PART II

Computers and Change

Richard Coyne

Department of Architecture, University of Edinburgh, Scotland

Sidney Newton

Department of Design Studies, University of Western Sydney, Australia

Sally McLaughlin

The Waikato Polytechnic, Hamilton, New Zealand

Ahmed Jumani

Sydney, Australia

This report examines the changes to design practice brought about by the introduction of computer technology. We conducted an interview survey of practitioners who use computers and investigated how computerisation is affecting work skills, what kind of stresses it is imposing on the firm, how firms make decisions about computers, what are the management and organisational implications of using computers, what kind of physical work environment is produced, and what trends and research issues need to be addressed. We report on the survey here.

We report on an interview survey of how professional designers, such as architects, use computers in practice, and how computers affect practice. As with any survey we can only report upon a slice in time of the state of practice. At the time of the survey we had to explain terms such as "multimedia" and "Internet" to some interviewees. At the time of publication new terms will have gained currency, such as "the information superhighway," "World Wide Web," and no doubt others. The rapidly changing nature of information technology provides the focus of this study. The "timeless" element of our study is the fact of change in the context of information technology (IT), and how practitioners cope with it.

We began the study with three key hypothesis. The first was that the influence of computers on design practice exceeds that of other technologies. Computers become pervasive. They touch on almost every aspect of an organisation: management, public relations, clerical tasks, and in the case of design practice, documentation and design. We aimed to discover the extent of this change. Second, we also hypothesised that computers are implicated in major changes to work practices. Computers affect the way people work together, they affect attitudes, and they impinge on power structures within organisations. We wanted to discover whether these changes impose any stress on firms.

Third, we hypothesised that computing exerts only a minor influence on the activity of design itself (or design decision making), other than in areas with a substantial engineering orientation. However, computers exhibit many indirect impacts on design by affecting designers, changing the dynamics or organisations and changing the physical work environment. Inevitably computers bring with them the imposition of a new culture that indirectly affects design. Furthermore, the promises, threats and possibilities represented by computers unsettle attitudes to design.

These hypotheses were mostly confirmed, and in the process we discovered some fresh insights into the way computers are implicated in the changing nature of design practice.

The survey was designed to test our hypotheses, that originally appeared in an article published in CAAD Futures (Coyne, 1992). The results of the survey may assist in accounting for the difficulty organisations experience in making decisions about personnel, capital and training relating to computing, and assist research groups in directing their efforts.

Our survey builds upon the work of others. Although considerable attention has been given to the impact of computer use on clerical workers (Otway and Peltu 1983; Turkle 1984; Wainwright and Francis 1984; Long 1987) there has been little in the way of formal studies of the impact of computers on the attitudes and work practices of designers. Some of the factors motivating computer use in architectural practice are discussed by Radford (1988), Radford and Stevens (1987), Stevens (1989) and Radford and Coyne (1989). Helpful summaries of the implications of computing in general from a social science perspective are provided by Kling and Iacono (1988), Danziger (1985) and a compilation of articles by Forester (1989). Gutman (1988), Cuff (1991) and Olley (1992) indicate the way that the profession of architecture is changing in general due to technological and other pressures. There are also many quantitative surveys of the extent of computer use in practice. Our survey is qualitative, focussing on the influence of computing on the work practices and interests of designers. In our reporting in this article we present quotes taken directly from the interviewees in support of our arguments, rather than focus upon numerical analysis of responses to each question.

We put some 54 questions (with sub questions) to 29 practitioners from architectural and engineering practices and CAD consultancies during intensive face-to-face interviews. The majority of practitioners were from architectural firms, but we also wanted the scope of the survey to be representative of the changing forms assumed by practice under the influence of computers. The majority of the practitioners interviewed were involved in the day to day management of CAD and IT. Some were dedicated CAD or systems managers. Others combined CAD/IT management roles with design or drafting roles or with more extensive management roles within the firm. One of the interviewees was a designer who made extensive use of IT but was not involved in the management of IT, others were involved in the strategic decision making in relation to CAD and IT but not with the day to day management. All but five of the practitioners had been involved personally with computers for over five years.

We structured the survey under four major areas of concern: (i) changes in the workplace-the extent of change, how practitioners cope with making decisions about computers within a climate of change, how computers change work practices, (ii) the management implications of using computers, (iii) the character of work spaces occupied by computers, and (iv) the trends, ideologies and research agendas set by computers in the workplace. The pattern we followed in the interviews was to suggest a series of propositions about computer use, to which we invited a response, which in turn solicited further insights from the practitioners. We follow this pattern in the reporting in this article.

