Univa UD, Inc., a company established to deliver grid and cloud computing solutions.

Monday, November 29, 2010

Superfast Broadband Not Worth a Subsidy

Several media outlets, including the ABC have reported that the study "Superfast: Is It Really Worth a Subsidy?" calls into question the value of government subsidising high speed broadband projects, such as Australia's National Broadband Network. This is just a working paper, but it presents a well reasoned case:

Governments around the world are investing multiple billions to support the roll-out of fiber to enable high speed broadband. These subsidies are based on the premise that fiber to the home (FTTH) brings substantial externalities. It is argued that FTTH will support economic growth and is key to national competitiveness; that it will benefit education, healthcare, transportation and the electricity industry; and that it will be the TV platform of the future.

In this paper we argue that the evidence to support these views is surprisingly weak, and that there are several errors that are made repeatedly when making the case for FTTH. In particular:

  • The evidence that basic broadband contributed to economic growth is decidedly mixed, and some of the studies reporting greater benefits have significant flaws
  • Time and again, data that basic broadband brings certain benefits is used to justify investment in fiber – but the investment in fiber must be based on the incremental benefits of higher speed, since (in the developed world) there is already near universal basic broadband
  • This error is compounded since other high speed broadband infrastructures (such as cable, and in time wireless) are often simply ignored when making the case for fiber
  • Fibre is credited with bringing benefits that would in fact require major systems and social change in other parts of the economy, such as a widespread shift to home working, or remote medical care. In practice, these changes may never happen, and even if they do they will have significant additional cost beyond simply rolling out fibre
  • Frequently business or government applications, such as remote medical imaging, are used to make the case for FTTH. But these applications require fiber to certain major buildings, not to entire residential neighborhoods (and these buildings often have high speed connections already)

We do not argue that there is no commercial case for rolling out fiber, nor do we argue that fiber brings no societal benefits. But we do believe that those benefits have been grossly overstated, and that therefore, particularly in a time of tight budgets, governments should think very hard indeed before spending billions to support fiber roll-out. A decade ago telcos wasted billions of shareholders’ money on telecoms infrastructure that was well ahead of its time – governments are now in danger of doing the same with taxpayers’ money.

From: Executive summary, Superfast: Is It Really Worth a Subsidy?, by Charles Kenny and Robert Kenny, Communications Chambers, Sat 27 Nov 2010 12:08:02 EST

What is programming

Dr Guy SteeleDr. Dick GabrielDr Richard P. Gabriel and Dr Guy Steele will present "50 in 50: 50 Programming Languages in 50 Years", at the Australian National University in Canberra, 9 December 2010. To register for pre-lecture drinks (commencing 5.15pm). Please respond by Tue 7 December:

ANU College of Engineering & Computer Science and the Australian Computer Society (ACS) present:

ANU Public Lecture Series 2010

50 in 50: 50 Programming Languages in 50 Years

Dr Richard P. Gabriel and Dr Guy Steele

Thursday 9 December 2010, 6-7pm

Haydon-Allen Lecture theatre (the tank) building 23, University Avenue, ANU
This lecture is free and open to the public.
Enquiries t: 6125 6601
E: Heather.mcEwen@anu.edu.au

People keep inventing new programming languages. What is programming, and how can the design of a programming language help or hinder that process? We have learned a lot over the last five decades: principles, conventions, theory, fashions, and fads. As George Santayna stated, “those who cannot remember the past are condemned to repeat it.”

In this multimedia presentation, the presenters survey numerous design features and important lessons from the past that future programmers - and future programming language designers - ought not to forget. They illustrate each lesson by discussing specific programming languages of the past, and endeavour to shine what light we can on the future.

Dr Guy Steele is a software architect who researches programming languages at Sun Labs, Oracle. He is the co-author of several books on programming languages, including The Hacker’s Dictionary, and Common Lisp the Language. He is a Fellow of the American Academy of Arts and Sciences, a member of the National Academy of Engineering and a Fellow of the Association for Computing Machinery.

Dr Richard Gabriel is a distinguished engineer working on programming language and software engineering at IBM Research. He is the author of five books on various topics including Lisp benchmarking, software and patterns, creative writing, a poetry collection and software development.

Please note there is limited availability of parking at ANU and guests should consider alternative forms of transport or allow sufficient time to find appropriate parking. For a map of the ANU campus visit http://campusmap.anu.edu.au To join the ANU Public Lectures mailing list and to listen to selected podcasts,
visit www.anu.edu.au/publiclectures

The views expressed in this lecture are those of the presenter and do not necessarily represent the views of The Australian National University. Printed on recycled paper. CRICOS# 00120C MO_10363

Sunday, November 28, 2010

Proposed F-35D Long Range Variant

Proposed F-35D Long Range VariantThe Australian Government selected the Lockheed Martin F-35 Lightning II single-seat, single-engine stealth multirole fighter to replace the F-111C and F/A-18 Hornet. While a capable aircraft, the F-35 does not have the range of the F-111C. This will be made up for, to some extent, by the use of in flight refuelling, but a larger aircraft would be desirable. It would be feasible to produce a twin-seat, twin-engine, long range derivative of the F-35.

The F-35 "Joint Strike Fighter" was planned in three variants: F-35A conventional takeoff and landing (CTOL), F-35B short takeoff and vertical landing (STOVL) and F-35C carrier aircraft. A fourth variant was added in 2010, the F-35I for Israel, derived from the F-35A.

It would be feasible to add a fifth variant, the F-35D, derived from the F-35C, which has larger wings and tail control surfaces than the F-35A and B. The proposed F-35D Long Range Variant would differ from the F-35C by having a larger fuselage and wings, twin engines and a crew of two. The range of the F-35D would be three times that of the F-35C with the same weapons payload.

The F-35D development would be relatively rapid, low cost and low risk, compared to a completely new design. The F-35D would follow the approach taken with the successful development of the F/A-18E/F Super Hornet from the F/A-18C/D Hornet. More commonality with the existing aircraft would be possible than was the case in development of the F/A-18E. The sensors, computers and other electronic components of the F-35 could be used unchanged. Some components of the F-35 would be duplicated for the F-35D, such as the engines and cockpit instruments. Other components would be scaled up, such as the landing gear and the fuselage.

The Australian air-force requires only 24 long range aircraft to replace the F-111C and development of the F-35D could not be justified for this small production run. However, the US Air Force is investigating options for a Next-Generation Bomber. The F-35D would meet the requirements for such a medium bomber. It would have advantages in commonality of training and maintenance with the F-35.

In 2006 the Australian Department of Defence invited participation by Australian researchers in the F-35 project. One area for research could be for the Australian government to fund a small study into the feasibility of the F-35D and see if this sparked interest by the US Government.

Saturday, November 27, 2010

NBN Business Case Summary

On 24 November 2010, the Prime Minister announced that the NBN Business Case would be released. Unusually this was not provided via a government web site, but instead through several commercial media sites and a link from the ALP web site., as well as from NBN Co. This, and spelling mistakes in the document, indicate the summary was prepared with some haste.

It should be noted that the business case is not a cost benefit analysis and does not consider social costs, or benefits, of the NBN, but is a commercial style document. There are few new details in the summary, which were not already provided in previous, carefully prepared NBN Co. documents.

NBN Co. have a technically sound plan for building a national fibre optic network, with supplementary wireless for regional areas and satellite links for very remote areas. The plan for roll-out is credible, if ambitious in terms of the rate of progress.

