Men and machines: which is master which is slave?

 

Robert Henderson

The British public has become weary of state-run  computer projects either failing to be completed  in whole or part after immense amounts of  public money has been spent or public service  IT systems being incompetently operated  where they are put into use. Common as such failures are, the present fiasco with the Her Majesty’s Revenue and Customs (HMRC – the British equivalent of the USA’s IRS) attempting to recover tax from 1.5 million people that has been under-deducted in the past two years because incorrect tax codes were issued is probably the most spectacular operational failure to date besides having a projected final cost of £389 million compared with the original cost projection of £140 million and being introduced a year late.  It is a classic instance of IT failure. There is the cost over-run, the delayed implementation, the failure to design the system properly, the inadequate training of staff and the seeming inability of anyone involved with the project to understand the system in its entity.

The operational failure was unsurprising because as an investigation by the National Audit Office (NAO) “established that the new system was put into operation with more than 50 identified technical defects. Up to 500 staff had to be reassigned to carry out the “manual workarounds” that had to be devised to correct the errors, which included creating “erroneous work records” for many taxpayers ” and  “The [NAO] audit  discovered that data on taxpayers was loaded on to the new computer system last year without being checked for accuracy, raising the risk of workers being issued incorrect tax codes”

When the system went live in 2009 “HMRC officials expected the new system to generate about 13 million new tax code notices.” In the event, it produced almost 26 million.

“According to the National Audit Office, a “significant proportion” of those codes were wrong.

“Many of the incorrect notices were generated because the new system allocated a new code for each job that a person had done. That led to many people receiving several different tax codes.” (Ibid).

The failure of the HMRC system was the result of (1) the IT professionals who designed the system not understanding the requirements of tax collection and/or the HMRC people in charge of the project not understanding computerised systems and (2) the staff running the system, especially those responsible for inputting of data, being inadequately trained.  The circumstances also suggest that either the HMRC management in charge of the project allowed the system to go ahead despite knowing that there were many program shortcomings or the IT contractors did not properly inform the HMRC management of the defects in the system.

The consequences were the diversion of   500 people to make manual adjustments and   God alone knows what further expenditure of time and money in the future to deal with the millions of incorrect codes being issued many simply because the system had not been designed to link up multiple employments.   Using a centralised computerised system has created a degree of chaos which no manual system could create nor in all probability that which could be created if computerised wage and accounting systems were used by HMRC at a local level to calculate tax because with small local systems problems such as different codes issued for multiple employments would be almost certainly spotted quickly.

Such monumental IT failures are symptomatic of a general problem with digital technology, namely, our ever growing reliance on it and the increasing complexity of the hardware and software   which outstrips both the expertise of so-called IT professionals and the abilities of the average person to simply operate the programs.

For the private computer user the frustration is even greater, because at least an organisation of any size will either have its own in-house computer expertise or can afford to buy in IT expertise to deal with IT problems.  The private individual often has no access to such expertise because the cost is prohibitive, but even where cost is no barrier the expertise of supposed experts is often found wanting.  (I have paid for two new computer systems to be installed in the past ten years and on both occasions the supposedly simple task took multiple visits. Both occasions involved large well known retailers).

The general consequence of our ever growing reliance on digital technology is that we are increasingly being controlled by the needs of the technology rather than using technology to serve us.  It is very difficult to escape such control. If a person is in work they will almost certainly have to use it. If they are in education they will definitely have to use it. Even if a person does not encounter digital technology  in their work or education, they find it increasingly difficult to avoid it in their private lives even if they refuse to use a computer or a mobile phone, ,  not  least  because  businesses  and    government  increasingly  require those dealing with them  to  do so by computer.

But Increasing numbers of people do buy computers and other digital equipment for private use.  Why do they do that if the machines are so unreliable and demanding? Simple: once a significant minority uses a technology it becomes increasingly difficult for the rest to resist.