Changes in the Workplace

Computers clearly introduce substantial changes into the work place, though practitioners were ambivalent in their reporting of the change. The majority of practitioners attested to changes in the skills of the designer due to computing. Half of those who attested to change reported that the changes were minimal.

Several changes were reported. It seems that computing (particularly CAD) has brought to light new economic considerations that practices did not have to address previously. Practices have had to address the economy of using CAD-using CAD for what it does best and less expensively. A further change is that a new sense of team work develops around CAD. CAD also enables some firms to diversify in particular ways. Some firms were diversifying into computer-related areas such as desktop publishing and audiovisual work. One firm reported that they were able to charge larger fees for desktop publishing than for design. A further change is that with desktop publishing the production of text documents has assumed greater importance, with some firms employing a graphic designer. Other areas of diversification include: facilities management, audiovisual and graphic work, 3D computer modelling, shadow studies, documentation only services, "selling knowledge" about a particular building type through customisations to CAD systems, software development, consulting on CAD, and training. Most practitioners did not think that computing had substantially affected design-only presentation and documentation.

For the individuals interviewed the changes were more pronounced than for the firm as a whole. Some of the interviewees had moved from more traditional design and drafting roles into CAD management. The change was often dramatic. There are those who are no longer involved in design and those who think that they can only design with "the full range of technology." On the other hand, some reported that their design skills hadn't really changed. They were just working with different media. Some practitioners reported that the technology detracted from their design skills, others reported that their design skills were enhanced.

These reports of change came largely unsolicited in the interview. However, we were also interested in testing our own assessment of where change may lie.

The Relationship Between Professional and Support Staff

We asked about the change to the roles of the professional designer and manager in relation to what were formerly exclusively secretarial or administrative tasks such as typing, communications, filing and simple accounting. Some of the practitioners thought that the relationship of professional staff to administrative personnel had been restructured. Most of the practitioners said that they now carried out what were formerly administrative tasks. Most also said that they had observed this with others in the firm as well. In some cases this change had led to a reduction in administrative staff. In some cases the role of administrative staff had been extended. One of the practitioners from an engineering firm stated that the number of support staff had actually increased because the firm now produced more information of higher quality. The quality of documents was also important to one of the architectural practices, which asserted that "every document that left the office should be desktop published." The administrative assistants typed the material and then passed it on to graphic designers who did the layouts. A minority of practices said that designers should be designing and that other tasks should be left to support staff. According to one practitioner:

We don't employ secretaries or typists. Previously we would have clerks [for filing and typing], technical people and professionals. That doesn't exist any more. We have administrative people and technical professional people. I type all my own words, I maintain my own budgets using spreadsheets.

In general, the roles of "middle management" and support staff appear to be diminishing.

Viability of Small Firms

Another apparent change to the structure of practice is that computing has made small firms more viable. The majority of practitioners confirmed this view. According to one practitioner:

Yes, the era of the back yard operator has arrived and we need no longer feel guilty about it. I know of several circumstances where two competent architects with a CAD system are doing large projects extremely competently, and I would imagine extremely cheaply. So it empowers people with the skills to be closer to whatever it is that they want to be closer to rather than hanging around a huge office full of architects.

Some thought that small practitioners can now produce work of the same professional quality as bigger practices thanks to CAD and desktop publishing.

Making Decisions About Computers

IT poses a new regime of decision-making into an organisation-what equipment do we invest in? do we rent or lease? do we buy now or wait until later? We wanted to test whether it is difficult to make decisions in this climate of change and uncertainty. Our expectation was that decision making about IT would place some stress upon organisations. In order to test how firms were coping we first asked if there were people in the firms who are sufficiently up-to-date with product availability and IT trends. The majority of interviewees said "yes." We also asked whether or not there was a shortage of expertise in the area of system selection and management in the firm. The majority of interviewees reported that there was no shortage of computer expertise. The practitioners who thought there was a shortage reported that the main shortfall was in system management.

We suggested to the practitioners that those who know most about system selection and management may be fairly junior in the firm, because energetic new employers find that computing offers a means of career advancement, and they may have had some training at university. There was general disagreement with this. One practitioners observed:

CAD's been around long enough now but it was a problem five years ago.

We asked if the decision makers of the firm consult widely and well before making decisions in relation to systems acquisition and management. The practitioners generally reported that decision makers did consult widely and well. Admittedly, the practitioners interviewed were successful users of CAD. We did not interview those who may had tried it and failed, or who were unable to cope to the extent that they went out of business.

We targeted specific issues in our questioning to see how firms coped with decision making. We asked about the difficulty of planning for the obsolescence of the equipment, maintenance costs, and costs of training and management, which are easily neglected in decision making. Only three of the practitioners admitted that acquisitions were made without taking all of these factors into consideration. According to one of these practitioners:

We are still suffering the consequences of leasing a computer system over 5 years. That was wrong. Five years is too long to predict, it doesn't leave you flexible enough. You have to pay as you go.

