Modernising Justice through IT, 2

22 June 2007 by Rod McLaren

Part two of my notes from GovNet’s Modernising Justice through IT event, 12 June 2007 (part one is here), and as before my comments are in square brackets.

[I’m at a conference on using IT to modernise politics and I can’t see any laptops out except mine: the justice community has different very practices and priorities at conferences compared to the reportage- and backchannel-centric behaviour of the internet conference community, for instance.]

Seminar: Delivering information to the front-line (Bedfordshire Police and BlackBerry):

• Beds has 1,200 uniform, 1,000 civilian staff; historically “computerised but not mobilised”
• activity-based costing suggests 50% of officer time spent on in-station updating of systems, paperwork
• Why IT projects fail (according to both users and management): lack of consultation; inappropriate training; insufficient support
• are users more forthcoming with outsiders/consultants? [in our experience police users are very forthright!] used C-Innovate as user liaison
• users wanted: email, PNC, Criminal intelligence, Command&Control
• considerations: other orgs using the solution?; ease of use; functionality; integration; ease of admin [ie does it match up to internal IT team skills. If not, who will support/manage it?]; security; battery life; cost
• buying off-the-shelf and lightly customising: the patrol car analogy [ie buy Fords and put radios in them, rather than commission bespoke PoliceCars]
• 40 users Nov06-Apr07 pilot
• seeking/allowing considerable user involvement but not mandating use. 10-15% of users don’t like it, and won’t ever like it [they tend to resist it from day 1, very hard to recover them with evangelism, even if the device and app are good]. ThamesValley’s WinMobile trial reports similar numbers.
• Standard PNC training programme cut-down because wasn’t enough available officer time; 2-3 days on PNC, 3-4 hours on BlackBerry [note that 1: going mobile might be the first time an officer has directly used PNC; 2: training becomes inconsistent by local needs to get mobility out, so what effect upon training compliance at macro level?] For new officers, mobile becomes the only way to check PNC they ever knew about.
• mobilised: PNC (by NDI briefings (by Airpoint urgent message; CMS2 (crime mgmt system); PNLDB (police legal powers db, via browser?); warrants (via browser); more coming. 5x more PNC checks in a typical shift.
• note neighbourhood officers may have no single, physical HQ to return to anyway, so mobile delivery is key
• Costs: (measured over 3 years) 249 cashable and 182 non-cashable/opportunity cost (training etc) = £431/device/year. [Devices themselves cost 100, and data 10/month, presumably in that 431 figure.] Total outlay 30-40k.
• Benefit measurement: “measuring to excess can cost more than the value of the benefits you’re trying to achieve” [this is the local, in-the-field story that counters the centralised-gov story and measurement culture of CPS, CJIT etc: if funding is local, barriers to proving ROI can be determined locally and set lower ]
• BlackBerry ROI study: “email saves users 63 min/day” [obviously it isn’t quite as simple as that: BlackBerry widows might disagree, and even users cf continuous partial inattention, etc]. For Beds: 1 hr/day @ £22/hour * 180 working days = 3.9k/device/year benefit. [also, what about the non-tangible ROI: when users like the app+device they feel more positive (viz Crackberry effect), also there’s inevitably higher usage, better quality data, and a much reduced training and device replacement cost because devices don’t mysteriously get broken or lost]
• BlackBerry: NPIA say there are 8,000 mobile data devices currently in UK policing, and BB claim to have 7,000 BBs deployed. [Reason: no large-scale Win mobile (why not?) or Airwave/WinMob deployments yet (not in market yet?)]
• CESG recently approved BB Curve for Restricted data use [and they’re working on Confidential data use, which IMPACT will require]
• BlackBerry think that mobile data business models will switch from flat-rate-all-you-can-eat as takeup grows [recent evidence suggests flat rate is growing though]
• Q: any evidence yet that speedy image dissemination gets results (aka the “golden hour”). A: yes, anecdotally.
• Q: use at courts. A: GSL’s prisoner transport and management at periodic courts – all done on BB [disclosure, that’s us]
• Q: Cheshire police user in audience pipes up to say his Blackberry is a godsend – got work done on the train on the way down etc
• [good seminar, great stats, and Bedfordshire are a good cheerleader for Blackberry]

Seminar: Practical Steps for Multi-Agency Criminal Justice Information Management (Sponsored by Northgate Information Solutions and Initiate Systems):

