Which comes first?
The digitisation or the transformation?
It is over twenty years since Nobel Prize winner Arno Penzias (then running Bell Labs) told PITCOM than computerisation never made anyone redundant. It was the organisational changes that were enabled by computerisation that made whole functions and their departments, redundant. So what have we actually learned since LEO 1 automated the production control pf Lyons Bakery and LEO2 automated the Army Pay roll? If this article is anything to go by …. not a lot.
Last year I promised, (while blogging on the problems with rural social inclusion as exemplified by the problems with using digital by default to identify farmers ) to comment on the report of the exercise commissioned to help the Labour Party Digital Government review .
I have been taken to task before by Chi Onwurah MP for not declaring my political allegiances. I therefore remind readers that whiIe I am Vice Chairman (Policy Studies) for the Conservative Technology Forum, I am also a levy paying, albeit now retired, member of Unite and of the Co-op. The views in this blog are my own. They reflect neither left, right nor centre but that area where old (non-Marxist) Labour meets tribal (open, but not necessarily “free”, market) Tory, round the bike sheds at the back.
I particularly welcome the call for those looking at “Digital Government” to focus on social inclusion and ethical standards rather than simple cost saving, although I would have welcomed rather more on how to measure performance and to hold government to account with regard to both. I would also have preferred more focus on the objectives than the technology, although I was personally interested much of the latter. I fear, however, that too many of those responsible for public sector IT systems, particularly those over-zealously outsourced by the last Labour government (such as under the National Plan for NHS IT and the many botched hospital PFIs) have mindsets akin to those who run the US Bureau of Indian Affairs.
I was therefore delighted to see the recent admission from Andy Burnham that Labour had taken the privatisation of the NHS too far. Perhaps he already knew that Circle was planning to withdraw from the running of Hinchinbrooke Hospital. I look forward to seeing further evidence that all main parties recognise that the outsourcing and offshoring of critical public sector functions, including the security of our personal information, has passed its zenith and that the time has come to rebuild the in-house systems skills of the public sector. That rebuilding needs to include the skills for end-users, not just IT “experts”, to use open source, interoperable and agile methodologies to support the integration and transformation of service delivery, under democratic control and open accountability.
Recommendation 31 of the Digital Government Review , five days training for all civil staff during the next parliament to become digital champions, is far too modest and has far too low a priority. All civil servants should have the equivalent (including both off-the-job workshops and on-the-job distance learning) of at least ten days a year to help them do their jobs better. The goal should be to involve end-users and their managers in driving incremental change within inter-operability frameworks using IT “professionals”, “systems experts” and outside “consultants” in support roles only.
That has been the ostensible aim of those developing what we now call “agile” methodologies for over forty years. The time has come to take them at their word and adopt the necessary disciplines – while recognising why this is so hard in practice .
I strongly agree with the authors of the review that the goal should be better service for those in most need (my personal rephrasing of their social inclusion goals). Given the state of public finances (including the overhang of bloated PFIs and other rigidly wasteful outsourcing contracts) that will have almost certainly have to be achieved by incremental change, on positive cash flow using software as a service over shared network and cloud services to cut new system costs by 30% (and more) above the savings on those they replace. We can then argue whether the additional savings should be used to improve services to the growing number of elderly (including me!) or to cut taxes.
I found it difficult to work out which of the other 34 recommendations were there to help achieve objectives and which were there to address assumed constraints. Many appear to very technical and capable of interpretation in a variety of ways, not all of them good professional practice. I did the programme management module on MSc06 (1971 – 3) at London Business School and subsequently ran the only one of Tony Benn‘s DTI tripartite industry strategy programmes to achieve its objectives (The Water Industry Computing Development Plan). I learned that if a programme has more than six priorities, it has none. More-over only the top three really matter. Most supposed objectives, such as health and safety, equal opportunities and even timescales and budgets, are constraints, not objectives. My own experience has been that the biggest constraints are the skills and time available, not the funding. I have seen too many expensive fiascos resulting from politicians throwing consultants and contractors with the wrong experience, motivation and management at a problem because they think that mortgaging the future will provide a short cut to success. I would love to see a bipartisan agreement to follow good professional programme management practice. But pigs might fly
Is the objective of Digital Government to deliver better and more socially inclusive automated services? Or is it to deliver better services, digital or otherwise, making use of technology to help human beings overcome organisational problems and resource constraints in addressing the needs of those in most need of help, support and/or treatment?
I suspect that some of the authors of the report did not recognise the tension between the two approaches. Others probably did, but could not agree how to reconcile the differences. I sympathise. Politicians, advisors and officials are subject to massive lobbying from armies of consultant and suppliers telling them that technology and outsourcing are the “answer” and the implementation should be contracted to them. Those currently at the top of most major suppliers to the public sector got there by winning such contracts, rather than working on their subsequent delivery. They now have grave difficulty in adjusting to the reality of a world where the public sector is not only broke but mortgaged to the hilt (PFIs and outsourcing deals).