We have long passed the point where a handwritten   document is likely to be read by most people in business.    Now, except   between social contacts,    everything   must   be word-processed to be acceptable.   A word processor or access to one has become a sine qua non for anyone who wishes to be taken seriously.   Even amongst private individuals a letter   is increasingly seen as unusual or even quaint.

With emails, we have not come to the stage that telephone   ownership reached a quarter of a century ago when not to have a phone became considered eccentric, but we are rapidly moving towards it.  Employers increasingly wish to contact employees by email wherever they are and this means the choice is often between having a computer and email at home and not having a job.

Those  with  school  age children,  whatever  they  think  of  computers,  find it next to impossible to deny their children  not only a computer but access to the internet,  both because the  children want it to match their peers and  because  they  have  been  brainwashed into believing that a computer  is  a   necessary educational tool.

In  short,  people are increasingly being  driven  to  become   computer  owners    not because they actively want  to,   but   because   they  feel  isolated and excluded  if  they  remain computerless.  Again, as with the analogy between  telephones   and emails, within the foreseeable future,  someone without a   computer is in danger of becoming in the eyes of the majority   as  much  as  an  oddity  as someone  without  a  TV  is  now  considered.

Despite all these pressures, there are still a large number of people in Britain who have remained distant from the digital world. According to a recent Office for National Statistics report nine million British adults have never been online. Nor is this simply the elderly for “Only 45 per cent of adults without any formal qualifications had used the Internet, compared with 97 per cent of those with a degree”.   It is worth bearing in mind that approximately ten per cent of the British population (6 millions) have IQs of 80 or less, and an IQ of  80 is the point at which most psychologists working in the field of  intelligence testing think a person would struggle to live an independent life in an advanced modern society such as Britain.

It is unreasonable in a civilised society to simply hang the computer ignorant or the intellectually underpowered out to dry as digital technology looms ever larger.  Yet that is precisely what is happening.

Technology as magic

To master computers to the degree where a person does not lie helplessly in the hands of experts is a demanding and   continuing   task.   It is unlikely that many could or would manage it without making computers their profession.   In fact,   even   supposed computer professionals   are   only   knowledgeable   in   their specialist areas:   a   hardware   specialist has no deep knowledge of software and vice versa.

The  science  fiction writer Arthur C Clarke  pointed  out  a  good few years ago that there comes  a point with  technology  when it became indistinguishable from magic  for all but  the initiates. The dangers of that are obvious: that which is not   generally understood gives the few who do understand a great power over those who do not.  That  potentially awards  private  corporations and governments a great stick with which to beat   their  customers  and citizens into  submission,  either  for  profit or political power.

Where the technology is as vital and central to a society  as  computers  have  become,   there  is  the  further  and  more  fundamental  risk   of society reaching a  state  where   the   technology  can  no longer be either properly  maintained  or  controlled.

Computers are like no other machine ever invented.  They have   a unique combination of an unparalleled public and private    use and   a central importance to economic activity   and        public administration.   The  potential  penalties  for  the   failure  of these machines  are vastly greater than  for  any  other  piece  of  technology.   Not  only  can  an  immediate   application  of a computer be ended,  as can happen with  all  machines,  but  computer users also  risk  losing  networking capacity  and,  if they have not useable backed up copies  of their computer data,   the loss of their entire records   and conceivably the loss of the means to continue their business.

Computer users are also vulnerable to outside sabotage though   hacking and viruses.   No other machine has ever exposed a society to such risks through its ubiquity and vulnerability to outside influences.

Computers are also vastly more demanding of time and trouble than   any other machine used by the   general   public.   Technological change has been making increasingly severe demands on human beings for around 300 years.   There was change before then of course, but it was slow and most people could live their lives without having to adapt to radically new ways of living.