We then asked about "human factors" that might undergo change due to the purchase of computer systems. Does the firm plan its acquisitions taking into account human factors such as: changes in work practices; staff obsolescence; work conditions? A majority of practitioners gave positive responses. Some engineers and the consultants were sceptical that this was the case in their firms. One said:

I don't think that they [the directors of the firm] actively made allowances for anything that may or may not have happened. I think they just decided to get into information systems and see how it went, and they had a management structure that coped with various levels of day to day management and they just took it form there. I don't think that there was any planned obsolescence. There was a turnover in the practice that allowed you to cope with that obsolescence fairly readily.

How do practitioners decide to invest in computerisation, particularly CAD? We took it for grant in our study that practitioners do not invest in CAD to improve productivity, and that the rationale people offer for their decision making is more subtle. We asked the practitioners whether the main reason the firm started in CAD was that more clients seem to require CAD. The practitioners were split on this issue. Those who disagreed with the proposition appealed rather to productivity and efficiency. But the question succeeded in bringing out some of the ways that CAD is introduced into a firm other than through carefully deliberated policy.

The reason that any organisation gets into CAD is that there are staff that want it and they exert pressure on management. It's hardly ever taken objectively. … It's the guy smuggling in a pirate copy of [name deleted] and putting it on the word processor that starts a firm in CAD.

A lot of them are taking the approach that CAD is coming so we might as well jump on the bandwagon now.

Searching for responses that identify problems in decision making we asked if the most difficult thing about selecting equipment was its immanent obsolescence. We were thinking of the problem of buying the latest computer equipment only to find that it has been superseded by something much better. There was some disagreement about this proposition, but the majority said "no." Some practitioners indicated that the most important component of a computer system is the software, which does not become obsolete quickly.

Good software is always upgraded. You buy your software and the hardware will look after itself. On the surface hardware is your biggest cost. In reality it's not. Your biggest cost is first the cost of staff gaining skills and the associated cost of training people when the staff leave.

Equipment selection appears to be less of a problem now due to the decreasing cost of hardware, and the high quality and flexibility of computers. Some practitioners posited a solution to obsolescence which is to purchase at the top of the range, rather than buy the cheapest. The other solution is to simply plan expecting the current machines to be obsolete in a couple of years.

We also asked about the role of computer vendors in the decision making process. We asked if practitioners were at the mercy of fast talking computer vendors in making hardware and software choices. Most architects and consultants disagreed with this proposition on the basis that they recognise it is not good practice to rush into such decisions without in-house knowledge. It appears that the skill is to have sufficient knowledge to ask vendors the right questions. Amongst those who see the pressure from vendors to be a problem, one remarked:

Yes, you certainly are to some extent. We didn't know which hard drive tape backup unit etc was going to be the best. The tape backup drive that we originally got, no one will service so we've had to change over.

We then focussed on the notion that design practices have not traditionally been capital intensive. Having to think about capital resources is a major change. The majority of interviewees agreed that having to make decisions about capital resources was a major change in design practice. However, this view is moderated by the scale of the capital involved, particularly the scale of costs involved relative to the staffing costs associated with IT.

It's nowhere near as costly as the staff member using it. Industry has accepted that for years.

Stress in Decision Making

We asked if decision making in relation to IT imposes any stress on the firm. The majority said that decisions about IT are now becoming everyday decisions. The stress comes from the initial move into computer-supported work. Most confirmed that the modular and low cost nature of computer purchases obviates the risk. Among those who saw decision making as stressful, one practitioner indicated that practice is undergoing a transition from a craft to a production industry, and we are not prepared for it.

It imposes stress because you are going from a craft based industry to a building production line process. You have to do things in a logical order, a lot of architects and people managing design don't think logically, and they get caught out.

In the light of these comments pertaining to management we asked if the particular firm in which the practitioner was involved had coped well with the changes brought about by IT. A majority of interviewees said that their firms had coped well with what was until recently an alien technology-though there were reservations. They reported some specific difficulties. Some difficulties pertained to false expectations about speed.

It's had its ups and downs. People think IT will be much faster, but putting the drawing on the computer initially takes as much time as on the drafting board. It saves time with changes and multilevels.

For one practitioner the introduction of the computer represented a change in culture to which architects do not easily adapt.

Average-mainly because architects do not think computers, they think art, which makes it difficult for them to maximise the benefits of using computers. Computers don't produce freehand wavy lines. They produce hard lines to exact dimensions. Architects by definition don't work that way. It's foreign to them.