The only realistic way forward is incremental change, using “agile” methodologies supported by low cost mobile technologies accessing cloud-based services. But this has to be funded by cannibalising existing contracts to save 30% and more on current outsourcing costs: hence the desire of major suppliers, and their lobbyists, to delay change while they shrink their UK sales and support teams and adjust to a new world. Hence also the enthusiasm for complex studies to buy time.
I therefore applaud the focus on social inclusion, but would simplify it down to a requirement that public service delivery systems (whether digital or not) should be designed for access by those in most need, using carers they trust.
No large scale roll-out should be committed unless and until the specification has been successfully tested on the target audience. The Secretary of State for DWP’s insistence on this basic principle is what lies behind the delays with Universal Credit. This appoach was alien to officials and suppliers, let alone the big management consultancies whose experts always know best. They insisted on cutting code and installing equipment under their existing, extended, contracts before they documented and tested the “pathfinders”. In other words, they ignored the reasons for Australia’s successful use of the Oracle methodology they were supposedly copying. Hence the core reason for £hundreds of millions, and more importantly, several years of unnecessary human suffering and waste.
In the small print of the 2012 budget the Chancellor mandated that no new system should go live after 2014 unless “the responsible minister can demonstrate that they can themselves use the system successfully” . Thus George Eustace is personally involved with the three week “agile” cycle to belatedly sort out the systems of the Rural Payment Agency. Similarly the Universal Credit systems cannot go live unless ministers can use them.
Hopefully the launch of the new Digital Accessibility Alliance will be followed by an extension of the policy, preferably in the pre-election budget with all-party support, to mandate the testing of all new systems with members of the target audience, not just the minister, before roll out is contracted.
I note the plans to budget large sums for teaching the “digitally excluded” how to use current technologies but regard this as less effective use of limited government funds than training civil servants to take public service delivery back in-house and to work with local authorities and the voluntary sector to run “joined up people systems” that meet the needs of those most dependent on them. Almost PITCOM’s first activity was an exhibition of computer-based aids for the disabled, in the Upper Waiting Room. It was opened by Sir George Young when he was a junior Health Minister.
Over 30 years on and we are still failing to make effective use of the technology to help those who could and should benefit most. Barnados and the Salvation Army are well ahead of the Government Data Service in the sophistication of their use of IT to help them serve and protect (their levels of delivered security also put Government to shame) those in most need. Perhaps government should pay leading charities some of what it pays to the big consultancies for advice on how to better use IT to meet the needs of those in most need.
That leads me on to the “ethical” dimension, which Labour would entrust to Cabinet Office.
The Charities are able to do such more at lower cost and more securely by enlisting the hearts and minds of those who work with and for them. Meanwhile no-one, other than a handful of Big data enthusiasts, trusts most Whitehall department further than they can be thrown . That is not because of the lack of probity of individual civil servants but because of the constraints within which they operate, including rotation to a new role as soon as they begin to gain genuine experience and build trust.
So how should we handle the issues of trust with the delivery of public services?
Abnd here I come why I agree so strongly with the ethical objectives behind the review ….
• The transition should never erode the quality of citizenship: it should enhance it or be neutral. It should be based on incentives rather than force …
• The digital divide is largely socio/economic. People should not be penalised for not using on-line services if they do not wish to: many have access problems including the elderly and illiterate. The state has to create the opportunities and incentives to acquire e-literacy and should not set an impossibly high standard (e.g. use of browsers, security tools etc.)
• E-Data should not be subsequently mis-used against the citizen, lost, mislaid, sold etc.
• Citizens are accustomed to exercise digital choice and older people prefer to use mobiles (including alarm systems).
We then homed in on six recommendations:
1) Governments have to be strongly encouraged to offer citizens online services via their choice of channel and of intermediary and these means have to be multilingual and secure.
2) Research programmes should be encouraged to ensure that the technologies used for e-Government services are fit to be used by the majority of the citizens. Given that the majority of those dependent on such services are disabled, this requires a focus on mixing audio, text and particularly video-streaming technologies.
3) Governments should use the e-participation technologies in order to gather views on the channels people would like to use, as well as on the concerns and priorities for services and to collect feedback on the quality and relevance of the services they receive.
4) It is unethical for Governments to demand information from citizens that they cannot keep secure and confidential.
5) There is a need for programmes to identify and demonstrate good practice for the secure sharing of data across organisational boundaries, including across national borders.
6) There is a need for greatly improved gradations of choice under the control of the individual: with allowance for changes of time and circumstance as well as with whom the information is to be shared under what conditions – rather than simplistic one-off choices or defaults. This approach raises many questions as to who authorizes or authenticates the choice as well as of cost and practicality.
These might have appeared challenging to some but some of the eastern European countries, lacking the technology backage of the West, had even then made serious progress. Today our smart phones offer all the necessary facilities at a fraction of the cost in 2008 so the research programmes are no longer needed. Neither do we really need mass training programmes to enable the final third to use technologies that are fast becoming obsolete. There is a good case for digital access centres (recommendation 5 in the Labour Digital Government Review) but it is more to do with people contact and support for complex distance learning packages.
