Have your say at computerweekly.com
How to turn round a demotivated IT team
More responses to Robin Laidlaw, who advocated sacking the
worst-performing 25% of your staff to revitalise the IT department
(Computer Weekly, 11 May)
I hope nobody takes Robin Laidlaw's swashbuckling advice in
Strategy Clinic seriously. I can see the "redundancies will
continue until morale improves" notices already.
However, Laidlaw does have a point. Einstein reputedly said that he
never met anyone so stupid he could not learn something from them.
I believe that there is no such thing as a bad worker; only a
person in the wrong job.
Within a team I have always allowed people to select their own jobs
based on the idea that a volunteer is worth 10 pressed employees.
Sometimes this will result in a change in the structure of the
whole project, but this is probably a good thing.
Anyone who does not fit is then usually only too pleased to find a
home on another project. Such a transfer is without any disgrace.
The costs, difficulties and unpleasantness of more formal action
are completely avoided.
A well-led team will often find innovative ways of solving the
problems and the approach can be so productive that only very few
new recruits will be required. The team is happy, the management is
happy and the customer is happy.
I once lost 40% of a team using this method but (following many
missed deadlines) the next deadline was met. For those who remember
Brook's law - adding manpower to a late software project makes it
later - I offer my own corollary: letting people leave a late
software project can make it sooner.
David Parkinson
Robin Laidlaw's solution to an unhappy IT team is the worst piece
of management advice I have heard for years. Ignoring the fact that
the new IT director will spend the next nine months embroiled in
employment tribunals, dragging the company's name through the
papers and that it would probably result in getting a P45 quicker
than punching the chief executive, it demonstrates a complete lack
of understanding about morale and performance.
If morale is low, that is the fault of the leader of the
organisation, not the staff. It is the leader who sets the
direction, controls the workload and decides on the processes,
methods and tools that will be used.
It is the leader who decides on the workplace environment and sets
pay and rewards to motivate staff and, most importantly, it is the
leader who decides on the type of atmosphere in which their staff
work.
The oppressive rule by threat environment that Laidlaw seems so
intent on creating will never result in a productive workforce, but
merely one that hides its inefficiencies and relies on his
micro-management to uncover improvements. "Heads down, pretend
we're working", will become the motto.
Robin Wilson
Gaining access to public services websites
In response to Antony Savvas' article, which reported on Stockport
Council's new website, which can be accessed using speech for
visually-impaired users, keyboard-only navigation or via digital TV
or Wap (Computer Weekly, 11 May)
I was interested in Antony Savvas' article about Stockport's new
speech-enabled website, and particularly struck by council
e-services head Andrew Kirkham's comment that this provided access
"for all users".
I would be more sympathetic to that claim and the positive tone of
the article if the software were not Windows-only; if the pages
could gain at least two stars from access testing organisation
Bobby; and if the pages were valid HTML, as according to the World
Wide Web Consortium's validator.
DR de Lacey
Command over NHS IT is not controlling
costs
Your report "Contract problems hit £1bn London NHS plan" (Computer
Weekly, 25 May) is just one more manifestation of the difficulties
which plague costly public computing projects.
These problems stem from the "command and control" culture which
dominates relationships between politicians and civil servants, and
on down the implementation chain to the business providers in
building projects and public sector computer projects. It is an
awful but reformable culture.
In February 2004 at the select committee of the Department of Works
and Pensions into public sector computing contracts, chairman Archy
Kirkwood at question 276 in the oral evidence stated, "Political
timetables for computing projects are all about general election
dates. Nothing else matters."
Earlier the committee had heard how difficult it was for the
provider industry to deliver on cost and time unless the
specification was complete. To start work early before "concept
viability" had been properly established by independent assessment
was to ignore the very high risk that the end-users might reject
the project in the final phase of implementation, leaving high
costs and no added value.
With some "buffer time" put in as a mandatory initial phase,
chronic waste would be avoided. Other jurisdictions have already
rejected "command and control" and replaced it.
In the world of contemporary British computing there is the
prospect of waste of possibly as much as £12bn with the present
project for setting up a national electronic database for NHS
patient records, initially funded at £2.3bn. No account has yet
been taken of the huge costs of migration and data loading for this
flagship project.
