House of Commons - Administration

Select Committee on Administration Second Report

ANNEX: COMMITTEE DISCUSSIONS

17 October 2006: Opening discussion with Director of PICT

The Committee received a presentation from Joan Miller, the Director of PICT.[85]

Issues emerging in the course of the ensuing discussion included:

The inadequate size of Members' parliamentary mailboxes.
The possibility of allocating e-mail addresses for Members' staff to Members rather than to the individual Members' staff.
The mobile computing project, whether devices could interact with external e-mail addresses, and whether Members' own telephone numbers could be transferred to the new service.
The idea of an electronic room booking system. The current booking system worked well, and any new system would need to be stable if it were to succeed.
Provision of service to constituency offices: there was general agreement that this was in no way comparable to the service provided at Westminster and that this discouraged Members from basing their staff away from Westminster.
The resource implications of changing this situation.
The need for clearer information about what PICT could provide to Members and what it could not.
The idea of a 'regional office' model of constituency support.
Failure by PICT engineers and contractors to keep appointments or to arrange them sufficiently in advance.
The possible provision of instant messaging as a work tool to alleviate e-mail traffic.
The need for a lessons-learned analysis of the Members' IT refresh, including the procurement process.

21 November 2006

The Committee held discussions as part of its inquiry into Parliamentary Information and Communication Technology.

OPENING DISCUSSION WITH MR JOHN MILNER AND PROFESSOR JIM NORTON, SPECIALIST ADVISERS TO THE COMMITTEE

The two advisers gave their initial impressions of the evidence received so far. The following were among the issues raised:

John Milner:

PICT was having to manage a difficult legacy in bringing together disparate staff from different parts of Parliament.

PICT needed to carry out creative work, market research and communications to establish Members' service requirements.

PICT needed to develop a proposed standard service offering to present to Members for their feedback. This could include pick-and-mix elements, but infinite variety was not cost-effective or supportable.

Jim Norton:

Better mechanisms were needed through which Members and PICT could communicate effectively with one another, perhaps including a users' group.
The Committee should try to help PICT identify clear goals, with well-defined standards of service for Members on the Estate, in the constituency, on the move, and possibly in Government Departments as well. Some compromise would be bound to be necessary. A clear route map needed to be developed.
There seemed to be some easy quick wins for PICT: providing public and private e-mail addresses for Members, and increasing server space available. The website currently lacked basic features, such as the ability to track visitors.
It was not difficult to support a wide range of standard hardware. Members should be able to choose equipment from more than one supplier: the resulting competition would be a useful lever to ensure product quality.
A wide range of software and operating systems would be much harder to support. PICT was struggling to support a very wide range of applications, which needed to be culled.
The problem with providing a wide range of peripherals was the cost of the consumables (printer cartridges, etc).
Some oil companies were now getting rid of central procurement altogether - but also of central support.
Well-provided online help could reduce demands on telephone helpdesks. User-configured services (such as broadband provider Plusnet) could be more popular and successful than services which relied on others for support. The Intranet could be used to provide a wide-range of self-service options to customers.

Members:

Members had no choice over the IT equipment provided to them, yet the published figures made it appear to the world as if it were money they had spent themselves.
Much dissatisfaction appeared to concern the provision of peripherals rather than computers: connecting a wider variety of printers to a central network ought to be of less concern than connecting computers themselves.
Members' staff as well as Members needed to be an important part of communications with PICT.
There needed to be benchmark figures for equipment cost and customer service.

The Committee suspended its formal meeting in order to hold informal discussions.

***

INFORMAL DISCUSSION WITH CHRIS MONTAGNON, EXTERNAL MEMBER OF THE JOINT BUSINESS SYSTEMS BOARD (JBSB)

Mr Montagnon made the following points:

There had been three central reasons for establishing PICT: the vulnerability caused by concentrating skills in a very small number of people; missed opportunities to share data and systems; and the recruiting and retention advantages of providing improved career opportunities for ICT staff. But bringing together these disparate groups of people was a tremendous challenge. Improvements were already beginning to be seen: for example, better record-keeping by the helpdesk would allow problems and opportunities to be identified more effectively.