The Industrial Revolution changed that and arguably someone living between 1815 and 1914 saw more radical technological qualitative change than any generation before or since.  But   that change   was the difference between living in a still   largely pre-industrial society (in 1815) and an industrial society in its early middle age (in 1914).   Moreover, the   change did not require the vast majority of the population to master complicated machines at their work, let alone in their own homes.

In 1914 the most complicated machine most people would have   had   to operate was probably the telephone and vast swathes   of the population would not even have had to go that far into the world of technology.  Not only that,  because  machines   then  were  either  mechanical or part  mechanical, that is,  not  electric,   just looking at the way a machine was made  often   allowed the intelligent  observer to have a fair guess at how   it   worked  and  to  see   what   had  gone  wrong   if   it  malfunctioned.    Even  work-related machines which  required  skilled  operators,   such  as  machine  lathes,   were   not  fundamentally difficult to understand, although the dexterity  required to operate them often took time to acquire.

In general terms, things stayed much the same until the age of the personal computer, and even beyond.   Machine became more and more predominant in advanced societies but they were  not,   in most instances, complicated to use.  This was particularly true of those machines used in private life.

Telephones just required the user  to dial;  washing machines  gave  you  a dial with a program on it and  a  start  button;  televisions  and radios required simply needed switching  on;   cars were simply  designed to travel.   Even today, when they are   increasingly packed with microprocessors and menus of function “options”, such machines are simple to use compared with a computer.

When did the computer rot set in?  It is a stunningly recent phenomenon.   Most people even in the West   would not have used a computer before 1985.  Probably a majority had not   done so by 1990.  By the end of the 1980s the nearest most would have got to a computer would probably have been   bank   ATM machines.  The internet was esoteric and laborious, the   web barely more than a gleam in Tim Berners-Lee’s eye.   Even  in  the  world  of  employment  computers  were  still   used  sparingly.

As with computers, so with the other machines which cause   most grief now?   The mobile phone was a status symbol and the size of a brick, while landline phones were still phones boringly restricted   to   simply   phoning   rather   than   mini-computers with a tendency to bemuse.   Microwaves had a simple   choice of power.  Refrigerators did not offer to   remind you of what needed to be ordered.  TVs   tended to simply work when switched on.

In the past 20 years all this has dramatically changed.   We are now in a world in which computers are absolutely integral to business and public administration and they are now the norm rather than the exception in homes.  For most people, it   is literally impossible to escape them.   Worse, they have become ever more complex and demanding to use and invade ever   more of our lives as microprocessors are inserted into the most unlikely things such as clothes.  In fact, all machines are more and more demanding.

 The constant learning process

Personal computing began in the mid-seventies.   A person starting them would have had to learn the BASIC programming language.     By the early eighties they would have been using DOS (Disk Operating System) which required the user to write lines of code to do things such as copy a file, move a file, delete a file. . By 1990 Windows expertise was necessary.   Since 1990 successive   editions of Windows have varied considerably from   the   previous version requiring further learning.

What  goes for  operating systems applies also to most  other  programs,   which   when  they  are   upgraded   often   bare  surprisingly   little  resemblance to the  version  prior  to   them.  Certainly,  if one moves from an old   program   to  a version  which has been uprated twice,  the chances are  that  knowledge  of the original program will be of little  use  in understanding the new one.  In addition to this burden of learning and frustration, one added the need to familiarise oneself with the Internet and the Web.

Apart from the effort needed to constantly learn new programs   and to attend to such things as installations of software and  hardware, the other great drawback of computers is the amount  of time which can be spent on purely maintaining and learning   sufficient to use them.   It is all too easy to find a day or two slip by just sorting out a single relatively simple computer problem or learning how to use a new program.

 The nature of what is to be learned

The burden of learning is   especially heavy because of the nature of that which is to be learned. This is what might be   termed   dead information.   There is no intrinsic interest in what is to be learned. It is merely a means to an end.  To operate a program all that is needed is knowledge of   the menus and function keys.   That is precisely  the  type  of  information  which  is least palatable to  the  normal  human  mind.  Hence, it is the least easy to learn for most people.