The major questions with the NBN are not ones for NBN Co, but for the government:
  1. Is the project of net benefit to the community? In my view, an analysis would show that it is, but no such analysis has been done.
  2. Will mobile wireless overtake the fixed fibre network? In my view, wireless networking will become the predominant method of access to the Internet, rendering the NBN obsolete before the project can be completed. The NBN infrastructure will still be of use, but like the copper network it is replacing, this will require ad-hoc adaptation to the new service. While the NBN is correctly described as a Fibre To The Premises (FTTP) network, almost all the customers will be using a wireless link to connect their devices in the home, so it could be better described as Wireless To The Customer (WTTC). The fact that there is a fibre link to their home will be irrelevant to most users, who will want to be able to walk around with their hand held information devices. The NBN service will only work in the customer's own home, which will be a major annoyance. Most customers will require a separate wireless service when using mobile devices outside their home. It is likely that many, perhaps most, will avoid the complexities of maintaining two broadband services and so not use the NBN. As a result it is likely the NBN will need to be retrofitted with a wireless service. It would therefore be better if the fibre system was engineered with this future requirement in mind. Which wireless technology will be used is unclear, but that it will require Femtocells is not.
The summary was provided as a 36 page PDF document of approximately 1 Mbyte (size varies between the different versions). The PDF version is difficult to read on screen and to download, so I have prepared a HTML version (without diagrams):

NBN Co. Business Case Summary

24 November 2010

Table of Contents


1 NBN Business Case

The NBN Business Case is based on detailed engineering, financial and business analysis undertaken by NBN Co. over the last twelve months.

The Business Case includes robust sensitivity analysis throughout the plan, which shows that the key conclusions (outlined below) as measured against the stated objectives of the project are robust to a range of assumptions.

Information outlined below has been redacted from the Business Case with the view of providing more information without comprising the market sensitive aspects of the Business Case including those that will be impacted by outstanding Government decisions, especially a decision around the final number of Points of Interconnect (POI).

The POI decision is subject to external review by the Australian Competition and Consumer Commission (ACCC), which will be concluded on 30th November. The discussion of the conclusions below outlines the range of impacts that different decisions on the POI may have on the project’s finances.

Irrespective of the decision on the POI the NBN will realise a return higher than the long term average government bond rate, the decision itself has significant implications for the structure of the NBN and the telecommunications industry.

This Business Plan sets out the key objectives and priorities for NBN Co for the three years from 1 July 2010 to 30 June 2013.

It is anticipated that the Corporate Plan will be updated at least once a year.

The Corporate Plan is a critical part of the process of designing, building and operating the National Broadband Network (‘NBN’) and achieving the Government’s objective of providing affordable superfast broadband to all Australians through structural reform in the Australian telecommunications industry.

The Corporate Plan is prepared in accordance with the requirements of the Commonwealth Authorities Companies Act 1997 and Governance Arrangements for the Commonwealth Government Business Enterprises (June 1997).

2 Business Environment

Current forecasts anticipate that there will be more than 50 million end users connected by fibre to the premise (FTTP) technologies worldwide by the end of 2010 compared to 43 million at the end of 2009, mostly in Asia.

The key observations emerging out of the experience from global fibre deployments include;

  • Government policy plays an instrumental role in overall fibre development;
  • Open access network models are seen as being important in driving fibre competition at the retail level, with copper access seekers needing fibre wholesale products and a clear migration path to fibre;
  • Retail pricing structure for fibre products is based around bundled (cheap or free) voice, fast broadband access and multi-channel TV. Retail price relativities are determined by broadband speed, broadband usage limits, and size and quality of the overall package bundle;
  • Cost effective fibre builds and connections remain a continuous focus for operators; and
  • fibre development is a long-term play with competition from cable operators to remain strong in markets where cable is entrenched.

2.1 The role of NBN Co in transforming the Australian telecommunications industry

The establishment of NBN Co and the roll-out of the NBN is a key element of the Federal Government’s plan to transform the Australian telecommunications market.

The NBN will deliver a significant, once in a generation restructuring of the industry, resolving the current infrastructure and investment bottlenecks that have seen Australia fall behind its international peers.

The rollout of a national high-speed broadband network, via fibre and wireless and satellite, will enable greater competition, services and innovation for consumers at the retail level. This is expected to lead to a new industry structure, entrants, positioning, pricing & demand drivers, and revenue and profit shares.

The NBN will also facilitate a major reallocation of capital in the telecommunications industry, which has historically been dominated by high Public Switched Telephone Network voice revenues.

The decline of voice revenue is already underway, and is expected to accelerate with the advent of high quality Voice over Internet Protocol (VOIP) under the NBN. However, declining voice revenues are expected to be substituted by increasing broadband revenues, as business models continue to shift from toll calling charges to access charges.

In addition, the significant investment by Internet Service Providers in Digital Service Line Access Multiplexers (DSLAMs) and other equipment in order to provide broadband services over copper is expected to be redirected to content, service differentiation and value added services over the NBN, fuelling the development of new applications and innovation that will drive consumer demand.

As a wholesale provider of services with no participation in the retail market, NBN Co is set to address the problems currently arising from the primary infrastructure owner and wholesale access provider also holding the position of market share leader in almost all segments of the industry. This separation should in time allow a simplification of the regulatory regime and greater certainty for industry participants at all levels.

NBN Co will provide Layer 2 wholesale services only, providing flexibility to support a range of wholesale and retail business models. Larger retail service providers are expected to acquire Layer 2 products from NBN Co and use their own infrastructure to provide retail services to their customers. Smaller retail service providers may opt to use a Layer 3 intermediary for incremental wholesale services. The diversity of possible business models is expected to result in lower barriers to entry for retail service providers and to open up competition both in the major population centres and in regional areas.

3 Key Objectives

The Government has stated its broad objectives for the NBN as follows: “The new superfast network will: Connect homes, schools and workplaces with optical fibre (fibre to the premises or “FTTP’), providing broadband services to Australians in urban and regional towns with speeds of 100 megabits per second – 100 times faster than those currently used by most people extending to towns with a population of around 1,000 or more people;

  • Use next generation wireless and satellite technologies that will be able to deliver 12 megabits per second or more to people living in more remote parts of Australia;
  • Provide fibre optic transmission links connecting cities, major regional centre and rural towns;
  • Be Australia’s first national wholesale-only, open access broadband network;
  • Be built and operated on a commercial basis by a company established at arm’s length from the government and involve private sector investment; and
  • Be expected to be rolled-out, simultaneously, in metropolitan, regional and rural areas.”
  • To design, build and operate the broadband network required as the foundation of the Government’s NBN policy the Government established NBN Co on 9 April 2009 as a Company under the Commonwealth Authorities Companies Act, 1997.

NBN Co’s objectives can be summarised as:

  1. The network should be designed to provide an open access, wholesale only, national network, covering all premises;
  2. The technologies utilised should be fibre to 93 per cent of premises (including Greenfields developments), fixed wireless to 4 per cent of premises (delivering at least 12Mbps), and satellite to 3 per cent of premises;
  3. The pricing principles to be employed should ensure uniform, national wholesale pricing accessible on non-discriminatory terms; and
  4. The network expected rate of return should be in excess of current public debt rates.

NBN Co has designed a layer 2 network that satisfies the four objectives set above.

Once NBN Co’s Corporate Plan is approved by Government (in this first year of operation NBN Co’s Corporate Plan primarily relates to the design of the network,) NBN Co will commence full-scale build and operation of the network.

3.1 Achievability of the Objecties

The major objectives and timeline are dependent on the rapid resolution of Government policy dependencies as well as the progress of the negotiations with Telstra, all of which appear to converge towards the end of the 2010 calendar year.

The charts below show the coverage under each technology.

93% Fibre Coverage

[Diagram omitted]

93% Fibre + 4% Wireless

[Diagram omitted]

93% Fibre + 4% Wireless + 3% Satellite

[Diagram omitted]

4 Timeline and Critical Dates

There are four distinct phases in the early delivery of the National Broadband Network:

  1. Establishing NBN Co;
  2. Planning and Design;
    1. Products & Pricing;
    2. Network Design & Testing;
  3. Construction; and
  4. Commercial Operations.