• Rob Watts, Northgate
  • 29% of analyst time spent searching for information. “Lots of data, but little information”.
  • [and in summing up:] success requires thoughtful procurement, incremental approaches [rather than BDUF], and agile development needs flexible procurement.
• [good common sense from] Daniel Batts, Microsoft CJ and Public Safety
  • Shows the [nastily complex] communities of interest slide from the Home Office’s Information, Systems and Technology Strategy 2007-8 report
  • what’s the right information?: Knowing what to gather, what to do with it, what to pass on [and to whom], and how to value the result.
  • younger users have higher expectations. [Google generation; and the thought often occurs that a good short-term fix to some of these information challenges might be to run a Google Enterprise box over the data source and provide training in search queries…]
  • success factors: secure proactive executive sponsorship; leadership to drive [top-down?] collaborative culture, partnerships, standards; define info mgmt flows by business process not by hierarchical silos; more decision making to the process [local, lateral]
• Chris Westphal, Visual Analytics [which looks like a fairly simple words-on-sticks approach to the underlying data, though that may be an artefact of the seminar format. Chris should get Stamen in to take the visual aspect further.]
  • [However, this is good stuff:] Data quality issues: poor collection instruments [and the inevitable disconnects between context of collection and that of eventual use]; information not verified; inconsistent or incomplete data (typos, no cross-refs, blank/test data); errors in data; misrepresentation of data (un/intentional). Process issues: how to make data meaningfully actionable]?
  • larger the data set and the more “human interaction” it has, the more data inconsistency there will be [then demonstrates the power of sqling to show humans are inconsistent. Or lie.]
• James Wilkinson, Initiate – Initiate Hub entity resolution to achieve “the golden nominal … single trusted view of the ‘truth’” [ie it’s a sieve for meaning that should result in genuinely unique identifiers, the database state’s dream. Trusted by enforcement, but what about citizens?]
• Simon Blades, Northgate: Intelliframe [horrible visual metaphor of a set of modules that looks like a street map view or grid but boundaried by pipes that seem to weave in and out of each other in the manner of Escher. Edward Tufte would have a heart attack]. Expect COTS not code.

Managing Justice: C-NOMIS – Linking the Prison and the Probation Services – Mike Manisty, Director, Offender Information Services, CIO, National Offender Management Service :

• [Very encouraging to hear him start by mentioning people: those that pass through the system and the wider body of citizens the system serves]
• C-NOMIS – 75,000 users, 1.500 locations, replaces 140 separate instances of prisoner db. [C-NOMIS based on Syscon’s TAG application, buffed up by EDS]
• Offender mgmt model delivers – (for offenders) continuity, consistency, commitment, consolidation; (for offender mgmt) control, collaboration, commissioning, cost [reduction]
• insiders: 300,000 staff on GSI or CJX-based islands, often closed silos; outsiders: 200,000 users
• “coherence in complex world of offender management” – people on the ground making genuine commitment [he’s a believer, and funny]
• C-NOMIS live at Albany prison Dec 2006 [will be interesting to hear if NOMS is able to force C-NOMIS upon privately-run prisons without re-negotiating the contracts]

The IT Industry Role in Modernising Justice – Mike Grundy, Managing Consultant, Public Sector, Steria :

• scope of modernisation: joining up; delivering services to citizens directly; shared service agenda[e]
• core IT: CJS infrastructure; modular CJS process design and apps; info access and exchange
• Steria activity: probation 13 years; CJIT 2yrs; Pol N.I. 20yrs; command and control for half the forces in UK.
• clear acceptance of responsibility and risks on both sides; geared to lowest overall risk profile for the programme
• organisational and policy [and user] demands for quick results VS the extended timescales of major change projects; in CJ orgs natural objectives can conflict with local reqs
• Inhibitors to IT [success]: shortage of inhouse resource [so often!]; no access to users [so often!]; over-reliance on 3rd party advisors; model contract form (OGC), which can promote unnecessarily adversarial relationships; procurement process, which encourage big bang projects and inflexible contracts whilst discouraging proper allocation of risk
• Intellect framework
• Critical success factors: full scoping of change factors; clarity of start point; strong leadership and joint governance; full funding; delivery
• ‘We plea not to be held at arms length, but to be really involved and engaged,’

Wiring Up Youth Justice – supporting improved end-to-end sentence management in the youth justice system – Brendan Finegan, Director of Strategy, Youth Justice Board for England and Wales :

• 25% of people passing through Crim Justice system are juvenile
• historically: paperwork in yellow envelopes transferred with the youth (high % lost), soon to be replaced by secure email-based electronic system, Eye

Closing Keynote Address – Lord Laming, Chairman, The Victoria Climbie Inquiry :

• “every crime is a threat to the social fabric” [manages to say this without sounding like an authoritarian!] – criminal acts change social behaviour
• low income not necessarily cause of crime [Telegraph has a recent, controversial piece on the four causes of crime how to reinforce social fabric?
• Organisational structures are less important than values and leadership, however independence of judiciary is key [this is his heartfelt and angry aside: the gov is making a mistake]
• case study: social service departments in councils, for cradle-to-grave service. “one door easily accessible to all”. Now: BlackBerrys, laptops, all info electronic. No paper case files were transferred to the new call centres. The cost saving is going back into service delivery.
• [he’s passionate about improving things]

[The second half of the day allays some of my fears that the human element won’t be entirely obscured under the weight of information collating, analysing and sharing technology. I would have likely to have heard more about the budgets for these programmes: many police services are talking of very tight budgets and some are operating at loss currently. Presumably then, IT programmes must demonstrate cost-saving as well as expanding capability.]

Afterwards we emerge from the conference to find a couple of blocks of Victoria cordoned off by police: part of a building has collapsed and at the time, no-one is sure if it’s an accident or a bombing. The police, fire, ambulance and urban search and rescue teams all seem to be fairly “joined-up”, but by their actions more than their technology, and with their good humour with grumpy commuters.

Modernising Justice through IT, 1 Do touch interfaces increase user expectations?