“SYSTEMIC BUSINESS RISK”:
90% of Qld Govt’s ICT needs to be replaced
Total cost: $7.4 billion

31

news Ninety percent of the Queensland Government’s ICT systems are outdated and will require replacement within five years at a total cost of $7.4 billion, the state’s first comprehensive ICT Audit released today revealed, as Queensland continues to grapple with the catastrophic outcome of years of “chronic underfunding” into its dilapidated ICT infrastructure.

In late May 2012, the new Liberal-National Party State Government in Queensland announced it would conduct a six month whole of government Audit into ICT systems used across the state public sector, in a bid to identify potential savings and efficiencies ahead of projected rationalisation of its ICT assets and processes. The Audit, announced by then-Queensland IT Minister Ros Bates, who has since resigned her post, was to cost $5.2 million, use 32 public servants to complete and report in October last year.

This morning, in what will be perceived as a landmark moment for Queensland’s ICT industry, the state’s new ICT Minister Ian Walker finally released the full ICT Audit document to the State Parliament (you can download the PDF here), in addition to two subsidiary documents: An information memorandum on application and technology duplication in the state’s ICT operations (PDF), and the Government’s formal response (PDF) to recommendations contained in the Audit.

The Audit document is extremely large and complex. However, a cursory reading appears to show that the State Government’s ICT systems are in a disastrous state and require urgent intervention and government funding across the board.

“The complex, under-maintained and ageing environment has collectively reached a state of decline, where 90% of the current ICT portfolio requires replacement within five years – leaving the government with a total replacement cost of at least $7.4 billion,” the report states. “The task of addressing this issue is complex and expensive and has created a systemic business risk for government.”

The document notes that the Audit provides a unique view of the current state of ICT within the Queensland Government which was not previously possible to obtain.

“It has revealed a portfolio of ICT systems in a state of decline as a result of a long period of fragmented and minimally constrained growth, combined with chronic underfunding,” the Audit notes. “Underfunding has translated into minimal maintenance and upgrades of existing ICT platforms, resulting in ICT environments which are ageing and difficult to maintain. The fragmentation and lack of constraints has resulted in an eclectic mix of technologies and systems that complicate the management and upgrade of the total ICT portfolio.”

“The Audit revealed a number of systems that present a critical risk to government. These systems are typically long overdue for replacement, large and complex, and underpin critical activities. These systems will require urgent attention to avoid rendering key parts of government inoperable.”

The Audit lists several dozen major IT applications within the Queensland Government which are at severe risk — if they fail, the state will not be able to keep on providing critical services to the community.

For example, the Department of Community Safety (which houses ambulance, fire and rescue, corrective services and emergency management functions) is in urgent need of a new HR management system to manage its workforce. The system was listed as reaching its end of life status five years ago in June 2008, and needs to be replaced at a projected cost of $150 million.

The need for the deployment of this style of system is particularly troubling for the Queensland Government, coming as it does after the disastrous failure of a similar implementation at Queensland Health, which ballooned out in cost to the point where it is currently estimated that it will cost some $1.2 billion in total by the time the project is complete. The failure of the Queensland Health payroll systems upgrade left many of the state’s medical workers without pay for lengthy periods, and contributed to the public’s poor opinion of the previous Bligh Labor administration.

The Department of Energy and Resource Management needs $40 million to replace a “mission critical” system for all land and property-related transactions, developed in Visual Basic 6, which was released in 1998 and will shortly reach its end of life. The Department of Justice and the Attorney-General has a cluster of important legal systems handling areas such as juries administration and case management which need to be replaced, although the department did not allocate a cost to the upgrades.

The Department of Communities, Child Safety and Disability Services similarly has a cluster of systems which need replacing in areas such as client case management (for families and the disabled under its care, for example), as well as systems dealing with the management of children placed in foster care. Tenancy management systems are also out of date at the department. In total, several hundred millions of dollars in funding is needed to replace IT systems at the department.

Queensland Health’s woes are not limited to its botched payroll system. The department also needs $250 million to upgrade its patient administration system in hospitals, as well as a cluster of smaller systems supporting medical care.