4.1 NBN Co has identified five critical dates in the overall program plan for FTTP:

December 2010Complete Telstra negotiationsFinalise agreements with Telstra on decommissioning of copper and hybrid fibre coaxial (HFC) networks and infrastructure usage; subject to Conditions Precedent.
April 2011Start Customer TrialCapability to connect at least one mainland based retail service providers (RSPs) with trial customers offering a free subset of products to test preparedness. All NBN Co support with manual processes.
June 2011Completion of Telstra Definitive AgreementsAll Conditions Precedent satisfied, including enabling legislation, ACCC and Telstra shareholder approvals.
September 2011Ready for First Commercial ServiceCapability to fulfil, activate and assure a limited number of products with multiple RSPs and up to 6 per cent of premises passed. Supported with a combination of basic semi-automated and manual processes.
December 2011Ready for Business as Usual Roll-outCapability to fulfil, activate and assure an increased number of products with multiple RSPs and up to 19 per cent of premises passed. Supported with a combination of advanced semi-automated and manual processes.
August 2012Ready for MarketFully automated systems, no limitation in activating as a percentage of premises passed. Multiple RSPs certified; critical volume available and predictable.Operations capability can fulfil and assure the NBN Co suite of products at scale.

The period to August 2012 will be focused on the establishment of key systems required to support the roll-out of the NBN. This includes detailed testing of NBN Co’s network design and construction methodologies through the establishment of a test lab, the early roll-out of FTTP in twelve First Release Sites (five mainland First Release Sites and seven sites in Tasmania). It also encompasses completion of essential support systems - including operational and business support systems and enterprise resource planning - that need to be in place before full scale customer acquisitions can commence.

NBN Co is currently scaling up the FTTP deployment from the First Release Sites to full-scale network construction. It is anticipated that, following evaluation of the lessons learned from the First Release Sites, a series of ‘Second Release Sites’ already announced (14 new sites on the Australian mainland in addition to five first release sites), will be used to refine construction methodologies and systems once critical support systems are in place and operational. Based on current timetables, detailed design of the Second Release Sites will commence in November 2010, with subsequent construction commencing in February 2011. The Volume Roll-out construction will commence in June 2011.

During FY2011, NBN Co will also be finalising the design and accelerating the deployment timeframe of its wireless and satellite solutions for the ‘Last 7 per cent’. NBN Co is currently progressing negotiations for wireless spectrum acquisition, which are expected to close before the end of 2010. After a simplified procurement process, the Company expects construction of the main wireless network to start in November 2011, following a series of proof of concept and First Release sites aimed at finalising key decisions around spectrum and wireless network build options.

The long lead times in satellite construction and launch mean that NBN Co does not expect to have its own satellites in orbit until FY2015. However, the Company expects to be able to offer an interim satellite solution, called Satellite First Release Sites, from June 2011 using spare capacity on existing satellites in order to provide continuity to the existing Government Australian Broadband Guarantee program expected to cease on 30 June 2011.

4.2 Annual Updates

The assumptions made by the Company, which underpin the Plan, together with the business strategies and development of capabilities of the business, how the Company will measure its achievement of the financial and operational objectives, and the risk management and its mitigation strategies, will be reviewed on a regular basis to take into account the latest major developments; it is anticipated that the Plan will be updated at least once a year.

5 Telstra Deal

The Financial Heads of Agreement between Telstra and NBN Co paves the way for a faster, cheaper, more efficient rollout of the National Broadband Network.

Taxpayers benefit from the deal with Telstra because it reduces the overall cost of building the network and will result in higher take-up rates and revenue for NBN Co.

It also means that a greater proportion of the NBN rollout will be underground, with less overhead cabling.

It would mean Australia’s largest telecommunications company, Telstra, will become a participant in the rollout of the NBN. Telstra will become NBN Co’s largest suppliers of infrastructure and is likely to become NBN Co’s largest customer.

As no binding agreement has yet been entered into with Telstra, the Business Model includes extensive analysis of NBN Co’s No Deal scenario (which is a scenario set out in Business Model) and a comparison between the two cases.

The Corporate Plan is predicated on the execution of Definitive Agreements that give effect to the Financial Heads of Agreement with Telstra. There are two major risks with regards to this transaction:

Delay in completion: The Plan assumes that the definitive and binding agreements will be signed by end of 2010, with completion and satisfaction of all the Conditions Precedent by 30 June 2011. Any delay will impact NBN Co’s ability to finalise its network design (deployment targets, transit backhaul, fibre access network (FAN) sites and POIs locations), and therefore the Company’s ability to achieve the Plan deployment targets; and

Unwinding of the Financial Heads of Agreement: The Financial Heads of Agreement are the basis for negotiation and, if agreed, the terms of the Definitive Agreements. The terms of the commercial schedules are not legally binding. No legal obligation arises unless and until all of the Definitive Agreements are executed by the parties. Whilst negotiations and drafting of the Definitive Agreements are progressing well, there are no legal obligations on either party to agree and sign binding documentation other than to negotiate ‘in good faith’.

In addition, the Definitive Agreements will be subject to satisfaction of Conditions Precedent, including the dates by which the Conditions Precedent must be fulfilled; these include due diligence, shareholder approval as well as enacting of legislation and regulatory and other matters between Telstra and Government. In the event of the Definitive Agreements not taking effect, there will be significant impact to the implementation of the NBN and NBN Co’s Business Model; the period required to achieve full deployment will also need to be extended.

5.1 ACCC Special Access Undertaking Process

NBN Co’s announced intention is to utilise a special access undertaking approved by the ACCC to determine pricing for NBN.

The special access undertaking cannot be finalised and lodged with the ACCC until key policy matters such as the number and location of POIs and pricing of products and services are finalised.

NBN Co considers that it would be preferable if it did not lodge its special access undertaking until the CCS Bill and NBN Companies and Access Arrangements Bills are passed. These Bills contain amendments which affect the way NBN Co would operate and the powers of the ACCC in relation to the special access undertaking.

NBN Co is planning lodgement of the special access undertaking as soon as feasible after the Bills are passed.

The ACCC is subject to a statutory timeframe of six months to accept or reject the special access undertaking, subject to extensions of time and ‘stop clocks’ (for example while it is waiting for information it may request from NBN Co) NBN Co is actively engaging with the ACCC in developing the special access undertaking, including to try and reduce the likelihood of an extended consideration process once the finalised special access undertaking can be lodged.


6.1 Product definition and Pricing

NBN Co has developed a rigorous process to ensure an attractive product set is offered to meet market demand.

NBN Co will also provide Uniform Wholesale National Pricing across fibre, wireless, and satellite technologies.

6.2 Products

The NBN Co product set will be offered as follows:

  • A uniform product construct across fibre, wireless and satellite, featuring the same four product components across each access network, and based on the technology-agnostic Ethernet Bitstream framework.
  • A 12Mbps downstream and 1Mbps upstream entry-level offer across all three access technologies (i.e. fibre, wireless & satellite), at the same price (network ubiquity).
  • An initial fibre product suite with committed speed options of up to 100Mbps and peak speed options of up to 1Gbps (performance certainty and speed).

6.2.1 Product release road-map

The NBN Co product Roadmap & Release Schedule

The Product Release roadmap commences in April 2011 building on core capability and value proposition with the introduction of the high speed broadband and telephony capability. The product set then evolves with 4 additional feature and functionality ‘drops’ creating a strong value proposition for Retail Service Providers to address consumer through to high-end business markets.

Product Drop One: High Speed Broadband & Telephony

Product drop one delivers a core set of NBN Co Fibre Access Service features to the market. It is intended to enable Retail Service Providers to “On-Board” and commence interoperability testing of NBN Co services. The Retail Service Provider will be able to offer high-speed internet packages with access speeds of up to 100 Megabits per second and telephony capability. This will enable a range of migration scenarios from existing broadband and telephony plans.