According to one practitioner the effects of IT were negative.

It has had a negative productivity and financial impact. In terms of the staff, things have been fine, but in terms of getting the product out of it, we have not coped. We are not unique in this.

The worst scenario is where the computer is purchased but not used.

You see firms that buy a system and then have it sitting in the corner, whereas other firms buy the system and installation and training to get over that first hurdle and then they can pursue it themselves. In the early stages especially the training and support is essential.

We asked what common mistakes practitioner had made in the purchase of equipment. The majority did not admit to making serious mistakes. But the mistakes mentioned included being committed to elaborate and long term lease arrangements, underestimating the benefits of purchasing Macintosh PCs, purchasing equipment not supported in the country so that it had to be sent overseas to be repaired, wrong choice of back-up system, and giving insufficient priority to systems management. Some regretted being committed to the text interface DOS system in the light of the user-friendly Macintosh and Windows environments. Firms that got in early and had mini computers seemed to regret that they could not now shift readily to personal computer systems because they had already trained their staff on the old system. Many of these problems do not appear to be abiding. Some practitioners appear to have been caught out by the transition from large custom built systems to flexible, modular, powerful and user-friendly personal computers and workstations.

The next transition will involve laptop computers, computerised notebooks, advanced digital communications and multimedia, which appear generally to be compatible with the current style of personal computer and workstation, and there is a growing compatibility between operating systems. We expect that practitioners will report fewer regrets with hardware choices in the future.

The Cost of Computer Systems

We asked if the rapidly falling cost of computers had made the selection and management of computers less critical now than it was five years ago. Some reported that costs have not dropped.

Hardware costs have dropped but software hasn't. The cost of training has gone up. You don't go out and buy a PC any more. You buy a 32 bit workstation sitting on a very expensive network.

Others suggested that as the costs drop you buy more equipment, or more expensive equipment. It appears that software costs are still high. Also, due to the economic climate, firms are working with lower profit margins and there is less money available. These responses suggest is that it is the modularisation and flexibility of modern computer systems that makes the selection process less critical now rather than simply cost.

There are indirect problems with the reduced cost and hence reduced priority given to hardware. One practitioner reported that the reduced cost of hardware means that less money is spent on managing it.

Five years ago with the kind of investment you made you had to have someone experienced in the field to manage it. Now that the cost of the hardware and software has dropped so dramatically they're not prepared to spend a decent amount of money on getting someone in to manage it or set it up for them. They're buying the equipment without buying how to use it.

To explore further this balance between people and equipment we asked if the practitioners agreed that computers are now "cheaper than people," so firms do not need to maximise the use of equipment (by scheduling after hours use etc). Most practitioners agreed, and some were highly critical of attempts to ration out the technology.

They [computers] should be like telephones, they should be on everyone's desk.

If you can afford a staff member you can afford a machine.

Practitioners reported that in the past they organised work shifts to optimise the use of computers, but not any more.

At first we had 3 eight hour shifts. Now they [computers] are more like typewriters. Its not a problem if the computer is just being used a couple of hours a day.

But, according to one practitioner

the 90s are going to be the time of the capital intensive office rather than the labour intensive office.

One of the practitioners who disagreed that computers are "cheaper than people" remarked that computers are only cheaper if the capital cost of the equipment and all the associated tax benefits provide a greater productivity than the equivalent cost of hiring somebody, at least for drafting. The practitioners who reported as much appeared to be those with large minicomputers or those involved in leasing arrangements. Another practitioner remarked that as long as there is a heavy workload (the firm is doing well) then the staff costs are higher than the equipment. With less work there is less intensive use, and the equipment represents a larger relative cost.

Training

One of the major changes we hypothesised was the commitment to training that firms have to undertake. Training was conducted in-house, on the job and through formal courses. Some of the practitioners regarded formal training as a way of getting started with a particular computer package, but that was no substitute for in-house on the job training. Some reported that training occurred out of hours and was sustained by the motivation and enthusiasm of the individual. One practitioner said that their firm focussed upon user friendly systems that do not require extensive formal training. Only secretarial staff are sent on training courses. Again, those for whom training was a major issue were those firms with specialised turn-key systems. Some firms even had to send staff to the US for training. We expect a trend away form this requirement as systems become easier to use, and more students graduate from universities and colleges with the necessary skills. However, there was a consensus amongst the practitioners that training is an ongoing issue. The factors that contribute to this ongoing concern with training include the problem of keeping up to date with new releases of software. According to one practitioner training is becoming more important.

As the system expands your productivity drop is more important with each person who does not know what to do.