If PICT was to understand its customers better, forums would be needed where representatives of customer groups could meet with PICT. Such forums would establish and prioritise those systems and applications that would help all Members to be more effective and efficient. PICT would agree a schedule and budget for delivering these.

Members might want to think about how to decide which applications they needed. A group of representational users might report to the Administration Committee.

In companies such as Sainsbury's, with 500 remote locations, there was no flexibility as to the service provided. Everything was decided by the central IT service. At Imperial College, although it initially seemed that everyone could have what they wanted, in fact any proposed new applications had to be submitted to the centre to make sure they would work on the network and that the licensing situation was legal.

New services such as wireless access sometimes took a little time to introduce because they needed to be tested to ensure they could not disrupt existing systems.

Services such as a 24-hour helpdesk could be provided, but only at a cost.

For his work at the JBSB, Mr Montagnon had no resources other than himself. His role was to challenge proposals: for example, for the new administrative finance system (HAIS), he had wanted to ensure that it was being customised as little as possible and that management had understood the importance of dedicating resources to changing ways of working as much as to the IT system itself.

Although the JBSB concentrated on systems for the Administrations of the two Houses, the division between the Administration and Members was not straightforward. HAIS, for example, was used to pay Members' salaries and allowances.

INFORMAL DISCUSSION WITH MICHAEL FABRICANT MP, ALUN MICHAEL MP AND ANDREW MILLER MP

The Members invited made the following points:

Michael Fabricant:

A decision had been taken before Mr Fabricant was chairman of the Information Committee that IT equipment should be provided centrally. It would be better for Members to have a choice between either buying a centrally provided package or buying equipment themselves which would need to meet certain criteria if it was to be connected to the Parliamentary Network. Clearly PICT would not be required to support hardware bought by an individual Member.

The standard of centrally provided support had improved immeasurably over the past few years, but was still variable. Better quality control was needed, and a full service at weekends.

Alun Michael:

It was important to look not only at how Members and their staff carried out their business, but also how they could carry out their business. All Members of the National Assembly for Wales used computers, because they had been available from the start. It was also important to consider services provided to the public: such as the facility to e-mail Members.
The variety of equipment being used had brought the previous system at Westminster to meltdown. The principle of standardisation was not the problem, but the quality of delivery. The mobile computing project showed how choice could be made available within a standard.
The IT service needed to be seen as a robust, dependable platform, with built-in security.
PICT worked hard and the service had improved. But Members were difficult to satisfy and there was no obvious channel for identifying service needs.
Members tended to stay stuck at their existing points of competence. Some Members were unaware of facilities already available to them, such as shared drives. Newer Members had higher expectations of IT than longer-serving Members. Peer coaching might be helpful.

Andrew Miller:

There were two key messages from the conversation held between PITCOM and EURIM and the corporate sector:
- Services needed to match Members' working hours. Support for when Members were not at Westminster needed to be strengthened: this might mean 24/7 provision; it would certainly mean better weekend provision.
- the demands Members placed on the system were like joining 650 small businesses to a corporate network. A standard package therefore made sense. Industry experts were horrified that a locked-down approach could not apply in Parliament. The diversity of systems and practices put the network at risk. Examples such as Reuters might be studied as a model.
Conclusions to be drawn were that:
- PICT needed to have a stronger dialogue with the political parties about their software, which was often used on centrally supplied parliamentary equipment. Casework management software had genuine parliamentary uses, but also party political potential. There were also data protection issues which were not always well understood.
- PICT should take up the offer by Intellect, the UK trade association, to conduct a concept viability study.
For PICT to provide leading-edge technology would strain the Parliamentary Network in a way that would be against the interests of the majority. Those Members who wanted to buy their own equipment should not have automatic access to the Parliamentary Network.
The Intranet was not very user-friendly, and could be greatly improved.

Issues emerging in the ensuing discussion included:

The Information Committee had looked at provision elsewhere, including the Scottish Parliament, the German Bundestag, Capitol Hill, and the private sector. Often countries able to start with a 'blank sheet of paper' took the lead: Estonia for example, with its paperless Cabinet.
Structures within PICT might need to recognise the different needs of different customer groups.