Product Drop Two: Emerging Entertainment Capability

Product drop two enables Retail Service Providers to build a triple play offering, by adding the capability to distribute their content (eg. television channels) simultaneously to two or more end-users as a single stream. This feature will greatly enhance a Retail Service Provider’s ability to deliver internet protocol television offerings. This multicast technique can achieve significant bandwidth savings when delivering the same content to many end users. NBN Co expect to see retail market innovation based on these features, particularly around the delivery of a “triple play” offering, eg. high speed internet, telephony and television channels for real time viewing.

Product Drop Three: High Speed Business Services

Product drop Three provides incremental functionality to support innovation in the small-to-medium-business market. These additional features are specifically designed to connect multiple business locations, so that they interact seamlessly and provide secure, high speed and reliable access to various business applications including video collaboration and conferencing.

Product Drop Four: High Speed Enterprise Services

Product Drop 4 delivers very high speed up to 1 Gigabit per second for high- end enterprise services. It will also provide incremental business-oriented functionality including point to point links and transparent VLAN. Enhanced Respond and Repair Service Levels in support of mission critical applications are targeted for this release.

Product Drop Five: Enhanced Reliability for Mission Critical Sites

Product Drop 5 delivers the final features to complete the initial NBN Co Fibre Access Service feature set. It includes access diversity for end-users. This feature enables mission-critical sites, such as hospitals, to achieve higher network availability and reduce their risk of outages. Additional operational capabilities will allow Access Seekers to provide greater reliability to their business end-users.

6.3 Uniform National Wholesale Pricing

One of the objectives Government has set for NBN Co. is “uniform national wholesale pricing.” This means that retail service providers (RSPs) will have to pay the same amount, irrespective of where they are based, to deliver their service to households and business throughout Australia.

RSPs currently have to pay for two “products” in order to reach households: (1) “access” through a network – currently primarily copper but in future fibre – and (2) “backhaul” to carry content from the RSP’s headquarters to a “point of interconnect” (PoI) with the access network. NBN Co sees it as part of its mission to make the sum of these two costs low and uniform.

If PoIs are served by only one backhaul provider, that provider can therefore charge very high prices to any RSP seeking access to the set of households served by that particular PoI. This greatly limits competition, and the capacity to innovate at the RSP level.

NBN Co’s recommended solution to this problem is to establish a small number of PoIs located in capital cities. In our business model there are 14. Since there would be competitive backhaul available between any RSP’s headquarters and all of these 14 PoIs, the cost of backhaul would tend to be both modest and uniform.

If the Government elects to mandate a larger number of PoIs than NBN Co plans, it will need to take other steps to achieve low and uniform national wholesale pricing. If that solution does not promote the same intensity of retail competition as the 14-PoI offering, NBN Co projects that its IRR could decline by 50-80 basis points because of slower take-up of broadband and slower introduction of retail services that require higher speeds.

NBN Co plans to charge the same amount for its basic access product – 12/1 Mb/s – across all three technologies: fibre, fixed wireless and satellite.

NBN Co will also charge whole uniform national prices at higher speed tiers across its entire fibre footprint, which will be serving 93% of premises.

6.3.1 Prices
The pricing policy is based on wholesale price levels allowing to achieve comparable or better retail prices than current market for equivalent performance.

NBN Co pricing points will allow any retail provider to offer substantially higher quality products (i.e. much faster speeds), at a highly competitive prices in today’s market.

The pricing structure and pricing levels have been set to achieve a viable internal rate of return (IRR) based on NBN Co’s estimates of take up of different speed tiers and connectivity capacity usage.

Based on take up and speed usage growth assumptions, NBN Co. anticipates being able to reduce real prices for all products and nominal prices for all products, except the basic service offering, while maintaining an internal rate of return above the Government long-term bond rate.

6.3.2 Internal Rate of Return

The project returns, on an unlevered basis, are expected to exceed the long term government bond rate.

This is based on a number of assumptions, the most significant of which are growth in speeds and demand and hence revenue. The stated internal rate of return is also dependant on the completion of the Telstra deal, which has a material impact on construction costs (see discussion below).

The internal rate of return does not take account of any external benefits anticipated from the NBN to the economy, productivity or social outcomes.

The NBN Co business plan and internal rate of return assumes NBN Co’s 14 PoI solution. There may be an impact to NBN Co’s depending on the uniform wholesale price solution adopted by Government. NBN Co’s estimate is that this impact on the internal rate of return could be in the range of 50-80 basis points.

Other remaining Government decision may also have some impact on the internal rate of return.

The internal rate of return is driven by the following revenue considerations.

Stock of premises and new premises

Total premise sizing has been based on G-NAF (Geocoded National Address File) national address index, which uses multiple address sources including Government land records, Australia Post and the Australian Electoral Commission. In conjunction with work carried out by the Department of Broadband, Communications and the Digital Economy, NBN Co has assumed a starting national premise count of 10.9 million at FY10, comprising 9.6 million residential premises and 1.3 million business premises.

The forecast for premise growth has been based on forecasts of new ‘Greenfield’ (undeveloped/raw broadacre land) and ‘Redevelopment / In-fill (net of demolitions / replacement stock) dwelling households and business premises. This is based on the BIS Shrapnel report ‘Overview of the Australian Residential Market’ dated August 2010; National Housing Supply Council 2009 report, ABS data and internal estimates. NBN’s long-term forecasts have been referenced to residential household growth rates, assuming a compound annual growth rate of 1.6 per cent per annum, from Australian Bureau of Statistics forecasts.

Premises growth is driven primarily by the expected growth in residential households, resulting in an average growth of approximately 177,000 new premises each year during the forecast period (to FY2025), or average growth of approximately 166,100 to FY2040.


Another key determinant of revenue is speed. Over the course of the build the demand for faster speeds will continue to trend upward as they have in the past.

This will satisfy increased demand for new application and devices at premises.

Chart 1 shows the increase in download speeds available over broadband and that speeds available is expected to continue to increase going forward.

Chart 1: Speed Historical Profile and Market Expectations Source: Alcatel Lucent.

[Chart omitted]

NBN Co Speed Projections Exhibit 2 illustrates the projections of average subscriber speed purchased (Mbps) during the forecast period. The market expectations presented in the previous Alcatel-Lucent diagram (Chart 1) have been superimposed over NBN Co’s two alternate speed projections (Option A1 and A2). This presentation illustrates NBN’s conservative expectation that the growth in demand for speeds will be considerably lower than the extrapolation of increasing speeds implied by the history of internet access technologies.

Exhibit 2 Download Speed Historical Profile and Projections1 Source: ABS, 8153.0 Internet Activity Australia for period pre – FY10, NBN Co Option A1 & A2 Projections, Approximation of Alcatel’s Market Expectations (per Exhibit 8.20) 1 Internet Activity, Australia, Jun 2010 http://www.abs.gov.au/ausstats/abs@.nsf/mf/8153.0/

[Chart omitted]

Data usage

A key driver of revenue is that there is growth in speed leads to a new applications and therefore increased data usage over time.

Chart 3 shows the dramatic growth in data usage within Australia over the past decade.

Chart 3: Australian Data Use, Historical Profile

[Chart omitted]

Chart 4 shows a stylized progression in intensity of broadband use commencing with simple websearching tasks, requiring low broadband speeds, and progressing to include increased access to features like smartphones, gaming and television delivered over broadband.

The number of GB/month in June 2010 is an average of fixed downloads (9.2 GB/month) and wireless downloads (1.2 GB/month).

Chart 4: Trends in Broadband Use Source: NBN Co, adapted from FCC, OBI Technical Paper No4 and September Meeting Commission 2009, ITIF, CSMG

[Chart omitted]

Projections of data usage used by NBN Co have been built from the ground up and factor in the relatively strong relationship between increasing access speeds and usage The starting point has been to take the historic usage level as reported by the ABS (9GB/mth for fixed broadband) in FY2009 allowing for uploads and downloads and the long-term growth trends seen in the Australian and international markets (Cisco VNI forecasts). This underlying growth has been adjusted to reflect the large step change in speeds available to customers coming onto NBN fibre, and an adjustment has been factored in over approximately ten years to reflect the gradual development and adoption of applications that can fully utilise the increased bandwidth.