For some practitioners training in the use of the CAD system is less significant than being trained in the organisational side of the system-office practices pertaining to standardising title blocks, use of layering systems, font sizes, etc.

We asked if the firm takes training into account when purchasing hardware and software. Most practitioners responded "yes." Some of those that responded "no" thought that the software should be user friendly, so formal training should not be an issue. Some thought that training should only be necessary at the outset, as a one-off cost, and thus should not be the major determinant in the selection of a system.

We had hypothesised that one problem with training is the mobility it gives staff. You are generally training staff for their next employer. The majority of practitioners agreed with the proposition, but they were divided as to whether or not this was a problem. A number commented that their firms were building up a pool of expertise which they would be able to draw upon. Further comments were made that it was short sighted not to train people out of fear that you would lose them to the next employer. One senior practitioner thought that if you train people "they recognise that you have the right attitude." According to another practitioner trained people leaving is not a problem as there is a pool of expertise out there sufficient to meet the demand.

Today it tends to even out. As many people have now been trained, we pick up just as many as those who leave us.

We proposed that enthusiastic staff feel they can advance their careers by acquiring computer skills. Computing is therefore career enhancing for some. This enhances their mobility. Only two survey respondents disagreed with the proposition that computing is career enhancing. For those who agreed it is no longer so much a question of 'enhancing' a career as being a necessary requirement. Those who disagreed saw it as a "dead end" job for some people, thinking perhaps of those who become cast as CAD operators.

CAD Operators and Managers

We also put it to the practitioners that perhaps CAD has introduced a new dimension of tedium to design practice, and that the position of CAD operator is to be avoided by professional designers. A majority agreed with this proposition, but many qualified their agreement based on the observation that this situation is changing as more professionals are entering practice with CAD training, and are using CAD as just another decision-support tool.

I do not think the idea of using CAD is to have a draughtsman sitting in front of an electronic drawing board instead of drawing on a drawing board. There is a lot more scope in the system which architects and designers should use, but it is the old problem. Architects feel it is a stigma and they are afraid of it. I think with time we will overcome those barriers, but not in the near future.

A lot of professional designers are scared of becoming a CAD operator, but to be able to use the computer properly you need to become an operator first before you use it to design.

According to one practitioner it depends on the way that you run your practice as to how the CAD role is defined. For large projects the CAD operator may simply be implementing the decisions of others. If the same staff involved in the project and responsible for parts of it are those who also draw it up then the CAD role is not demeaning. Another practitioner went so far as to suggest that for those firms where the CAD operators is "somewhere down near the print boy" the office is inefficient. For some firms the role of CAD operator is shared.

We expect everybody to be a CAD operator. We are not distinguishing between architects and CAD operators, anybody who cannot use a CAD system is unusable in the way we run things.

People end up as 'operators' because the other designers will not take the time to learn to operate the machinery. You will get to a stage where unless you can operate these machines you won't get the jobs.

Even where some staff end up operating CAD systems the job need not be tedious.

You still get a situation where you get people who are specialists on CAD whether it be structural analysis or computer-aided drafting. I find that they don't feel that their job is a tedious one, they enjoy pushing the buttons.

The more glamorous side of the CAD business is CAD management. We asked the practitioners if this is a growing area of specialisation. CAD management seems to include managing the computer system, and also deciding how the CAD system will be used-layering conventions, customising the menus, organising files, etc. Most agreed that CAD management is a growing area of specialisation, but some suggested that in the future it will only be of relevance to the larger computer installations.

Systems management is clearly a problem area. Problems reported include the fact that "all the wrong sorts of people are getting into it," people who are good at it are hard to find, for some it is perceived as a dead end job, and CAD management is sometimes "out of sync" with the goals of the directors of the firm. Those who thought that CAD management is not a growing area of specialisation contended that with CAD the requisite skills will be absorbed more into the skills base of the average practitioner.

There are few good people about in CAD management. I think it might be declining as CAD skills are absorbed into the profession.

One practitioner contended that much of the CAD management role will be met by more sophisticated software. The role of the CAD manager will therefore diminish.

In the next 10-15 years the increase in the use of artificial intelligence will automate the management of systems in terms of making sure that you've got the right layer, with the right things on it, in the right place, in the right file, on the right job, at the right time. A lot of the tedious stuff will be dealt with automatically by increasingly sophisticated software.

This is already the case according to one practitioner.

You do not need very specialised skills to keep a system going.