The Queensland Police Service appears to be faring a little better, but still has a clutch of IT systems listed as needing replacements. The Department of Transport and Main Roads appears to be in a similar state to Queensland Health and the Department of Communities, with several initiatives, including vehicle registration systems and roads management systems, requiring tens of millions of dollars each to be replaced.

“The program of work undertaken to address the Y2K bug in the late 1990s had the unplanned side effect of synchronising the end of life dates for many of government‘s key systems,” the Audit states. “By December 2015, 77% of the reported applications will be at or past their end of useful life, 84% by December 2016, 91% by December 2017, and 100% by December 2021.”

“The combination of poor technical condition, complex, custom applications all reaching end of life in a short time horizon dictates that a different strategy will be needed to provide the required level of capability. Continuing to source that capability through heavily customised or purpose-built applications software is no longer sustainable.”

When it comes to its fundamental IT infrastructure, Queensland is also far behind much of the rest of the Australian public sector. For example, only 54 percent of its desktop PCs are running the stable Windows 7 operating system, with the rest still using Microsoft’s outdated Windows XP platform, which the software giant has continually moved to stop officially supporting. The cost of completing the upgrade is estimated at running between $100 million and $160 million. This is a cost that Queensland is not sure it can afford; with the Audit noting: “It is incumbent on government to explore alternative arrangements for its desktop fleet.”

Queensland has had a whole of government datacentre consolidation program under way for some time, under the so-called ICT Consolidation project begun in 2007. However, the project has been a failure, according to the Audit.

“By any measure, the ICTC initiative has been unsuccessful. It has struggled to get agencies to comply with the requirement (set by Cabinet) to consolidate data centres24 and ICT infrastructure. Six years after the Service Delivery and Performance Commission‘s recommendations were released the mandate for infrastructure consolidation remains ineffective and 45% of the Polaris data centre remains underutilised. The cost of unused floor space at Polaris is $3.26 million per annum.”

The state’s telecommunications services contracts also appear to be out of control.

“The Queensland Government has a plethora of data and phone networks, all evolving out of local agency requirements,” the ICT Audit states. “These networks are supported by thousands of contracts with telecommunications vendors. With the exception of internet services and some multi- agency initiatives (e.g. Public Safety Network), decisions on telecommunications arrangements are made at the agency level.”

“From a procurement perspective, the level of attention paid to the quality of these arrangements across government has generally been substandard. As a result, government has a range of telecommunications arrangements33 in place that have been either strategically inappropriate or simply too expensive relative to current market rates.”

“The Queensland Government is paying for a large number of networks covering the same geographic area with no ability to aggregate capacity or speed. This agency centric investment in telecommunications prevents government from leveraging economies of scale.”

Ultimately the ICT Audit strongly recommended a fundamental change to the way the Queensland Government procures and implements technology solutions. In general, it recommends the Government stop developing its own IT solutions and move to commodity platforms which are more easily supported; that it stop providing IT services in-house and outsource them, and that it move to a cloud computing model for services as much as possible.

The Audit emphasises that action must be taken: “Debating subtle nuances of the approach will waste precious time that the government simply does not have.” If it is to maintain its ability to deliver key services to the community, the Audit notes that a number of key initiatives as outlined above must start immediately and in parallel.

opinion/analysis
Wow. Just wow. I thought that the Victorian Ombudsman’s report into IT project delivery in the Victorian Government in November 2011 was bad. But this ICT Audit delivered by the Queensland Government shows that the state’s ICT systems are in nothing short of a disastrous situation. Virtually every major IT system in the Queensland Government needs replacing, as does much of its fundamental IT infrastructure.

I have never seen such a nightmare situation in my decade of reporting as a technology journalist in Australia. The depth and breadth of the problem facing the Queensland Government right now when it comes to its ICT operations is simply staggering, and virtually impossible to get your head around.

There is a lot to delve into in this report, and I haven’t even really begun to scratch the surface here. Next week I’ll follow up with a series of stories looking at the potential way forward for the Queensland Government with respect to the remediation of its ICT systems and infrastructure, as well as the LNP administration’s response, and other associated matters.