A conservative approach has been taken with regard to long-term data usage growth, factoring in considerations including saturation of usage, slowing growth in online hours and increasing delivery of content on multicast applications. This sees growth fall to ~8 per cent p.a. from FY2030, well below the baseline increase in access speeds forecast.

NBN Co has also referenced usage projections against a number of independent sources:

  • Historic ABS data, showing a 36 per cent per annum compound average growth rate from 2000 – 2010.

  • Cisco international forecasts for internet traffic to increase at a 34 per cent compound average growth rate from 2009 – 2014.

6.4 Capex

At the end of the contribution and deployment period, the total capital expenditure (capex) is estimated by NBN Co to be $35.7 billion.

This is lower than originally forecasted as a consequence of the pending deal with Telstra. This deal reduces the overall capex due to efficiencies as a result of the re-use of infrastructure and also the use of longer term leases.

The Business Model and Corporate Plan are predicated on the assumption that a deal is finalised and approved between NBN Co and Telstra in accordance with the Financial Heads of Agreement in relation to the following:

  • Progressive disconnection of copper services and decommissioning of Telstra’s fixed line copper and HFC networks as the NBN Co fibre network is rolled out;
  • Utilisation of existing Telstra exchange space;
  • Utilisation of a significant portion of Telstra’s existing ducts and conduits; and
  • Access to dark fibre and managed services for backhaul.

The deal with Telstra represents a marked improvement to the Business Model and Corporate Plan relative to NBN Co’s No Deal scenario and substantially mitigates a number risks otherwise present in the build (notably in relation to demand forecasts and the cost and speed of deploying the fibre network).

The NBN Co detailed cost modelling estimates that the NBN can be built for a total Construction Capital Expenditure of $34.4 billion to achieve targets of 8.3 million premises connected and 12 million premises passed by fibre or covered by fixed wireless or satellite by the end of full deployment date in a deal case scenario.

When adding the Replacement Capex (representing the forecast Capex reserved to maintain the deployed network) for a total of $1.3 billion, (total Construction and Replacement Capex amount to $35.7 billion).

The Agreement between NBN Co and Telstra, worth an expected value of $9 billion, in net present value terms(NPV) provides for: The reuse of suitable Telstra infrastructure, including pits, ducts, conduits, backhaul fibre and space in Telstra exchanges, by NBN Co as it starts to rollout its new network - avoiding unnecessary infrastructure duplication.

As a result, NBN Co’s forecast of the estimated capex cost to build the NBN in a deal scenario is $35.7 billion compared to $37.4 billion in a no deal scenario.

The progressive migration of customers from Telstra's copper and pay-TV cable networks to the new wholesale-only fibre network to be built and operated by NBN Co.

Under a deal scenario, total capital expenditure per premise is lower than in a no-deal scenario. This is due to the upfront savings achieved through the increased use of Telstra’s existing infrastructure.

As mentioned above, the Capital Expenditure for Connections is higher in a Deal Scenario than in a No-Deal Scenario due to the increased number of customers being connected earlier in a Deal Scenario, which translates into both a timing impact (acceleration of customers connections) and a structural impact (implying more initial build Capex to Connect, as well as active equipment, together with more replacement Capex).

As a trade-off, Opex costs are higher in a deal scenario in comparison to no deal, due to the higher use of Telstra’s infrastructure (ducts, transit backhaul links, exchanges) as well as the acceleration of customer take-up (connections) generated by the deal.

Brownfields in the Build Capex Breakdown

For the purposes of the financial model a distinction has been made between Brownfields and Greenfields premises within the FTTP footprint, due to their different cost characteristics.

FTTP Brownfields premises are considered to be the 10.1 million existing premises that are to be covered by the FTTP network at the start of the volume rollout. Greenfields premises are all those premises not currently included in the 10.9 million total premises as at June 2010.

Targets & ProjectionsFull Deployment Key Metrics (Rounded) (Nominal Dollars)
Capex$35.7 billion total Capex (of which $1.3 billion for Replacement and Maintenance and $9.9 billion for fibre connections) by June 2020.
Telstra Agreements$13.8 billion decommissioning and infrastructure payments by June 2020

6.5 Opex

As discussed above the Opex for the NBN is impacted upon by the Telstra deal Specifically Opex costs are higher in the deal scenario in comparison to No Deal, due to the higher use of Telstra’s infrastructure as well as the acceleration of customer take-up.

The main drivers of Opex costs are described below:

  • up-front migration incentives payable to Brownfield premises which connect during the main construction period. Under a Deal scenario, the cost of an upfront migration incentive per Brownfield customer is half that of the No Deal scenario.
  • Decommissioning payments payable to Telstra at the time of disconnection of subscribers as an incentive for the take up to subscribers to NBN infrastructure. Decommissioning payments on occur under the Deal scenario.
  • The lease of key infrastructure including aerial fibre, ducts, backhaul, wireless operating expenses and satellite operation expenses.

6.6 Funding

NBN Co’s funding requirement is driven by the Company’s earnings before interest, taxes, depreciation and amortisation (EBITDA) and Capex profiles, including working capital.

6.7 Equity requirement

The equity requirement from Government based on our current plan is $27.1bn.

This is based on advice from Goldman Sachs that NBN Co should be able to arrange debt funding.

Irrespective of what amounts NBN Co decides to borrow, total funding requirements will begin to decline in 2021 (approximately when the rollout is complete), because at that point NBN Co will be generating cash from its operations in excess of any capital expenditure requirements.

From 2020 to 2022, NBN Co. will be replacing equity funding with debt funding until its reaches a steady-state debt:equity ratio of 1:1.

NBN Co. expects to pay cash dividends, beginning in 2020, which in the aggregate would repay the government’s entire investment by 2034, even if no shares of NBN Co were sold to private investors. Since such dividends would be paid out of earnings, NBN Co. would continue to be appropriately capitalized and capable of being floated in the public market whenever the government chose to do so.

6.8 Cost of Capital

The corporate plan assumes that NBN Co’s cost of funds will decline as its credit ratings improve as stable revenues begin to be generated.

The risk premium is expected to decline by more than 200 basis points over the course of the build.

NBN Co has undertaken a detailed analysis of the cost of capital through the Weighted Average Cost of Capital (WACC) methodology. The WACC value has been considered in the context of the changing nature of NBN Co over time, using 4 lifecycle periods which reflect different levels of risks and volatility.

The cycles are as follows:

  • ‘Proof-of-Concept’ period, which encompasses the Corporate Plan timeline - the cost of capital would not be based on relevant comparables NBN would have the nature of an infrastructure project and an element of Australian Government broader policy in the telecoms industry;
  • ‘Construction’ period, benchmarks linked to the construction industry are used to derive a WACC;
  • Transition / Market Development’ period as risk tapers; and
  • ‘Established Network Operations’ period, is estimated reflecting the established infrastructure nature of NBN Co by that stage (while keeping a conservative risk premium compared to regulated assets such as water and gas utilities).

Based on these parameters, a capital weighted WACC has been derived at 10 per cent-11 per cent over the 30-year period. This has been factored into the business model.

NBN Co’s weighted average cost of capital over time will fall as shown in Chart 5.

Chart 5: Movement in weighted average cost of capital over time

[Chart omitted.]


NBN Co is seeking to develop a sound system of risk oversight, management and internal control and the associated culture, processes and structures to support achievement of its business objectives. Embedding a risk culture requires time and attention, as disciplines take time to form business as usual activities.

NBN Co has established an enterprise-wide risk management system to facilitate the identification of significant business risks and implementation of appropriate risk mitigation or treatment plans and monitoring processes. The system is built upon the premise that all employees have responsibility for risk management in their job areas.