The computer is in effect forcing human beings to act like computers, something utterly alien to them.   Intelligence  is  of  little  use   when  it  comes   gaining  knowledge  of  the functionality of  programs (how  a  person  uses  the  program  once the functionality is  learned  is  a  different  matter).      Computers are information   driven   machines.  Put the most intelligent man in the world before a   computer and he will be utterly helpless if he has no   computer experience.  Even if the man has some computer experience, he will be as incapable of using an unfamiliar          program as the dullard.

That  computers are function rather than intelligence  driven   is  objectively  demonstrated  by the fact that all  of  what  might be called the administrative  operations of a  computer  – file management,  loading of programs etc. –  could be  done  by a computer program.

  The substitution of function for intellect

When I watch the young using computers, obvious or disguised in the shape of phones and the like, I get a feeling of deep unease.  They  so  obediently pull down  menus  and select  options  that I wonder at the difference between them  and  a  robot.  The  machine is driving the human being at  least  as much  as  the  human  being is  driving  the  machine;  brute  machine functionality is replacing intellect.

There is only so much any human being can learn, both in terms of time and mental impetus.  If increasing amounts of both are required by computers simply to operate them, where           will that leave intellectual development?    Worse, will the   ability to operate machines become seen to be the most   important activity of human beings?

The myth of youthful expertise

It is true that those who have grown up with computers are more comfortable with the machines than those who came to them in adult life – the latter still comprise, incidentally,  more than 50 per cent of the population.   However, the idea that the young are have any deep understanding of computers is a myth.

The young know how to use the internet and web can work a word processor and use programs which really interest them such as social networking sites and games.  But  let  their   computer develop  a  fault   which  renders   Windows  unstable or unusable or  a piece of hardware  fails, and  they are,  in most cases,  as helpless  the  generations  which did not grow up with computers.

What the young do have which older people do not have   is group knowledge.  A schoolchild of today can call on the   computer knowledge of their peer group and the assistance of teachers. Those  a little older who are in work  still  have   their  peer  group  to  help them   if  they  get  stuck.  In addition, if they work for a large employer they can call on   the expertise of the employer’s IT department or service   contractors.

Computers have only been in schools since the mid-eighties.    Anyone over the age of forty will  not  have  a  peer group on  whom  they  can  call  for     assistance  with  computers  (and  other machines)   because  almost  all  of  those  they  know  well  will  be  of  their  approximate  age  – few people have  close  friendships  with  those who are  much younger than themselves – and the  people  who  are  their age will have little computer  experience  or   knowledge.   The best they can hope for is assistance from    their children if they have any, and then it is pot luck as to how computer competent those children are and how willing   they are to help the parent.   If an older  person  has  no compliant computer literate children and  does not work for a  large employer,  he or she will  be utterly isolated from the  knowledge   needed   to  deal  with   even   basic   computer  developments.

The true cost of computers

The  common  arguments for computers in business  and  public   administration  are  that  they increase  efficiency  by  (1)   reducing staff,  (2) allowing faster working,  including  new working  practices,   and (3)  producing  information   which  would previously have been impossible to produce.

Despite  the fact that experience has frequently shown  these   arguments  to  be  invalid,  they  continue  to  be  made  in  religious fashion by both the computer industry and the  many  uncomprehending executives who have bought into the  computer  dream  without  troubling  themselves  with   even  a   basic  understanding  of the subject.  (Thankfully we have at least  got past the pipe dream of the paperless office).