But let me say this: If you work in IT in the Queensland Government, you’re probably facing a big choice right now. Either get out and go somewhere more sane immediately (the preferable option from a personal perspective), or face the reality that you possibly work in the worst place in Australia when it comes to the use of technology in a government or business context, and that things are probably going to fall apart in increasingly disastrous ways in the near future. Shit is going to hit the fan.

I don’t swear lightly in Delimiter articles. But holy flying death lizard: What the fuck, Queensland? How did you manage to screw up almost every single aspect of your ICT operations? How did things get this epically, fundamentally, catastrophically bad? This is a nightmare which any chief information officer in their right mind would run screaming away from.

Image credit: United States Geological Survey, public domain

31 COMMENTS

  1. “only 54 percent of its desktop PCs are running the stable Windows 7 operating system, with the rest still using Microsoft’s outdated Windows XP platform”

    Wow, 54 percent is actually far higher than I thought… can’t remember if I was reading here or somewhere else that many businesses had no intentions of upgrading from XP anytime soon because there just isn’t the money for it.

    also a typo:
    “As a result, government has a range of telecommunications arrangements33 in place”

  2. “How did things get this epically, fundamentally, catastrophically bad?”
    As I have said before here, the problem is that the requirements to be served are constantly changed, without notice and funding. Report on this. Manage that. No extra staff or funding.

    The fact that systems work at all is a credit to the staff.

    No government will commit to setting out requirements in advance and sticking to them. Here comes scope change.

    The Qld Health payroll system was meant to be a lead agency implementation of a gov wide payroll system. IDES to replace various outlook/exchange systems. The agency specific systems are just that – very specific needs and tough requirements – by law all gov funds & assets must be tracked, decisions recorded and accessible potentially forever (subject to the Public Records Act).

    For example:
    “Department of Community Safety (which houses ambulance, fire and rescue, corrective services and emergency management functions) ” 10 years ago or so ambos were not even part of the Qld government. And the repercussions are still being felt. 5 years ago Corrective services was its own dept. So when merging depts, functions and agencies, do you say – no emergencies, empty the jails while we sort out our IT systems and please fund us to replace our currently working systems with untried systems we can’t fully spec out?

    • Some excellent points here.

      Ren, you day you can’t understand how QLD managed to s rew up so catastrophically – I see business as usual. The only reason you haven’t seen anything like this in your career is because this is exactly first time any government has performed an audit as comprehensive and as honest as this. I guarantee an audit of similar scope performed by other state governments would expose similar ineptitude, underfunding, deeply segregated departments, infrastructure overlap, waste, ad hoc fixed and expensive bespoke systems requiring unique expertise that are complex and costly to upgrade.

      Very very few people would have the experience or scope of powers to have enough knowledge to piece together a comprehensive overview of the problem for each state, but it wouldn’t be at all difficult to find people with direct experience with such systems from within various departments. My experience and anecdotal testimony from relevant people suggests that an up-to-date, reliable, comprehensive system that caters to stakeholder requirements and has been adequately planned, implemented and isn’t a pig to support is the unicorn of government ICT – it simply doesn’t happen. The norm is precisely indicative of what has been exposed in this audit.

  3. “How did things get this epically, fundamentally, catastrophically bad? ”

    Hmm… let me see:

    “The cost of completing the upgrade is estimated at running between $100 million and $160 million. This is a cost that Queensland is not sure it can afford;”

    Any time even basic computer upgrades are getting ‘too expensive’ you know your mindset is fundamentally flawed.

  4. “About 2000 Queensland IT workers will be sacked as the government looks to outsource their positions, the state’s public sector union says.”

    http://news.theage.com.au/breaking-news-national/2000-qld-govt-it-jobs-to-go-union-20130607-2nuq8.html

    If I was the cynical type this would reek of the government attempting to manipulate the public into accepting the report as justification for sacking thousands of IT workers and handing it all off to private business.

    Mike

    Mike

  5. Perhaps a good time for a watershed decision?
    Perhaps even follow Munich’s etc. example and not only save money, but employ Australian talent to use Open Source software, rather than to keep importing commercial software from those clever Americans?
    If we did it right, we might even develop a software export industry, as I don’t think that Australians are really stupid.