The idea of using structured e-participation via mobiles to consult target audiences (making a reality of recommendation 15), via their choice of channels and to also organise “acceptability testing” before the mass roll-out of on-line services is committed, appears, however, to remain “as alien to IT suppliers and consultants as it is to government departments and policy advisors, who know that they know what is best – but for who is it best.”
There is a lot in the review on big data and open data – but who should police the governance of data collected by government and its agents under statutory powers? Confidence is crumbling in the wake of the revelations from Bradley Manning, Snowden and a growing flood of leaks from both public and private sectors. The current dialogue of the deaf with regard to surveillance powers is not helped by the inability of civil servants to do other than follow the departmental line once their minister have spoken, even if only to launch a consultation.
I rather like recommendation 34: the provision of a channel for civil servants to comment anonymously. But a channel to whom? The Select Committee on Public Administration attempted to find ways of using information on egregious behaviour provided by anonymous whistle-blowers in its ground-breaking report on public sector IT practice But getting material to the Minister, other than via a trusted intermediary to his or her parliamentary office, remains harder than leaking it to the press. Passing it to the National Audit Office is similarly fraught. I nonetheless think this recommendation is well worthy of further, bi-partisan, thought.
The recommendation (14) to “Create an ethical framework and governance for ethical issuers around the interaction of the state, its citizens and corporations” is excellent and does not just apply to digital technology. But once again who should have this role? Until the Public Records Act 1958 it could be argued that this was the responsibility of Britain’s third most senior Judge, the Master of the Rolls and his deputy, The Keeper of Public Records . It now sits with a mish-mash of under-resourced Information, Surveillance and other “Commissioners”, plus some residual functions within the National Archive.
In the face of threats from all those who think that our private information is their new oil (“big data”) what better way of celebrating the 800th anniversary of Magna Carta by putting its use by the state and its contractors back under clear judicial oversight and common (not roman) law? That could also do more than any amount of blether to help rebuild trust in the UK as a globally trusted hub for information processing.
This review is already overlong for a blog and I have yet to start on the sections on the role of the Government Data Service. Suffice it to say that I regard the GDS as a brave and overdue attempt to start rebuilding the “delivery” skills of central government after two decades of over-zealous outsourcing. But it has yet to demonstrate the skills to plan, design, procure or implement major change, as opposed to rationalising web sites and creating procurement frameworks through which little business flows. Most big departments and their suppliers are prolonging legacy contracts in the hope that it go away after next election.
Meanwhile the GDS is currently facing its first real baptism of fire: turning round the Rural Payments Agency using an “agile” approach to ensure that the on-line interfaces are usable by the target audience. If it succeeds it will have demonstrated the skills to be a worthy successor of the CCTA in supporting those departments and local authorities not large enough to be able to develop serious in-house skills. It would be good if it were to then develop into an executive agency capable of helping smaller departments take back in house that which should never have been outsourced. But whether it can ever realistically help meet the needs of big departments like the Home Office, HMRC or DWP is another matter.
The suggestion of devolving autonomy to consortia of smart cities and local authorities is rather more realistic than the idea that these should be supported from within Cabinet Office. The best local authorities already outperform central government in terms of quality of service and price performance. It would be better to reduce the level of interference from Whitehall and allow them to grow their own co-operative consortia, building on the success of past exercises, such as those organised by SOCITM and the best of the former REIPs . We need to give them freedom to integrate the local delivery of services which are currently fragmented across the central government silos created by Lloyd George in 1917 and reinforced in the later 1940s in the second attempt to create a Whitehall planned world fit for heroes.
We need to recognise that GDS is still a fledgling organisation with little more experience of delivering reliable services to the most vulnerable in society than the outsource suppliers who were going to transform the health service – and brought it to its knees.
It is more important that it is rebuilt to help support that which only government can do. One example is to support and promote the development of inter-operability standards at all levels: from data, produce and service definitions, through communications protocols and security standards to performance measures. We need a central repository for standards, maintaining files on which products and services adhere to which (and who certificated them), for use (inter alia) a revitalised GDS to police their use across current and public sectors procurement, including frameworks, to help ensure flexibility, inter-operability and compatibility in a world where both technology and demand are evovling in unpredictable ways.
Perhaps this role should be given to the National Physical Laboratory, who should be funded accordingly, including to provide and support low cost participation and access routines to international (not just UK and EU) standards bodies for the innovative small firms who are expected to provide 25% of future public procurements.
To conclude, I would like to see the Labour Digital review used as the start point for a constructive bi-partisan dialogue on how to get the best from using technology to support ethical, socially inclusive public services. But I have no illusions over the challenge this poses to those whose gained their current positions by climbing professional and technical (or political) drainpipes. For once I am rather pleased at being accused of being a journalist at heart!