When  computers first began to be generally used in  business  and   public service the overwhelming majority of the  people  making the decisions about their introduction –  businessmen,   politicians and senior  public servants –  had absolutely  no  experience of computers,  mainframe or otherwise.   Not only did they lack experience they were terrified of anyone who did.  This meant that the snake oil salesmen of the computer   industry could sell them virtually anything. Things have not   changed dramatically even now.  That is one of the prime reasons why so many large computer systems, especially public service ones, go to Hell in handcart.   (One of the  lesser known  laws  of C Northcote Parkinson  states  that  the  time  spent on discussion of any agenda item in a committee meeting   is inversely proportional to the knowledge of those present ) Discussion  of  a  large computer system  last  five  minutes   because only the one expert present understands what is going  on;  discussion  of  the refreshments to  be  served  at  the  meeting lasts 45 minutes).

Buying computer systems is an expensive business for any organisation. Buying a badly designed computer system is  not merely  expensive  but potentially  disastrous,  because   to  prepare   for   its  introduction   the  structure  of   the  organisation  will  have  been changed and  the  old  system,  manual or computer will have been dumped.  So, when the new computer system malfunctions neither can the organisation use   it or revert to the previous system.  The same applies, at least temporarily, when a computers system simply goes down for a limited period. Costs continue, but work ceases.

But even where the computer system is properly designed and   works efficiently the costs are immense.    The readily identifiable costs are frightening enough.   There  is  the  initial  cost of the system (many public service systems  run  into billions),  the cost of its maintenance, the cost of its  upgrading,  the  cost in time and money of the   initial  and ongoing training of staff and  the cost of the employment  of   new people in new supervisory posts to oversee the system.

There are also the less readily quantifiable costs. Computers    generate   vast   amounts of data which   is   distributed   promiscuously.  This occupies time which would   not    otherwise be spent.  Email means that substantial amounts  of  time is spent by employees  answering or even simply deleting  emails,  not  least because everyone is  much more  prone  to send an email than a letter.

More  generally,    the   structure  of  working  within  the  organisation  will tend to  shape itself around the  computer  and  employees  will begin to develop a  different  sense  of        priorities   with  the  computer  looming  largest  in  their  mentality  rather  than  the overall needs and  ends  of  the organisation.

How often is any proper cost-benefit analysis done on the utility of a computer system? Very rarely,  not least because  once  a  computer  system  has been put  in  place  with  its concomitant staffing changes,  it is a daunting and hideously  expensive  task to change matters. Nor will those who have made the decision to purchase a computer system willingly admit they were wrong.  Not only that,  but we still have the  problem,  which will never go away,  of the people  with  the         power   within  organisations  knowing   insufficient   about  computers  to  make any meaningful decision on the  value  of  computers  to the organisation.  This ignorance also robs the  powerful  of  the self-confidence to challenge the  need  for  computers  or  the nature of systems being proposed  by   the  “experts”.

The problem of long-term data storage

Data   storage   bids fair to cause fundamental problems in both the medium and the long term. This is because the nature of the storage media, both hardware and software,   is changing so rapidly.

Even in the twenty odd years of the widespread use of   the personal computer, we have already had storage on hard   disks, 5.25″ disks, and 3.25” disks, zip disks of various sizes, CDs, DVDs and USB memory sticks, with the saved in a wide variety of formats depending on the software used.)   To the problems of access to electronic data may be added   the fact that fewer and fewer documents (as a percentage of the total number of documents created) are being and will be   saved in hard copy form.

The implications for the future are profound.  Until now historians have been able to look at documents because they   were written or printed.   In the future,  historians (or any  other researcher) will find either no documents or ones which  are  inaccessible  because  they  are  only  in  an  outdated   electronic format.

There is also a standing temptation for those in positions of   power   and influence,   especially   politicians,   to   deliberately destroy any record of their misbehaviour.   This is made vastly easier if the documents are only held on computer.

The failure of the market.

There is no better modern example of the market failing to provide what the customer both needs and wants.     If it was  driven by the customer,  the computer industry would  produce  hardware  and  software  which  was  easy  to  install,   had   continuity  of use,  was simple to use and was  supported  by   adequate help lines and manuals.  The industry signally fails to do any of these things.