    • Doc, I fully agree.
      Opmantek are one such provider of network monitoring, management and audit software. Headquartered in QLD, their products are used (and supported) by large multinationals and telecommunications carriers mainly overseas- a fantastic export product!
      There are many other OS projects capable of replacing expensive legacy/enterprise systems for a fraction of the cost. These products and their development are flexible enough to support and replace the features of numerous other suites while building a working knowledge base, at home, that actually matches what is needed!

  6. “How did things get this epically, fundamentally, catastrophically bad?”
    – or maybe, how to learn and avoid this happening yet again …

    It is very easy for outsiders to be critical and write off past administrations, or IBM contractors, and then suggest it should be straightforward to fix. As Trevor commented – this is normal! More examples:

    Government approved a decision to not pay annual maintenance for desktop licensing. Over the years, this saved tens of millions; it was known that the licences would need to be updated so they approved for an amount to be quarantined out of the IT pool to pay for that upgrade. They forgot – and now say they can’t afford it.

    Apart from the IBM contract, there are past examples of outsourcing gone wrong eg, Mincom provided IT hosting and support services to compete with CITEC, but eventually withdrew from the business. Another lesson that should be learned.

    MoG changes (reorgs) are made without regard to IT impacts. Each change of Government largely destroys what was achieved by the previous one. Managers change, systems become unsupported, valuable policy regimes scrapped. This ignorance is very destructive. Eg, at the start of this administration, a key minister claimed they used their home email for Government business.

    The dozens of recommendations are each two year projects. The difficulty is that timeframes do lapse (7.5 bn?), budgets do blow out and it takes two years for executive to understand a project is in trouble so – do they declare so, just before an election? Wait for afterwards? No.

    The key is better IT management: this cannot be outsourced. The executive needs experience in the Government environment and its successes and failures. Outsourcing may be ok if properly contract managed but this capability is not demonstrated at the moment (and it should be before you do it…)

    Application of IT to Government agency business is core business.

    Sackings and heading blindly into outsourcing risks irreversibly destroying knowledge and to Government capability that becomes doubly costly to replace. It would be folly to save a few $10m’s through sackings only to efficiently handle a $7.5Bn upgrade.

    • TonyX:
      I’m impressed by your depth of knowledge. I’m only an interested observer, albeit a 17-year Linux (Debian) user, so I can comment only on general principles. I’d really like to see home-grown software in Australia, rather than genuflecting to clever overseas corporations, however. Hell, if a busy medical consultant can write his own navigation software for a circumnavigation, what can our professionals do? You’ve given me a far greater insight. Thanks.

    • TonyX

      Your comment about managing outsourcing is spot on. Gov’t only see the pretty lights of somebody else is doing the work, not that the contract has to be managed, the SLA’s actually mean something and have to be enforced, a relationship has to be maintained so in actual fact staff are still required which could have done the job in the first place. Current Government is far too fragmented and siloed to be in a position to actually outsource in one great leap. It must be done in an organised and pragmatic way in stages while Government learns to adjust to the new way of interacting with a third player, themselves, the community and now the outsource.

      Who said bureaucracy was dead!! We will just shift the risk further from ourselves and pay 5 times the price – this won’t stop the internal machinations and approval processes that seem to go nowhere.

  7. I’m not surprised in the slightest. This is typical of government ICT, which I’ve experienced from both the inside and outside.

    The overall issue comes down to management and governance brought on mainly by some of our ridiculous salary laws. Governments can only pay statutory amounts for staff which are typically far lower than market average (particularly at senior levels) and hence attract really poor staff, with the more senior the staff the poorer they get.

    This results in the default cop-out of bringing in a major outsourcer (e.g. IBM, HP, Tata, etc.) to deliver solutions (since salary caps don’t affect contractors). These organisations are there first to make money, second to secure their position and finally to ALWAYS make money. They have no real drive to succeed in their projects since there are virtually no consequences of failure (see the recent QLD health debacle as a poignant example of this). One this starts to spiral out of control, it only gets worse as the outsourcers consume so much budget (that cannot be reclaimed) that – being strapped for cash – the next stage in the cycle comes on line which is to outsource further – this time to low-cost overseas resources for “less essential jobs”, this creates even most disconnects and chaos as language, distance and cultural issues are introduced into the mix. The end result is what Queensland is seeing now, and most other state (and federal?) government departments are seeing with their ICT.