Management and Organisation

We proposed that computing tends to heighten the need for personal and corporate organisation. We questioned the practitioners to test this hypotheses. First we tried to ascertain to what extent organisation was an issue for the practitioner concerned. We asked whether the practitioners regarded themselves as organised. Clearly organisational skills are an issue with those interviewed, though some expressed their desire to use work time more effectively. We then asked what impact computers have on personal organisation. Does the use of computers require one to be better organised than otherwise? A majority of practitioners said "yes," though one practitioner remarked that the profession of architecture requires a certain amount of organisation regardless of the technology. One practitioner drew attention to the ability of the computer to enhance organisation.

No, it doesn't require you to be organised but it keeps things organised for me. I use a Mac all the time now. When I feel like doing things, I flick through things I haven't thought about for a month and think more about it-your ideas don't just drift away.

For one practitioner computers allow you to be less organised.

The reverse is true we can keep things until the last minute.

We pursued the question of the computer as an aid to organisation by suggesting that computers are a valuable aid to organisation. The majority of practitioners agreed with the statement, but several interviewees noted that there are occasions when the opposite is the case.

If they're not used properly they can create more problems within an organisation.

We have used the computer as a tool in program management and it has helped there, but in other areas it is probably more of an encumbrance.

Adopting a different tack we asked if computers instil good organisational habits. There was general disagreement with this proposition, usually because computers "don't make a good organiser out of a bad organiser."

In themselves they can lead to chaos.

There was also the suggestion by practitioners that organisational skills and computing go hand in hand as "people who are well organised take to computers-others don't."

Perhaps computers have a negative effect in that they force rigid organisational habits onto people. A majority of practitioners disagreed with this proposition. It seems that the individual worker has a fair degree of control of how they use the computer. According to one practitioner "it's like the filing cabinet; it depends on how you use it."

Even if at a day to day level the relationship between personal organisation habits and computers is not critical then perhaps organisational "catastrophes" are more prevalent or more severe with computers. We suggested to the practitioners that mistakes due to poor personal organisation can be catastrophic when you use a computer (due to file loss, etc). The majority supported this proposition pointing out that even minor mistakes can cause major problems. But according to one practitioner, as long as "your security system is right, you have backups etc. the chance of a disaster is quite minimal." According to one practitioner:

It's no different to losing a letter, or spilling a cup of coffee over a drawing negative.

We then turned our attention from the organisation by the individual to the management of the firm. We asked if computers assisted in the management of the firm, and if so in what way. It appears that computers were used by all the practitioners interviewed in the management of the firm. The main areas of application are in job planning, accounting, financial projections and administrative tasks. One practitioner noted that their firm had undergone "a tremendous reduction in administrative personnel" as a consequence of computerisation.

Management of Projects and Documents

We asked if computers assist in the management of projects, and if so, in what way. It seems that computers are widely used to assist in the management of projects. The main use is in keeping track of the project administration and in improving the response time for variations or to requests for project-specific information. One practitioner articulated the benefits at length. The benefits include keeping track of variations on a complex project.

One of the great attributes of our system is that we can write macros very easily. All of the housekeeping functions that traditionally kept draftspeople busy in an architectural practice-amendments, variations and the tracking of variations. We wrote macros to control all of that so that a person couldn't log off their drawing unless the variation was logged on properly. The creation of amendment columns and the automatic numbering of amendments across all documents effected by the amendment we wrote macros for. So we would change something once on the 1:100s and wherever that detail was affected it was logged on automatically. That meant that we were able to run a complex commercial project ... which changed its leasing environment three times during the period of construction. There were 260 client induced variations all of which were tracked perfectly in our written database and on our drawings. The job captain, who was a director of the company, knew at all times that those amendments were up to date. The speed with which we could get those amendments back to site was very fast. That endeared us to the process of construction which meant that everything went smoother. We now believe that we could not have done that project without CAD. That's why the building bristles with commercial activity. All the life of the project was allowed to happen because it could be monitored, rather than frozen out because it was too hard.

All of this suggests that the management of a project that uses CAD is different to the management of a "conventional project." We put this to the practitioners, most of whom said that it was different. Some said there was no difference, but they generally considered the question in terms of overall management decisions rather than the process itself. It seems that CAD changes the way practitioners manage drawings, and as drawings are so central to the progress of a job the management of the project is affected.

There is a certain logic to the progression of drawings when done by hand which is different when done by a machine, and since the progress of the project is related to the progress of the drawings, you have a different sequence of drawings.

According to another practitioner it is necessary to structure the production of the documentation in "a more logical way" than on a manual project. There is a strict sequence imposed on the production process. You need the 1:100 drawings before the detailed overlays can be produced. With CAD there is a heavy interdependence between drawings.

If you're doing a project manually you do not have as heavy an inter-reliance as you do with a CAD project. For instance, you can start at any time on a new sheet, get something out of the drawing cabinet etc. In CAD you tend to take something that's already been done, enlarge it, put extra detail on it, etc. You are relying on something done previously so there's much more linkage, which means it has to be thought out and planned much more.