The approach adopted by NBN Co is consistent with the international risk management standard, Australia/New Zealand Standard ISO 31000:2009. NBN Co’s Risk Management Policy details the way in which the Company establishes the risk context, identifies, assesses, analyses, evaluates and treats risk to effectively manage its business, assign roles and responsibilities for risk management, and establish a risk register and reporting framework to facilitate the process and meet reporting obligations.

Risk profiles will be defined and managed for each of NBN Co’s business functions and major activities. The aggregation of these profiles forms the NBN Co Group Risk profile and also forms the basis for Management and Board reporting of risks and associated management processes. Risk reporting is subject to legal professional privilege wherever appropriate. Risks, the risk mitigations in place, and treatment plans are reported quarterly to the Board under the following categories:

  • Safety and Environment;
  • Demand and Revenue (Product and Competition);
  • Technology;
  • Execution and Delivery;
  • Security (Physical, Networks, Information Technology);
  • Policy and Regulatory;
  • Resourcing - Suppliers;
  • Resourcing – People and Skills; and
  • Financial.

The Risk Management Policy includes a consequence scale to assess the impact of a risk on NBN Co’s reputation Risk management is also being integrated with NBN Co’s Program processes, with weekly recording and reporting of issues and risks that also form part of the reporting above.

NBN Co’s operating environment is dynamic, and subject to many external factors, many of which are outside of NBN Co’s control. The risk management system will play an important role in identifying and responding to risks as the operating environment changes in both the short and longer terms. Finance provides policy, strategies, tools, guidance and a cross-company reporting capability to manage risk, while additional risk specialists are distributed throughout the organisation with responsibility for addressing specific risk areas.

Whilst the execution of the risk management system aims to identify risks before they occur, for a number of reasons this is not always possible. As such, NBN Co also places an equal emphasis on risk readiness, business resilience and fostering a preparedness to respond to risk events; with the risk reporting and communication disciplines being established facilitate this. The regular risk reporting processes that have been established facilitate appropriate oversight and ownership of risks, and also serve as a mechanism for regular updating of risk registers and considering the completeness of risks identified and mitigation plans in place.

Risk management is seen as an ongoing process as part of all business activities and not a cyclical event. NBN Co seeks to foster a risk aware culture, open to risk identification and treatment of risk. This includes the promotion of education and awareness of the risk management system and key risk management principles amongst its staff.

The Risk Management Policy that establishes and underpins the risk management system will be reviewed at least annually and updated as required to reflect NBN Co’s needs. This will ensure the risk management system remains fit-for-purpose as NBN Co’s primary activities move through planning and design, to construction and commercial operation.

Wholesale and Retail Business Models

[Diagram omitted.]

[From Premise to Point of Interconnect.]

Friday, November 26, 2010

Open Access at ANU

Greetings from Parliament House in Canberra, where the seminar "Using Creative Commons in the Public Sector: Open Access to Knowledge, Culture and Public Sector Information (PSI)". Colin Steele, former ANU Librarian talked on issues with creative commons for research publishing. He expressed concerns over the effects of the new Excellence in Research for Australia scheme (ERA) on open access.

I gave a five minute impromptu talk after Colin, on how I use Creative Commons licensing for teaching. I started by explaining how ANU now accepts that course material can be released under a Creative Commons licence, as part of its open access policy. Then I detailed how I use a web search to find avialable open access material for teaching. In the case of training Australian Government employees and contractors, it is very useful to have HTML based government policy documents and manuals. This way students can work directly with the latest material. I forgot to mention how easy it is now to use web based material in e-learning systems and to create e-books. I will be teaching this in January 2011, at ANU: COMP7420: Electronic Data Management.

Australian Public Service Transitioning to Creative Commons

Greetings from Parliament House in Canberra, where the seminar "Using Creative Commons in the Public Sector" is on. Helen Daniels, Assistant Secretary, Copyright Law Branch, Attorney-General’s Department is outlining the "New Intellectual Property Guidelines and the Role of Agencies". The default will now be that government material is released with a Creative Commons "by" licence. Previously AGs administered a central register of licences for specific government documents. Under the new policy, agencies need to decide if material should not be CC, when it is created. So the AGs main task is to educate agency staff as to what to do. There is "Intellectual Property Principles for Australian Government Agencies" on the AGs web site. More material will be released in December. I hope to be able to use this material in teaching the ANU course COMP7420: Electronic Data Management from January 2011.

Electronic Records Management Needed for ABC Pool

Greetings from Parliament House in Canberra, where the seminar "Using Creative Commons in the Public Sector" is on. Sherre DeLys, Founder/Executive Producer, ABC Pool, just mentioned they have run out of storage. She said that they will now need to work out how to remove duplicated data and how to delete what was no longer needed. The ABC 2004-2007 Corporate Plan had the target to "Establish and implement a Corporation-wide records management system that meets legislative requirements.". The ABC 2007 Annual Report stated that this had been achieved. Pool may not be able to use the same corporate system, but should be able to obtain advice from the records management professionals in the ABC and use similar techniques and standards. Or they could enrol in my course in January 2011, at ANU: COMP7420: Electronic Data Management. ;-)

Wordpress plugin for museum content

Greetings from Parliament House in Canberra, where the seminar "Using Creative Commons in the Public Sector" is on. Seb Chan, Head of Digital, Social and Emerging Technologies, Powerhouse Museum, just mentioned they will be releasing a Wordpress plugin to allow metadata from the musium collection to be automatically included in blog posts. They also have been participating in the Museum Metadata Exchange (MME) project. He gave an an example of a possible mash-up would be to include historical photos from the museum with development application to the Mosman Council.

Sydney Android Rangers

Greetings from Parliament House in Canberra, where the seminar "Using Creative Commons in the Public Sector" is on. Bernard de Broglio, Internet Coordinator, Mosman Municipal Council just explained how their council rangers will be issued with Google Android smart phones next week.

Hansard in e-Book Format

Greetings from Parliament House in Canberra, where the seminar "Using Creative Commons in the Public Sector: Open Access to Knowledge, Culture and Public Sector Information (PSI)". Roxanne Missingham, Australian Parliamentary Librarian, discussed the use of Creative Commons licensing for parliamentary materials. She also mentioned in passing that there had been numerous recent requests for Hansard (the record of Parliament) in e-book format. This is driven by the availability of the Apple iPad. The EPUB e-book format is based on HTML and so it would be relatively simple to produce as a by-product of a web based Hansard. This is something I could cover in my course in January 2011, at ANU: COMP7420: Electronic Data Management.

Creative Commons from Parliament House

Greetings from Parliament House in Canberra, where the seminar "Using Creative Commons in the Public Sector: Open Access to Knowledge, Culture and Public Sector Information (PSI)" just started. Senator Kate Lundy, Parliamentary Secretary to the Prime Minister, was to open the seminar, but is in an extended Senate session on the National Broadband Network (NBN). This makes the seminar topic of access to information very timely, as to get maximum value out of the NBN, the community needs access to information carried on it.

Professor Brian Fitzgerald, Queensland University of Technology, introduced, Professor Anne Fitzgerald (his sister and also from Queensland University of Technology), as the first speaker. She mentioned that ANU has released an open access policy, also the Declaration of Open Government, by Lindsay Tanner, Minister for Finance and Deregulation (16 July 2010).