    Around 17 years ago, I was involved in the successful privatisation of a state government body . My role as the senior technical contractor was to slice all their ICT Infrastructure off the government network and into its own environment, I accomplished this in a 12 month period with the assistance of a team of 5 – only one of whom was a government employee (and he was nearly ineffective during the project). My manager – whose main job was to build our team and secure our budgets and business buy-in – was also a contractor and certainly earning far more than what would have been had he been an employee. The whole project cost less than $10 million including building a 16 site WAN along with a datacentre filled with servers literally from scratch. I have since spent a lot of time as a solutions architect for several of the larger Systems Integrators and – had I been representing them – would have costed the project at $25-30 million with double the time frame (or double the money to meet the deadline).

    The Queensland audit is a great first step. The next step is to get the RIGHT people to clean up the mess. The right people are people hired (or contracted) directly to the QLD government (and paid market rates) rather than being loyal to a mutli-national integrator – SIs will still be necessary but at a project level, not governance. This is obvious but will the ministers and senior bureaucrats be brave enough to do this? If they do, there will be a lot of pain (any project with 9 “0s” in it will be painful) but the end result will be a template for the country on how to get things right.

  8. “Ninety percent of the Queensland Government’s ICT systems are outdated and will require replacement within five years”

    So? All active equipment should probably be replaced every 4-5 years. Bit of a pointless and sensationalist remark, I reckon.

  9. “How did things get this epically, fundamentally, catastrophically bad?” Part of that would be the LNP’s relentless negativity regarding the Bligh government’s attempts to spend money on ICT (Health payroll aside). So, the government tries to spend wisely without spending too much and the other mob then get in and blame the previous one for not spending enough even though they were attacking them for spending too much. Politics in the IT industry is really stupid.

    • i fail to see how the lnp can be largely to blame for this fiasco.

      most of this IT non-action happened under the watch of the previous labour government.

      as a former state government employee, this audit result is no surprise. credit to the lnp for having the guts to perform an audit like this at all.

      • I didn’t say “largely to blame”. I said “part of the blame”. There’s a big difference.

        Then again, a lot of this “IT stupidity” doesn’t reach the level of MPs as the decisions are made without going any higher than a senior member of the department. What I’m trying to say is that a lot of this idiocy happens regardless of who was in power at the time as changing government doesn’t change every public service employee as well.

  10. The problem with government departments can neatly be summed up as “any decision as long as it is the right one”. In commercial businesses someone will (mostly) make a decision because not making a decision is simply not an option, whereas in government (almost) nobody wants to be responsible for making the wrong decision. The strong public service unions mean that decisions involving changes in work practice can be overturned, especially if they threaten job security. Add to these challenge, changes imposed by political reorganisations and legislation with short implementation dates.

    All makes vendors wary of becoming involved and wisely adding a premium to any government contract.

    As an aside, I’ve sometimes wondered if with all the reorganisations, are Stationary Printers strong contributors to political parties?

  11. the problem is – IT in gov and private business is seen as a cost centre.

    Therefore – the cheaper option is always preferred.

    Decisions are made by people with little or no idea about IT , and then it’s up to the IT department to somehow make it work.

    Everything fails , IT is blamed and the people actually responsible for the decision are never to blame (and usually the 1st to wipe their hands of the situation).

    My favourite part of working in the IT industry is when someone in management / senior management has this lightbulb moment … and a 1 minute idea for them = 1-6 months work for you …

    which in the end , wasn’t worth the effort..

  12. Told ya with was bad Renai!

    Source : Me Previous Frontline worker!

    Me: “Oh your running an outdated java. Hey why is the mouse moving. wtf”

    • For example

      Key observations: ICT replacement cost – technologies at risk

      I can confirm 6.31 Million was spent repairing Acer faulty laptops with the education department. Perhaps Queensland Government should read Fair trade and Warranty laws?