The big question is when will we see moves to reform the existing
command and control culture, which plunges a wide range of public
sector projects into chaos? Unfortunately, the quality of business
leadership in a democracy is difficult to reform.
Robert Erskine
Charity work can swell the skills on your CV
In Next Move, (Computer Weekly, 11 May) Tracey Abbott mentioned
volunteering as a means to get experience on your CV.
I thoroughly recommend it. The majority of job adverts ask for
experience and a newcomer to the profession has difficulty getting
the first job that will start their career. There are many
opportunities to gain valuable experience and put something into
the community.
I have looked at work with IT4Communities, which has been set up
specifically to connect skilled IT people to charities desperate
for assistance in some aspect of IT.
Its website features organisations from all over the country and
the opportunities to gain practical, CV-swelling, experience is
second to none. There is no pressure to volunteer - I was
registered on the site for almost a year before I found the
"perfect" opportunity.
I also recommend this route for people whose CVs tend to pigeonhole
them into diminishing job sectors. This is a route to show that you
can do more than your current employer requires and also that you
have spare capacity to devote to a new employer's projects.
Charles Peirce, systems engineer
XBRL will modernise financial
reporting
I read with interest Beatrice Rogers' opinion on reporting
standards in the financial sector (Computer Weekly, 25 May). There
is a real opportunity for compliance to become a business
facilitator for financial organisations.
Introducing XBRL as the sector-wide standard for reporting
heightens the business benefits by removing much of the duplication
and manual effort currently undertaken to meet reporting
requirements. This streamlined method of reporting will enable
financial organisations to be more efficient and accurate and, in
the long term, lower operating costs.
Understanding XBRL will be a challenge for IT departments, but
implementation need not be a costly or complex exercise. IT
directors must also ensure they are prepared to meet their new
responsibilities for submitting electronic reports.
It is about time that the vital task of reporting moves into the
21st century and the emergence of this new standard will help
deliver this.
Ann Hosford, business development manager, financial
services, Fujitsu Services
Confusion on chip and Pin rules delays
take-up
Chip and Pin may be a vital component in the fight against fraud,
but if UK banks do not speed up the accreditation process, the
January 2005 deadline will not be met, leaving retailers liable for
fraud.
Only a handful of retailers have achieved Acquirer Acceptance
Testing (AAT) in the past six months. Banks seem to have been taken
by surprise by the level of resource required to achieve AAT, which
in part explains the backlog. However, there are other
problems.
There is clear inconsistency in the interpretation of chip and Pin
rules, not just between banks but also between testers within a
bank. This inconsistency, combined with the fact that the rules are
still evolving and new technology requirements are being created,
means that software suppliers are faced with the constant process
of upgrading or tailoring software systems to achieve
accreditation.
The complexity of the chip and Pin accreditation process simply
adds weight to the resistance by mid-range retailers with
traditionally low levels of fraud who are finding chip and Pin
implementation costs hard to swallow. Yet without widespread high
street adoption, the value of chip and Pin in reducing fraud will
be significantly reduced.
All retailers have to work together with their software suppliers
to leverage bank relationships to increase testing resources and
get commitment to a faster, flexible accreditation process. If this
is not achieved the credibility of chip and Pin will be undermined
before it is even implemented.
Doug Hargrove, managing director, Anker Systems UK
Lessons on Chinook for computing in
aviation
Following your coverage of the Chinook controversy (Computer
Weekly, 1 June) to mark the 10th anniversary of the crash, I would
like to add some comments.
Since writing my report for the Tapper family, I have seen a
newspaper claim that the senior reviewing officer took the line
that "computers don't fly aircraft into high ground, pilots do",
even before the board had completed its deliberations.
If this is true it would of course explain why the technical and
maintenance issues were never followed up, or even properly
analysed, and it also shows that at least one senior RAF officer
may have a lot to learn about developments - both then and now - in
the aviation world's application of computer technology (civil and
military) and the increasing and sometimes total dependence on
computers in the airworthiness chain. However, it does nicely allow
the RAF to cover its tracks for the decision to continue operating
the Chinook Mk2 with all of its problems.