Later speakers planned are:

10:40am - 11:00amNeale Hooper, Queensland University of TechnologyCreative Commons and Government
Case Studies: public sector
11:00am - 11:20amDonna Nicholson, A/g Assistant Statistician, Integrated Collections & Dissemination Division, Australian Bureau of StatisticsCase Study: Australian Bureau of Statistics
11:20am - 11:40amRoxanne Missingham, Australian Parliamentary LibrarianCase Study: Australian Parliament
11:40am -12:00noonBernard de Broglio, Internet Coordinator, Mosman Municipal CouncilCase Study: Mossman Municipal Council, Sydney
Case Studies: educational, cultural and creative
1:00pm -1:20pmSeb Chan, Head of Digital, Social and Emerging Technologies, Powerhouse MuseumCase Study: Powerhouse Museum
1:20pm - 1:40pmSherre DeLys, Founder/Executive Producer, ABC PoolCase Study: ABC Pool
Implementing Open Access
1:40pm - 2:00pmColin Steele, Emeritus Fellow, Australian National UniversityCreative Commons in research and education
2:00pm - 2:20pmRichard Best, Legal Advisor, NZ State Services CommissionNew Zealand Government Open Access and Licensing Framework (NZGOAL)
2:20pm - 2:40pmHelen Daniels, Assistant Secretary, Copyright Law Branch, Attorney-General’s DepartmentNew Intellectual Property Guidelines and the Role of Agencies

2:40pm - 2:50pmCheryl Foong, Queensland University of TechnologyGuide to Creative Commons and Government
Wrap up
2:50pm - 3:00pm
Closing comments


ps: I noticed a locust crawling around the senate hearing room. I am not sure what omen this is. ;-)

Thursday, November 25, 2010

Femtocells for Wireless Broadband

Professor Mark Reed, Principal Researcher, NICTA, talked on "Next Generation Wireless: How can we solve the data crunch?" at the Australian National University in Canberra, this morning. This was timely as I was at a workshop with people from the Bangladesh government discussing "e-Government for Developing Nations". One key point is that wireless allows skipping generations of telecommunications technology.

Mark argued that Femtocells can provided greatly expanded broadband wireless and outlined research areas. However, it seemed to me the major issue was how this technology could be integrated into the business models of telecommunications companies (as illustrated by the lack of a business case holding up progress with the National Broadband Network). In other words the research question is: "How do we make money out of this?". As an example, one way would be for the customer who buys a femtocell to share in the revenue from others using the cell (this happens with some public WiFi systems).

Previously I suggested wireless be built into the NBN modems installed in homes. This would provide public femtocells and a very profitable supplement to the wired service.

One technical area for research is how to carry video efficiently. As Mark pointed out video is the major driver for wireless use. But video has very different characteristics to voice transmission and web access. It should be feasible to make video hundreds of times more efficient with a few simple protocol tweaks. Changes to the network topology would make it hundreds of millions times more efficient. Some changes are relatively simple, such as changing packet sizes and priorities, some will require hardware changes, such as putting caching in the cells. While there are many millions of videos which people might watch, there will be a relatively small number which most people will be watching at one time. Also the system can anticipate what people will want to watch and download it when there is spare network capacity.

Another issue is the use of mesh networks. With this arrangement, the consumer's handsets and base stations can communicate with each other, supplementing the fixed infrastructure. This could be used with intelligent and predictive caching.

Wednesday, November 24, 2010

e-Government for Developing Nations

ICT Management Handbook - A Guide for Government officers in Bangladesh, by Gregor, S., Imran, A., and Turner, T.
Greetings from the National Centre for Information Systems Research (NCISR) at the Australian National University in Canberra. I am taking part in a planning meeting for teaching e-Government to public servants. This is in conjunction with the Bangladesh Public Administration Training Centre (BPATC). BPATC is the official trainer for civil servants of the Bangladesh government and also trains senior staff in other public organisations. ANU previously provided a course for Bangladesh with part funding support from AusAID (with a book of the course published "ICT Management Handbook - A Guide for Government officers in Bangladesh".). The intention is to build on this with new and expanded courses. This may include some of my course-ware on Electronic Data Management and Green ICT.

In the past it was assumed that a lack of IT infrastructure was holding back e-Government in developing countries. However, ANU research shows that it is a lack of knowledge of how to use the technology for the business of government. In particular administrators need to understand this is not just about replacing manual clerical systems with computers, but also changing their decision making processes.

Very little technical infrastructure is needed to implement e-Government. Developing nations are skipping over generations of technology and adopting wireless mobile cloud based systems. This technology can be used by government, if appropriate security policies are observed. The expensive and cumbersome infrastructure of desktop computers, servers and bespoke software can be avoided, by agencies using mobile computers and free open source software.

However, the hardware and software is not the main issue, it is training staff to do the business of government differently. In that respect developing nations are not necessarily behind western countries. The solution for both is the same: empower key staff with education.

One issue which is common across the world is how to structure a course for working people. An ANU course is normally over 13 weeks, requiring about 8 hours of time from the student each week (some in class and some personal study). Courses can alternatively be run intensively over a number of weeks. This much study can be daunting, particularity for people who have not undertaken formal study for some years. It is therefore tempting to provide very short non-assessed courses, but these may not deliver the needed rigour and not be able to allow the students to demonstrate that they have learned what was expected.

Other issues are the delivery mode of materials, such as use of Powerpoint and a text book. In my view electronic formats and online delivery have advantages, but access to the technology will have to be considered carefully. One approach would be to supply the student with a netbook computer with built in wireless access for the course. The netbook would cost about a few hundred dollars and the data access abut $10.

Electronic Records Management at National Archives

National Archives of Australia (NAA) held a Government Agencies Information Network (GAIN) Forum in Canberra this morning. Barbara Berce from NAA began by introducing the GAIN initiative of NAA to communicate to agency staff. As well as face to face presentations, an online forum will be available in the next few weeks.

Barbara then reprised her presentation "From Red Tape and Reticence to Realisation" (from RMAA inForum 2010). She argued there are efficiency benefits from digital work practices and to "... stop living in the paper world".

NAA recently conducted a survey of government agencies. One question was if agencies had a comprehensive digital records management environment, with 30% saying yes, while 40% of the remainder having no plans to do so. Almost half of agencies have primarily paper records. Reasons for this included the difficulty of getting staff to use electronic records. A summary of the survey will be on the new GAIN web site.

Agencies are estimated to have 1,300 shelf kilometres of paper records. Much of this is outdated short term records and copies of records already stored electronically and so could be eliminated. Similar results have been found in the USA (testimony to congress by US government archivist) and New Zealand (archivists annual report).

One driver for change will be new FOI legislation. Agencies which cannot respond to FOIs quickly will have increased costs. Barbara pointed out the community now expects online and interactive access to government. This new information now needs to be retained.

The aim is that capture and disposal of records is automated. This was in "Note for file: A report on recordkeeping in the Australian Public Service" (Management Advisory Committee 8, from 2007). But apart from appropriate software (which is improving) this also requires organisational commitment. One point I disagree with the presentation on is cost. I suggest that better integrated records management could greatly reduce IT costs in agencies, rather than increasing it. As an example, eliminating the use of general purpose "office" software, particularly Microsoft Word, for the creation of documents would greatly reduce costs and also result in better records retention. Using web based software could eliminate most Ms-Word documents. Apart from the saving in licence costs (only about 10% of agency staff would need an actual copy of Ms-Office), this would would reduce storage requirements by 90% with much more efficient document formats used and and processor requirements by about 75%.

Barbara emphasised the cultural change needed. She suggested making the most of what equipment the agency has and considering records management requirements when new systems are acquired. But I suggest a culture change is needed at NAA, from thinking of themselves as being the ultimate repository of old records after they are no longer of use by agencies, to an active participant.

An example of how NAA could take a more active role would be to sponsor an online "cloud" system for smaller agencies. In conjunction with AGIMO a commercial provider would offer a complete system. All data in the system would be held in Australian Government approved facilities, under the control of Australian nationals on Australian territory. A small agency would then just need a credit card to pay for an account on the system, a web browser for each employee using it and a secure link to the cloud system. The agency would not need to purchase any records management or other software, nor run any servers of its own. Such a system could come complete with free open source software for running the basic operations of a government agency, including records management, word processing spreadsheets and email. NAA would then have a showcase for good electronic record practice, as well as an easy solution to the problem of supporting the archive needs of many small, and sometimes short lived, agencies with orphan record formats.