***

The Committee resumed its formal meeting.

CONCLUDING DISCUSSION WITH THE SPECIALIST ADVISERS TO THE COMMITTEE

Jim Norton:

The suggestion that Members could be responsible for hardware support and PICT for software and network support would be difficult to manage. For example, if a Member lost data from a hard drive, who would be responsible for seeking to recover it?

Good business practice for bringing about IT change suggested that 20% of the resource needed to go on the IT itself, and 80% on the people using the systems.

By improving the Intranet so it could be used for work functions, it might be possible to save administrative resource in the long run.

Some training and information for users could be provided over the network itself.

John Milner:

Imposing control on a system in the way that Imperial College did could save a little money, but new technology allowed for some flexibility. Coherence was more important than control: this was the approach being adopted at Cambridge.
It was possible to have a small portfolio of PCs, peripherals and software which would provide email, calendar document access and management and this portfolio could include certain key applications (eg in support of casework) in a coherent fashion. In this context it should be possible to cater for work within the Estate and within the constituency with equal facility.
This needed to be considered alongside PICT's back office work on applications that might also need to be accessed by Members, but were primarily aimed at administrators.

5 December 2006: Informal discussion with Richard Allan

Richard Allan, Head of Government Affairs, Cisco Systems, and former Chairman of the House of Commons Information Committee gave a presentation to Members of the Administration Committee as part of the Committee's inquiry into Parliamentary Information and Communication Technology. The presentation concentrated on technologies used by staff at Cisco:

Mobility — access to a virtual network via a software token; a trusted wireless system; PDAs; soft phones; instant messaging; desktop videos with telephony; and internal television. Telephony was often forgotten but a critical tool. Cisco employees took their phone number and extension with them wherever they went in the world - by using VoIP, calls could be routed internally to reduce the cost of calls. Instant messaging was used in a variety of situations: to avoid telephone tag when wanting to talk to someone; to ask questions privately of people on a conference call; and to set up meetings. Internal television could be used to access meetings and also to convey important messages from executives.

Collaboration — audio and video conferencing using high quality screens and spatial stereo. Audio-conferencing had become a standard tool in the business world; Parliament had perhaps needed it less because of the expectation that people would generally be prepared to come to Westminster. 'Videopresence' was a better than HDTV solution which allowed people in two different rooms anywhere in the world to feel as if they were occupying two halves of the same table. 'Virtual Margaret' was a worker who had a desk in Cisco's California office but who actually lived and worked in Texas. Her desk in California had been replaced by a screen.

Security — backup via a background application; encryption to ensure stolen or lost equipment does not give up confidential information; and 'clean' equipment provided to travellers to avoid business secrets being stolen in transit.

Management — web-based expenses, supplies and pensions systems; customised off the shelf HR and recruitment services; and on-line training. An advantage of providing training online was that you could ensure complete coverage.

Flexibility — the real challenge for corporate networks which seek convergence of systems used by staff. Within Cisco, all systems had the same look and feel wherever you were.

A discussion followed, during which Richard Allan made the following comments:

The common platform had been introduced at Westminster as an attempt to resolve the complete inconsistency of equipment and software being used by Members. The platform had been creaking and it had been vital to re-establish stability.
Employees at Cisco expected to receive the same quality of service wherever they were based - including at home. Citrix had given Members a bad experience and led to an expectation of a poor quality constituency IT service.
It had been recognised in the past that providing a service to constituency offices would prove a challenge. Users of the VPN should expect the same level of service as was provided in Westminster.
Within Cisco, there were no restrictions on the applications staff could install on their machines as long as the security platform could support them and they did not disrupt the network. But Cisco would only rebuild machines to the standard build if things went wrong. It was not easy to empower technicians to say no to powerful clients. There was no reason why Members should not have flexibility in the applications they could install on their centrally provided computers.
As in every organisation, there were different categories of user with different demands: some Members used standard tools; some used non-standard but supported tools; and some used non-standard, non-supported and non-approved tools.
Where connections were slow, the challenge was to identify and remove the bottlenecks. But a central IT service could not be expected to improve the speed of a broadband connection itself.
Members should not expect to be able to connect their own devices to the Parliamentary Network. The flip side of this was that PICT needed to get the contracts with its hardware providers right to ensure that the equipment being supplied met Members' needs. The House of Commons had an unusual capacity problem because of the electoral cycle: it might be necessary to pay a premium to ensure adequate service levels immediately after an election.
Cisco had saved 40% on its real estate costs by investing in technologies to allow remote working, shared desk space, and shared printer systems. No-one at Cisco had a desktop computer, but rather a laptop which could automatically identify the nearest printer. It might be possible to save space in the House of Commons by providing wireless offices with shared desk spaces and printers, for example for groups of Members' staff; and by enabling people to work offsite. But this would require significant cultural change, which might be difficult to bring about as there were not the same competitive commercial pressures to react as in a business environment.
If PICT was having difficulties in setting up a secure wireless network, this could only be a capacity issue. There were no technical or security difficulties - except the fabric of the buildings. A big team could set up a network quickly; a small team would be slower. Other government departments were also struggling with similar issues.
The ICT set-up in the House of Commons combined 646 SMEs and one large corporation. Suppliers tended to be focused on either one or the other, but not both. In other words, the market did not easily provide for the kind of set-up found at the House of Commons. Quality suppliers of services to SMEs tended to be locally based; but the House of Commons was a national institution.
A possible model to investigate as a comparator would be a retailer with a lot of branch shops.
There were logical issues around the use of the Parliamentary Network for Members' party political work which could be explored. Why was it not appropriate for the House of Commons to provide shared server space for Members of a political party, but ok for an e-mail with an attachment to be sent over the network to 350 individual Members? Applications, casework software for example, could be used for both parliamentary and party political purposes: how should this be provided and monitored?
Most Members found the governance structures for the House opaque. The Information Committee had been the place where pressure was put by Members on the House Service to achieve progress.
A single ICT function made perfect sense, but there could be specialist groups within this, with one concentrating on Members' services.
Members might have little clout with providers such as BT when moving office. Problems might be more easily sorted out centrally through PICT itself, if a special contact point was developed.
It was an accepted wisdom that investment in ICT delivered productivity benefits. But ways needed to be identified of measuring productivity.
It was important to move forward with consensus, rather than having a queue where the noisiest customer won - which would lead to junk. PICT needed to have the authority to ration and to authorise. Allowing Members to have non-standard equipment was much more of a problem than allowing them to have non-standard applications.
A corporate extranet, as used by Cisco to allow other organisations access to sensitive information, might not be suitable as a model for interaction between a central parliamentary service and Members as individual business units, because Members would want transactional as well as informational services.

12 December 2006: Informal discussion with Andrew Hardie

Mr Hardie made the following points:

For cultural and historical reasons, ICT provision at the UK Parliament was more complex than in many other parliaments.

PICT was working hard to achieve a networked model of computers and storage, but this model was fifteen years old. It was difficult to get such a model to function remotely, especially internationally.

There would be a cost to improving the system - most of the investment would need to be spent on the people providing the service rather than on the technology itself. However, with a simpler service provision model such as that offered by Web 2.0, it ought to be possible to reduce the overall number of people needed to provide service support.

Most services could be provided over the Internet using Web 2.0 - which was a different, flexible, way of providing applications and information and storing data. About 80% of ICT needs could be met using this approach.

Flexibility was a significant advantage of this approach. At the user end, all that was needed was a computer with a browser and an Internet connection, with no need for a standard build. This would reduce central maintenance and update costs significantly. The use of open source software could also reduce licensing and licence compliance costs.

A problem with the networked system was that it required lots of different kinds of server, which were difficult to support. The server requirements for a Web 2.0 based system were much simpler.

It was easier to solve speed problems caused by server overload in a Web 2.0 environment, but not problems caused by a slow broadband connection.

It might be possible to run the system in parallel with the current parliamentary network in order to test the technology. 2-3 years remained before the next planned upgrade of Members' computers; this time could be used to plan and test new services. Online word processing software would be a sensible area with which to experiment, as there would be little interaction with existing systems.