I suggest that computers can cause accidents either directly or
indirectly through all the normal range of RAF and civil accident
categories ranging from human error, such as programming, system
architecture and system mishandling through to maintenance errors
and technical defects such as the integrity of aircraft wiring. In
particular, this applies to the analogue signals where the
integrity monitoring available on the digital bus may be more
difficult to achieve, and is every bit as important as the CPU in
terms of calculating and transmitting the right answer.
I would also suggest that there will even be a measure of natural
operating hazards in the analogue-to-digital and
digital-to-analogue conversion between analogue sensors and
mechanical devices - not least because one of the analogue devices
is the pilot or other aircrew member.
If we look to the future when military aircraft missions may become
effectively managed and possibly even flown by external sensors in
the network-centric warfare environment, the role of computers and
the correct validation and verification of their software is going
to be even more central to the safe and successful completion of
military missions.
I still think that a plausible scenario for ZD576 is that at the
point when the pilots changed the waypoint and started to make the
turn (and the vibration levels increased) the latent wiring fault,
that we are fairly certain from the analysis of the maintenance
documentation was present on the aircraft, caused an engine
run-away.
By the time the crew could get the aircraft and engine back under
control, and by now in cloud and with no terrain references, they
might have had nowhere to go but to try to get over the Mull, which
tragically, they clearly failed to do.
I understand from the comments of former RAF test pilot Robert
Burke before the House of Lords select committee that such a
scenario would fit the latest Boeing simulations.
It is possible of course that the pilots never got the engine under
full control before entering cloud, losing all terrain references
and hitting high ground, and hence did not have anywhere to go. In
this scenario the pilots would have tried everything they could to
avoid the accident and would have been "heroes" in the tabloid
press - not grossly negligent.
Like the reviewing officers I do not know what caused this crash,
and I am certainly not suggesting there was gross negligence on the
engineering or maintenance side. However, I believe that there was
a significant measure of negligence in the board's processes by
their failure to even look at such scenarios.
These scenarios, in my view, have just as many if not more facts to
support them as that of the decision on the part of the pilots to
over-fly the Mull.
John Blakeley, air commodore, RAF, retired
Asset management can help monitor licences
Why is it that large, respectable, even model companies, are still
being caught out by software licensing issues? Chase Credit Systems
found itself facing a £163,000 fine for illegal software use and
the Business Software Alliance and the Federation Against Software
Theft have a seemingly unending list of targets to pursue.
Companies which ignore software compliance can easily fall foul of
the law, but can also risk disrupting business, damaging their
reputation and wasting up to a third of their IT budget on
over-payment of licences, and support and maintenance
contracts.
These risks are unnecessary; automated asset management systems
that are not over-specified can deliver a return on investment
within a single audit.
Glenn Stephens, executive vice-president, business
development, Centennial
Webmail could become an online swag bag
As Google, Yahoo and Lycos engage in open battle to win the hearts
and minds of Webmail subscribers, businesses may end up as innocent
casualties.
As storage capacity soars to 1Gbyte and beyond, information thieves
are eyeing up online e-mail accounts as virtually bottomless
electronic swag bags.
E-mail is already the method of choice for information thieves -
53.1% of business professionals who have stolen corporate
intellectual property have done so using a personal e-mail account.
And with vastly increased Webmail storage limits, it is becoming
even easier for disgruntled employees and those leaving the company
to steal documents and larger files including entire
databases.
Businesses must therefore urgently review internet usage policies,
ensure incident handling procedures are in place and consider
making it corporate policy that all suspected cases of intellectual
property theft will be investigated by computer forensic experts.
Knowing data detectives will be on their trail would be a strong
deterrent.
Chris Watson, senior forensic investigator, ibas
Quality data gives a clean customer record
I read with interest Ed Wrazen's article (Computer Weekly, 18 May)
on the affect of dirty data on the success or failure of customer
relationship management projects.
Data quality is a vital aspect of CRM projects and Wrazen is
correct in saying that data needs to be approached on a strategic
rather than tactical level.
Too many organisations look at data cleansing as a one-time
exercise when in reality it is a constant process. One way of
ensuring it is constantly updated is to ensure that every
interaction is an opportunity to verify, correct, or add
information.
By allowing customers to update their own information you can also
improve data quality with little ongoing investment. Constantly
verified and updated data will ensure that successful CRM projects
provide an accurate view of the customer.
Mark Baker, CRM product marketing, EMEA, PeopleSoft