A further major change is that known and precise information can make its impact felt right at the outset of the project rather than later on as the project is refined.

Yes, completely different because fundamental information can be fed in so early. We start modelling our buildings immediately on CAD, that might be just by definition of an area, or a space, or a commercial premises of some kind. As it turns into an architectural process we keep on modelling it. So the reporting is early and accurate. I could show you schematic designs for projects that we've done in a week that are virtually working drawings. That's not hard to do.

Clearly CAD challenges the practice in which a design progresses from imprecise and ambiguous beginnings to a refined and precisely defined form.

There's a lot more thought required initially in a design. You have to make a decision as to what type of construction a wall element is made of a lot earlier.

This emphasis upon precision imposes new ways of working.

You have to organise the way the information is arranged on the CAD system differently to the way that you would on a normal hand done project. The information that you supply has to be clear and concise, in such a way that an operator can follow it easily.

The information has to be correct.

You can't just fudge information. If you get a survey from the surveyor and you put it in the computer and it's not quite correct, you have to get it right.

The payoff is that CAD introduces various efficiencies.

Rather than doing things manually over and over again we're doing it on a computer which gives us the facility to shortcut things. If you were doing a schedule of furniture items manually you would do a schedule of each item. In a database you can put that information in once and extract it in many different ways.

These comments are at variance with those of one practitioner who noted that firms using CAD tend to mimic what they did in the past but at a faster pace.

Clearly CAD is a resource that has to be managed. We wanted to find out if the fact that access to the computers has to be managed and organised places an extra burden on the organisation of the practice. Few practitioners considered access to the computer to be a problem especially where computers were on everybody's desk. As long as computer are ubiquitous in the office then there is no problem.

You don't manage access to drawing boards. You put one on everybody's desk.

We then asked if perhaps the purchase of hardware and software, training, maintaining databases, software and manuals place a management burden on the practice. Most interviewees agreed with the proposition. However, several of those who agreed indicated that this may be because management does not take an active role in such decisions.

We asked if design practitioners are adequately equipped to cope with the management issues imposed by computers. There was an even balance between those in agreement and those who disagreed. Responses ranged from a concern that designers are "scared of computers" and shouldn't be exposed to management issues, through to a recognition that

designers are designers and then there's another breed of people called design managers. They're the people who have to help produce the project. When you look at other industries-advertising agents, movie producers-they have creative directors and they also have producers and account directors. They manage the client and the process and to a certain extent the designers. If you're a good designer you have your mind on one thing. You don't necessarily want to go off and manage the process.

Then there is the extreme view that computers simply don't raise management issues.

I think they liberate designers from all of that.

Bottlenecks and Systems Failures

We then tried to detect indirect evidence of management problems. A common symptom of management problems appears to be poor provision for contingencies. We asked the practitioners: What is the weakest link in your computer system? What would happen if there was a systems failure in this area? There was a range of responses. The most common concern was with output facilities such as printing and plotting. Other technical problems mentioned included file transfer and software integration, remote access to a mainframe and networking, integration between the technical (office administration) and CAD systems, the power supply and software failure. Only one practitioner mentioned backups as a problem. Some responses highlighted the user interface problems of certain operating systems as a major bottleneck, and others reported the difficulty of raising adequate finance. Several practitioners reported problems between the people involved-the weak link being the potential for breakdown between designers and CAD operators, and between the systems people and general management.

For more specific responses we asked what contingencies did the practitioners allow for system breakdown, catastrophic file loss etc. The majority of practitioners reported that they rely on a back-up regime in the event of breakdown. The frequency of back-up mentioned (and therefore the severity of file loss in the event of failure) varied from monthly, to daily and even twice daily. Other forms of contingency included taking out an insurance policy against data loss, and running a parallel system or network of discrete systems between which each job might be swapped. One practitioner spoke of the backups as "windows" into the progress of the job.

I tend to believe that you rebuild your whole business so that you have windows of your business. If someone wanted to look at something you'd ask them when it was, say, May '89. You'd get the printout of the directory of everything that was in the system at May '89 (20 pages or so). They'd flick through that and find the file that they wanted. What it gives you on major jobs is windows of the progress of the jobs-important in legal disputes etc.

One practitioner also relied upon a network of other practitioners to assist when equipment goes down.

We have a network of other offices we know, so in case of a breakdown we can borrow or hire equipment. Our back-up system and our network friends are our recovery program.

Systems Managers

The role of systems manager had already been raised by most practitioners. We asked if the firms employed a systems manager, and if so what was their role. Half of the practitioners employed a systems manager. Amongst those who did not, some had a person on staff who was both a design architect or engineer and a systems person.