      Source: I was one of the trial techs

  13. Hmmm … good, and sobering, post Renai. Expletives even!!

    Observation of similar, if less severe, patterns of activity in other jurisdictions over the past decade is what leads me to my enthusiasm for cloud services for government. The basic problem is that many (most?) agencies need to be saved from themselves.

    Our Westminster systems of funding and accountability in government place the primary bias of decision making on agency heads. It is a highly devolved approach. Agency heads are accountable to Ministers, who come and go on a regular basis. When push-comes-to-shove agency heads are motivated to preserve their capacity for autonomous action in order to be able to deliver whatever is requested by the Minister of the day. This is just the way things are. Physical consolidation to create mega-departments is the only sustainable way to change this dynamic.

    Agency-by-agency decision making is not all bad – it can enhance responsiveness to agency-specific policy and service delivery imperatives. It does, however, lead to fragmented policy and service delivery and to fragmented infrastructure and applications. As well, of course, it leads to under-investment due to an inability to leverage economies of scale and to difficulties in attracting and retaining specialist staff.

    So … what to do? The traditional ‘fix’ is to attempt to create governance arrangements and matrix structures which overlay ‘horizontal’ planning and coordination across the ‘vertical’ agency stovepipes. In ICT terms this would be an Office of the CIO, combined perhaps with operational shared services for common infrastructure and applications – on a whole-of-government (WoG) basis or in clusters of similar agencies. Has this worked? Beyond phases of success for 5 years or so NO … it has not. The problem is sustainability. Even if a Wog CIO or shared service can get traction they swim against the tide. The dominance of the devolved agency-by-agency decision making model eventually wins … and the WoG strategy starts to fray at the edges and ultimately fizzles out. It requires strong and sustained leadership, but this is rarely achieved. Leaders come and go …

    The reason I think cloud services are useful in this context is because they provide a circuit breaker for this unwholesome dynamic. Mature enterprise-grade cloud services are externally provided shared services that work. They are used by many different organisations, the more the better, and are inherently sustainable because they are insulated from the dysfunctional governance dynamics of any one of their individual customers. The services enforce a degree of standardisation and also iteratively evolve to stay current in technical and functionality terms without individual customers needing to invest unilaterally. Staff that were previously occupied carrying out low-level, relatively-non-value-adding, technical activities are freed up to focus on activities that are more directly related to the delivery of citizen outcomes – new applications, information management, process reform, service innovation etc.

    This leads me to state the obvious. IF EVERY NEW $ THAT AGENCIES INVESTED IN ICT WAS SPENT ON USING CLOUD SERVICES THE RESULT WOULD BE TO CREATE AND SUSTAIN SHARED CLOUD SERVICES IN AUSTRALIA THAT WOULD BENEFIT ALL OTHER AGENCIES IN ALL JURISDICTIONS. This is the only strategic solution to the mess that Queensland finds itself in … and to the less spectacular travails of other jurisdictions. Cloud services are simply shared services that work because the vendors are able to focus on delivering shared services that work … rather than having to respond to the twists and turns of each customer’s individual ‘unique’ requirements in a climate of under-investment.

    The cloud services model is perfectly attuned to the agency-by-agency decision making dynamic of governments because the economies of scale and standardisation are service attributes that accrue to any individual agency. Agencies can act unilaterally and still access high quality, sustainable, ICT capabilities. Critical mass and APIs lead to both de facto standardisation and to a reasonable degree of interoperability.

    This model can, and does work. Of course there are challenges in the transition and not all applications can be delivered in this model … but many can … and should be. There is hope!

    For cloud services to become more widespread, however, agency executives and staff need to be re-educated to become intelligent consumers of both shared and cloud services. The critical change needed is to make more practical decisions to adapt structures and processes to align with the standard functionality of cloud services … rather than to impractically and unaffordably insist on customised, dedicated, systems. [It actually IS this simple. Senior executives just need to read the writing on the wall and change their behaviour. They have run out of alternatives.]

    This is why cloud first ICT policy is a fundamental first step for states like Queensland. Symbolically, it is a signal that the rules of the game have fundamentally changed … and that the Government has woken up to the 21st Century global digital economy.