RMAA Canberra Meeting

The NAA meeting was followed by a meeting of the Records Management Association of Australasia (RMAA) Canberra branch Lunch Bag Education Update Seminar. I was invited to stay for the meeting and talk a little about the course COMP7420 "Electronic Data Management" I am designing for ANU.

This was an impromptu talk so I put the course web page up on screen and made these points:
  1. I am not a records management professional, but an IT person who chaired a committee on electronic document management when a public servant. I designed a short intensive course run at ANU (one of my former students was at the RMAA session) and have been teaching records management and metadata to IT and business students.
  2. ANU is the new kid on the block in terms of RM courses. It is not offering a full program at this point, just a couple of subjects (the other is COMP7410: Data Mining and Matching by Dr Peter Christen). These subjects are six weeks (half a standard university unit). The student needs to spend the equivalent of a least one hour a day studying.
  3. Courses can be done one off, or contribute to a certificate, diploma or masers, with courses from across the campus as part of the ANU Graduate Studies Select program.
  4. These courses are based in the Engineering and Computer Science with close ties to the cutting edge of new digital systems. As an example, David Hawking who developed the Funnelkack web search engine used by the Australian Government is an adjunct in the next office. We don't need electronic communications as we can shout through the wall. ;-)
  5. Courses are using advanced e-learning techniques and the most advanced Australian developed software (Moodle).
  6. Assignments will be work based: "wrote a report for your executive on the current practices in your organisation ... make recommendations for improving practices in your organisation.
Apart from being able to promote my coruse to potential students, the RMAA event was very useful as it presented details of the University of Canberra, CIT and other records management and information management courses. One point this made was that RM courses are spread across different disciplines at different universities and even in the one university.
One interesting point was that the NAA also runs courses, which while aimed at the pubic service, are available to the private sector. NAA also has teaching materials and e-learning modules.There was a discussion of professionalism which was similar to the discussions I have had at the ACS. A problem is that records managers are seen as people who push bits of paper around in the basement, rather than their role as information managers.
ps: NAA e-meeting room: Archives have equipped their meeting room with a hi-tech interactive lectern. The computer functions provided are very good, but like many such installations, the ergonomics could be better. The pen operated display screen has been mounted in the centre of a standard sloped lectern surface. There is a cordless mouse and compact keyboard provided. It would be better if the lectern was replaced with a more horozontal surface, with the screen towards the back. This would provide more space for the keyboard and more room for the mouse. The screen could be placed on a shelf with room for the keybaord to be pushed out of the way underneath. A larger, easier to operate keyboard could then be used. But that is a minor quibble and the equipment worked very well.

Tuesday, November 23, 2010

Electronic Patient Records for Better Health

Professor Hercules Dalianis and Sumithra Velupillai, from Stockholm University will speak on how e-health records can assist doctors, 11am, 8 December 2010, at NICTA in Canberra:

How can we use clinical corpora to assist the clinician, her managers and clinical research? & Modeling factuality levels of diagnoses in Swedish clinical records for information access

Professor Hercules Dalianis and Licentiate Sumithra Velupillai (Department of Computer and Systems Sciences, (DSV), Stockholm University, Sweden)


DATE: 2010-12-08
TIME: 11:00:00 - 12:30:00
LOCATION: NICTA - 7 London Circuit
CONTACT: Hanna.Suominen@nicta.com.au

Speaker: Hercules Dalianis, Department of Computer and Systems Sciences, (DSV), Stockholm University, Sweden

Title: How can we use clinical corpora to assist the clinician, her managers and clinical research?

Abstract Today a large number of Electronic Patient Records (EPRs) are produced for legal reasons but they are never reused neither for clinical research nor for business (hospital) intelligence reasons. Moreover, it is also alarming that the clinicianas daily work in documenting the patient status is rarely supported in a proper way. We are aiming to change these facts. Clinical corpora form an abundant source to extract valuable information that can be used for this purpose.

The Stockholm EPR Corpus is a huge clinical corpus written in Swedish, containing over one million patient records distributed over 800 clinics encompassing three years from the Stockholm area. We have explored subsets of this corpus with the aim of understanding the whole corpus and its domain(s). In one experiment we annotated a subset of the corpus for de-identification, and we created a gold standard for training and evaluation of automatic de-identification tools. In another experiment we investigated the relations of diagnosis codes (ICD-10) for co-morbidity analyses and found interesting results. We have also developed a method for automatic support in assigning new ICD-10 codes on newly entered clinical text, but also for evaluating already assigned ICD-10 codes. Finally we have tried to understand what exactly is written in the corpora, with the aim to construct information extraction tools that can distinguish between the factuality of diagnoses. Is the diagnosis certain, negated, or uncertain to some extent? Two annotators with clinical background have annotated a subset of the corpus for factuality levels.

Speaker: Sumithra Velupillai, Department of Computer and Systems Sciences, (DSV), Stockholm University, Sweden

Title: Modeling factuality levels of diagnoses in Swedish clinical records for information access

Retrieving relevant information from Electronic Patient Records (EPRs) is a challenging task, since there are different information needs in different situations. Moreover, this document type is a good example of where traditional information retrieval methods such as those applied in search engines are not sufficient; simply searching for keywords and retrieving ranked lists of documents is an insufficient way of exploiting the knowledge, experience and information contained in EPRs.

We have initiated the creation of Swedish clinical corpora manually annotated for factuality levels in clinical records. The first experiment was carried out on a sentence and token level, manually annotated by three laymen. Following this work, a second experiment has been carried out, focusing on assessment descriptions from clinical records from an emergency department. In this task, the annotators (clinicians) are given a diagnosis to be judged for factuality levels (certainly, probably and possibly positive or negative).

The created corpora will be used for automatic classification experiments. We want to be able to answer the following questions: is it feasible to automatically classify factuality levels of diagnosis descriptions in Swedish EPRs? Which diagnosis types are harder to judge when it comes to factuality levels? Which features are indicative? Are these different depending on diagnosis types? Are some diagnoses inherently uncertain/speculative? What does this imply?

In the future, we envisage information access systems that are able to distinguish these types of factuality levels automatically, information that could be utilized in information access systems, (semi-)automatic summarization applications, hypothesis generation and clinical research, etc.
Dalianis is an associate professor (docent) and tenured lecturer (universitetslektor) at the Department of Computer and Systems Sciences (DSV) at Stockholm University, Sweden where he heads the research area IT for Health. Dalianis received his Ph.D in 1996. Dalianis was a post doc researcher at University of Southern California/ISI in Los Angeles 1997-98. Dalianis held a three-year guest professorship at CST, University of Copenhagen during 2002-2005, founded by Norfa, the Nordic council.

Dalianis works in the interface between university and industry with the aim to make research results useful for society. Dalianis has specialized in the area of human language technology, to make computer to understand and process human language text, but also to make a computer to produce text automatically. Examples on applications are automatic text summarization and search engines with built in human language technology support as for example stemming, spell checking, compound splitting to improve the information extraction. Currently Dalianis works in the area of text mining and medical informatics focused on electronic health records. Dalianis has more than 20 years of experience of his research area. Dalianis has been project leader and received funding for over 15 national, Nordic and European research projects.

Velupillai is a PhD student at the Department of Computer and Systems Sciences at Stockholm University since April 2007. She successfully defended her Licentiate Thesis Swedish Health Data a" Information Access and Representation on the 6th of October, 2009. Velupillai is also affiliated with the Swedish National Graduate School of Language Technology (GSLT), has participated in several research projects, and is currently part of the Nordic research network HEXAnord. Velupillai has a background in Computational Linguistics and specializes on research covering both Language Technology, Information Access and Health Informatics. Velupillai has published and presented eighteen articles in renowned international conferences and journals.