Hardware and software are of course purchased in ever greater volume and computer services,   including   maintenance,     continue to swell.  But that is not an indication of computer  satisfaction.  Rather,  it  is simply  a  reflection  of  how   computers  have become an inescapable part of our lives,  not  only as obvious computers but also in the guise of   so  many  of  the  other machines we use – everything  from  phones  to   intelligent clothes.   Business and public administration has become so dependent on their use that they cannot do without them.   That  being  so,   whatever  is  on  offer,   however   unsatisfactory,  is  bought  out of  sheer  necessity.    The   computer companies have the modern world over a barrel.

The  power  deriving  from the ubiquity and  utility  of  the   computer is bolstered by the fact that the computer  industry  is  in  some  respects  a natural  monopoly.  Once a single operating system (OS) gained dominance,   the chances of any other system effectively competing were very small.   This is because the weight of programs available to run under the dominant OS soon became much greater than those which could be run under any other OS.    Thus, it became inefficient to choose any other OS.  That in turn meant most of the software   was written in a way to make in “friendly” to the dominant OS’s systems users.  This further excluded OS competitors and  the  software  to run under them  because  users,  especially  employers,  did  not want to spend the  time  training  their  employees on completely new systems.

 What needs to be done?

In a sane world governments would act to prevent the introduction of massive centralised   networked   computer systems providing vital services  because  of  the  dangers  of  a  general  failure of  such systems from cyber-attacks.  This applies especially in public services but also to privatised business such as the water and energy companies and to private enterprise businesses such as banks.  Governments should also take the lead in both costing their own computer projects honestly (see above) and encouraging large businesses and not-for-profit organisations to do the same.

As for the private user, they will be asking themselves questions such as these: why should using a computer be such a demanding and hit and miss business?  Why should we tolerate a machine which is not in any meaningful   sense “fit for purpose”, in the words of our current consumer   protection law?   Why do we allow  ourselves to be  fobbed  off  with  precious little information about how  to  operate   programs, both in terms of free instruction by the vendor and  in the laughably inadequate instruction manuals,  whether  in   hard copy or from the all too frequently risibly named “help”   functions?  Why do we put up with hardware and software which    go out of date (and hence become unusable) in a year or   three?  Why do we have to pay an arm and a leg for computer training   or repair? These problems could be substantially ameliorated by legislation to ensure that:

1. Windows (or any other operating system) is written so that any version of Windows will encompass every previous version of Windows. Users should be able to choose from a menu which version of Windows they use. The already vast  and  rapidly  increasing storage capacity on computer drives means that  an  increase  in  the  size  of  the  programmes  would  not   be  impractical.  The effect of this would be to reduce the need   to constantly learn to use new software.
2. Windows or any other operating system is written to ensure that all software written to operate under any operating system can operate under any version   of the operating system.  The effect of this would be to (1) reduce the  need   to  learn  new  software and  (2) reduce the  need  to  buy  new  software .
3. Windows or any other operating system is designed to accept any peripheral regardless of age.
4. Hardware is designed so that any hardware can continue to be used for as long as it works.

Are any of these four things likely to happen? Sadly, no.  The problem is simple:  no single country, not even the USA,   could   insist on such laws globally.

There is one thing the government of any advanced country can and should do, create circumstances in which those who cannot come to terms with digital technology can live in an ever more computer controlled world. They can do this by maintaining non-computer access to state funded organisations and forcing through legislation larger businesses and not-for-profit organisations to do the same. A good start would be to knock on the head the clearing banks’ proposal to stop clearing checks by 2018.

Robert Henderson September 2010

See also

https://livinginamadhouse.wordpress.com/2011/07/01/robotics-and-the-real-sorry-karl-you-got-it-wrong-final-crisis-of-capitalism/

https://livinginamadhouse.wordpress.com/2010/10/13/the-geepees-a-cautionary-tale/