    • Steve- what does cloud delivery of a service have to do with the public service’s inability to develop the software to run either on the cloud or in house hardware?
      The cloud is just a place to host your software, you still need to build it (or buy it if it is generic).
      My own 2cents from viewing these goings on from inside the tent and outside is that Government depts are incredibly good are diluting any form of accountability.

      • Hi Douglas, what I am referring to is cloud services … IaaS, PaaS and SaaS. The highest value comes from SaaS solutions which deliver advanced and configurable application functionality. This is much more than “a place to host your software”. It is a shared service that provides the application functionality needed to transform policy and service delivery possibilities. PaaS solution significant improve application development productivity and accelerate the pace of innovation. IaaS solutions improve flexibility and reduce operational costs.

        • SaaS for relatively generic things like CRM (salesforce.com for example) is one thing, but for very specialised, specific systems (DIAC case management or Customs Cargo Management for example), it’s just not something you buy off the shelf (or already living in the cloud as SaaS)

          Sure, host it in the cloud, but somebody has to develop the system in the first place.

          I think SaaS is great and definitely has it place, but to just suggest that every service should be purchased as SaaS, problem solved, is a little naive I believe.

          • Hi Douglas, the thing is that nobody is suggesting that all workloads, data and applications can or should be sourced as cloud services … not today anyway. Rather than saying “cloud services are no use because they cannot solve for all needs” we are better to consider which needs they can solve. There are case examples now of successful use in agencies … the Victorian Department of State Development and Business Innovation uses Salesforce for grants administration for example. This is not just ‘generic CRM’, it is the use of Salesforce (in this case) as a flexible, configurable, and robust platform. The NSW Department of Trade and Investment is implementing SAP Business ByDesign SaaS as a core ERP solution for a cluster of agencies … finance and HR etc.

            If agencies start insisting on SaaS solutions they will stimulate the market to produce a wider range of solutions that are relevant and useful for government work. [yes, of course, not tax and immigration etc. etc. … but there are many many other applications that agencies need … no?]

          • Hi Steve,

            Nowhere did I suggest that “cloud services are no use because they cannot solve for all needs”. That’s a silly statement.

            If I actually quote verbatim from you and not make statements up:

            “IF EVERY NEW $ THAT AGENCIES INVESTED IN ICT WAS SPENT ON USING CLOUD SERVICES…”

            My point was and still is that it is simply not possible for “EVERY NEW $” to be spent on SaaS as Ovum recommends, as the really big ticket items just don’t exist as SaaS, as you later concede.

            Obviously Ovum are invested in selling White Papers that promote cloud services, but it is not right to suggest they are a panacea for everything that ails ya.

            Regards,
            Douglas.

          • Hi Douglas, hmmm … apologies for putting clumsy words in your mouth (as it were). I think the bigger picture view, however, is that we have a mode of operation where legislation is drafted without consideration of implementation costs and practicalities etc., politics creates ever changing and highly specific requirements, and the culture of the public sector is to respond to this reality with customised processes and systems. This all flows downhill to treating most applications as something that must be handcrafted to conform to ‘unique’ requirements. The result is expensive and inflexible … which is OK when there is money to burn … but is (as graphicly demonstrated in Qld) inherently unaffordable and unsustainable for most jurisdictions.

            From a strategic perspective, my view, is that we need to find better ways to challenge this dynamic by re-framing the question from “How do I commission the building of an application to exactly match my current view of our requirements?” to “How do I source ICT capabilities which will meet my evolving business needs?” The first approach is ‘internal-out’ , the second is ‘external’in’ – involving earlier assessment of how standardised pre-existing solutions might be leveraged and thus how requirements might be pragmatically challenged and re-assesed to find a sweet spot of business outcomes, cost, time etc.

            I agree that this is simplistic if you are taking about redevelopment of the entire immigration case management system … a huge and complex task … and not the place to start. But we need to start somewhere to rebuild the logic of how ICT capabilities are sourced and managed in government.

            I stand by my statement “IF EVERY NEW $ THAT AGENCIES INVESTED IN ICT WAS SPENT ON USING CLOUD SERVICES THE RESULT WOULD BE TO CREATE AND SUSTAIN SHARED CLOUD SERVICES IN AUSTRALIA THAT WOULD BENEFIT ALL OTHER AGENCIES IN ALL JURISDICTIONS” as a challenge … most particularly at state government level.