Innovation could not be outsourced. Any contractual arrangement that inhibited innovation, either internally or by other contractors, was undesirable.

The current Parliamentary ICT system provided a number of different services in different ways to different people: the system was brittle by design, and was very hard to support as a result. A balance needed to be struck between providing only the lowest common denominator and always seeking to respond to the highest plea: it would help to reduce the number of different classes of user with different needs.

30 January 2007: Concluding discussion with Director of PICT

The Committee held a discussion with the Director of PICT as part of its inquiry into Parliamentary Information and Communication Technology.

The Director of PICT told the Committee that several main areas had been identified where improvements could be made:

Service desk: improving the quality of the service provided. PICT aimed to provide a one-stop shop for all issues, including those currently dealt with by contractors, and were considering how best to provide a 24/7 service. There was a continuing programme of recruitment and training of staff. Whilst the time taken to answer calls to the service desk had been reduced following the improvement programme put in place in July 2005, the aim now was to improve the quality of the first-time response to Members' queries.

Constituency support: although PICT was not funded to provide a full range of services to the constituency offices, PICT were investigating what services could be provided and what funding would be required. These included: enhanced remote control access by PICT engineers; online ICT training and mentoring for Members' staff (both of which were relatively low-cost); and possible regional response units (which would be high-cost). Other than resources, the main constraint in this area was the need for clarity over what could be provided during a dissolution of Parliament.

New developments in technology: eg the availability of video conferencing and internet telephony for Members. Members could be informed of new developments through various means such as papers, workshops and exhibitions, taking into account the wide range of experience of ICT amongst Members.

Other common queries from Members which were being investigated by PICT included: standardisation versus customisation of the desktop environment; provision of web-based services; and provision of storage.

Issues raised during the ensuing discussion included:

The need to define what PICT should provide for Members in terms of a service requirement rather than a technical specification.
That further standardisation of the system would be unlikely to meet with support among Members.
Whether a web-based service could replace the current VPN-based system for remote working.
The service provided by the Service Desk, and the possibility of introducing a dedicated phone line for Members.
The possibility of providing a DVD recording facility for Members, such as was available in the Bundestag and Congress.
The possibility of providing a second private parliamentary email address for Members.
Systems for ensuring that e-mails sent to Members' staff reached the office of the intended Member, for example where a member of staff had stopped working for one Member to work for another.
The cost of the mobile computing devices available to Members.

The Director of PICT responded by stressing the need for a regular forum in which PICT could demonstrate technical solutions to Members and Members could explain their requirements to PICT. PICT needed to take time to understand better the daily and weekly working patterns of Members and their staff. PICT staff hoped to visit constituency offices to improve this understanding. It might be helpful to continue the work begun in the survey of 22 constituency offices conducted in November 2005.

Members had a variety of requirements, from the simple to the complex. PICT was looking at ways of offering a range of service options with a corresponding range of costs, from the basic (bronze) requirement to the highly advanced (platinum). Clear processes for problem escalation were well understood by service desk staff, but how these staff and technical teams interacted could be improved.

The difficulties experienced by staff involved in the establishment of PICT had largely passed. There were opportunities for cross-fertilisation created by being able to 'see the whole pot in one place'. For example, PICT had been able to draw on the experience of former Lords IT staff to offer online training to constituency offices. A main aim was to ensure that people within the Department talked across the management structure, and that silos were avoided.

On the specific issues raised, the cost of the mobile computing service provided to Members included back-office support and warranty elements as well as unlimited use for a fixed monthly charge. Replacing the VPN with a web-based system was possible but would need the technical base to be re-established.

The Director of Operations, PICT, told the Committee that the only obstacle to providing a second email address for Members was capacity: funding would need to be found for the provision of support for 646 new addresses on the system. E-mail addresses could be configured to ensure that Members retain control of emails sent to their staff. This could be offered to Members each time they requested a new email address for their office.

PICT looked forward to receiving guidance from the Committee on priorities.

85 Available in the House of Commons Library, and online at http://www.parliament.uk/documents/upload/PI21AnnexPICT.pdf Back