            I’m not saying, literally, that every $ can be (today). The Qld audit report, however, reveals that todays approach is not really working … so we need to consider alternatives … which means challenging the status quo.

        • Unless of course you can point me to a Immigration Case Management SaaS which replicates the SfP or ICSE DIAC functionality? In which case I am totally wrong!

          If it doesn’t exist, someone has to write it the first time. This is where all the dollars are spent (just ask IBM).

          Regards,
          Douglas.

  14. Steve, Can you explain how “cloud computing” and a cloud first approach will solve the accountability and governance issues that are really at the heart of the decade long failure of ICT in this state and else where. For example, I have sat in senior executive meetings within government as a strategist and advisor – Its my observations that ICT has historically been discussed at the highest levels in two contexts – when /how do execs get the latest toys (eg previously blackberries, now iphones/pads). And secondly, Oh F, who can we hand the responsibility of ICT to – its a black hole – oh that’s the CIO – IE executives hand over their executive oversight to the CIO and its set and forget.

    Furthermore, In the Queensland Government environment governance of ICT has during the period of these failures sat within the IT department – that’s the same area responsible for delivery. Hence the cover up and lack of questions because those deliverying ICT services are also governing them.

    This issue has been continually raised but not addressed – including by the Queensland Audit Office.

    If this culture and executive failure to own outcomes is not addressed its likely that the Queensland Tax Payer will continue to bleed money. MSP and Prince 2 only give the illusion that things are being done. Its process oriented nature dominates often at the expense of outcomes.

    How will a cloud Ist approach address these route causes of failure. Wont outsourcing simply push these failures under the carpet via commercial in confidence contracts?

    • Hi ‘Ex Gov & Observer’,

      You are right that the nub of the issue is executive accountability for ICT decisions and their implementation. I have written on this issue as well. Too many executives have ‘tuned out’ of any responsibility for ICT – regarding it as a low level plumbing issue, messy and smelly … best avoided to protect your career. They have essentially lost hope. The safe thing is to obfuscate accountability via committees and make sure that no individual can be ‘fingered’ with the blame. For any executive, particularly agency heads, to say that they “do not know how to manage the successful application of ICT” is akin in 2013 to saying that they do not know how to do their job.

      A big part of the reason for this, however, is that ICT-enabled business change projects have had too high a failure rate for too long. Avoiding accountability for, and involvement in, these projects is a learned (and understandable) behaviour. Turning this around requires creating successful projects that deliver business outcomes without drama, cost blowouts and disappointment. My case studies of successful use of cloud services in agencies in Australia reveal some hope because they demonstrate a new path where executives have re-engaged in the use of ICT to deliver innovation in policy and service delivery. In most cases, the projects were owned and driven by a senior executive who understood that cloud services enabled a better, faster, less costly and less risky way forward compared to traditional ICT methods. The key thing is the ability to leverage a pre-existing platform in a more agile way – iteratively and by making pragmatic tradeoffs of required functionality, cost, timeframe etc.

      The thing that executives can grasp is the new calculus of transformation: Agile Thinking + Cloud Services Platforms = Innovation and Productivity. This enables them to focus on business outcomes not technology procurement, ownership and maintenance. Not for all applications … but for some. The result is a re-emergence of successful experience, which will rebuild confidence … and hence accountability (over time).

    • Outsourcing doesn’t solve anything, you are right.
      A big multinational is picked who has huge teams of lawyers with contracts that basically let them get away with murder.
      On the Govt side, the dept creates multiple committees who meet regularly, achieve very little, and dilute any form of accountability or responsibility. Usually the people on the committees have very little idea of what they are actually doing, but they got seconded to this division and put their hand up to “look busy” with more meetings.
      Years later, with nothing achieved other than paying the regular bills from the outsourcer which now sometimes mount up to hundreds of Millions, it all falls in a heap and everyone looks for someone to blame.
      And we in the public all throw up our arms and write/read articles like this asking “how did it all go so wrong???” :)
      Rinse and repeat ad infinitum.

Comments are closed.