In smaller firms it tends to get added on to somebody's job.

One large firm employed several systems managers.

Yes, there are several people in that role. One looking after the hardware (backups, making tapes, physical hardware); one looking after software on the CAD side; one looking after software on the Mac side and keeping inventory of software releases, licence numbers etc.; and a senior CAD operator appointed on a project by project basis.

In another firm

The systems manager innovates the writing of macros. The CAD manager organises the resources and the training and the running of the system and is our most proficient operator.

For some the systems manager role is also that of CAD manager. One firm defined the role of the CAD manager as a production manager, analogous to the production manager in manufacturing.

A CAD manager is really a production manager. Because design firms have been craft oriented in the past they've managed to get by without production managers. They're responsible for the machinery and the effective flow of work through the business.

To test the importance of this role we asked what would be the immediate effect on the firm and its use of computers if the systems manager suddenly left. The responses ranged from "catastrophic" to "no problem at all." Those practitioners for whom it would be no problem said this was because the relevant expertise was distributed throughout the firm, or at least between more than one person. For some the loss of the systems manager would be only a slight problem as there were others in the firm who could take over, if they had some specialist re-training. In other cases firms would need to employ a new person. But this was not seen as a major problem as there are plenty of people with the necessary skills on the employment market. We asked specifically if expertise in systems management was distributed throughout the firm. In the majority of cases it was not.

Systems Manager Training

We asked if there is a need for specialised training for systems management. The majority of practitioners said that there is a need for specialised training. Some said that it should come on top of a professional education. We asked which form of training would be useful for systems management. The responses were (in order of most to least importance) for professional development courses, university postgraduate degrees, and part of an undergraduate education.

It's a strand of management. You have to have management training. Management as a whole and practice as a whole in universities is weak.

Clearly, there was also a preference for the systems manager to also be trained in the design discipline-architecture or engineering.

All architects and students should receive training in this area at all levels as distinct from having specialised training programs for professional system managers.

It has to be somebody with a background in design, a designer or an architect.

Some practitioners also hinted that the role of systems manager is one that is diminishing in importance. We asked if the role of the systems manager will increase or decrease in importance in the future. There was an even balance of responses. The general reason given for increased importance was the growing number of users and volume of information to be supported by the technology. On the other hand the importance was seen to decrease as systems become easier to use and more users become familiar with the technology. Practitioners in large firms or who use large systems saw the role as an increasing one.

The introduction of networking with its devolved computing power, as opposed to stand alone or star systems, has assumed that system management responsibility would be devolved also and I just don't believe that's true. Moving from a 17 station star system to a fully networked system has increased the management requirements ten fold. I don't see that changing but I'm hoping that, for example, you won't have to know Unix, or if you do that you will have learnt it at school as a ten year old.

Amongst those who see the role of systems manager declining:

I think the large firms will probably always need somebody-at the moment they're telling us it needs to be one for every six people-I think it will probably be less than that as people become more familiar with it.

Systems Managers and the Firm

We also inquired into how well the systems manager is integrated into the firm. We asked: how would you describe the relationship between systems management personnel and the rest of the firm? We asked the practitioners to respond to a number of possibilities: (no problems; amicable relationship; systems personnel appear protective of their expertise; or other staff are not very understanding of the pressure a systems manager is under.) None of the practitioners disagreed with the first option-that there are no problems. None disagreed with the second proposition-there is an amicable relationship. It appears that some systems personnel are protective of their expertise, but most practitioners did not see it as an issue. The majority agreed that other staff are not very understanding of the pressure a systems manager is under, though at least a third of the practitioners did not see this as an issue. (Perhaps they did not understand the pressure a systems manager is under.) One practitioner who saw problems said that

the systems manager must be trained as an architect and have a professional rapport with the rest of the staff.

For one practitioner the situation is different now to several years ago.

The systems manager used to be in an "ivory tower" but now tends to come from the ranks so it's a different relation to that in the past.

There appear to be different problems in large as opposed to small firms.

In small firms because they are part of the team there are generally no problems. In the larger firms it tends to be a separate person or a separate division. In those cases you can get isolation, people being protective.

For another practitioners problems with systems management in large firms can be attributed to poor management at the director level.

That [systems personnel appear protective] depends on the attitude of the directors of the company. If management throw up barriers and you have an organisation that doesn't give out information, you'll find the system manager is protective of his expertise. The systems people wanted to give out the information but the other people didn't want to know. They didn't perceive it as being part of their job.

One of the interviewees had the role of systems manager, and complained that people did not understand his role, but that did not impede his work.

I don't think other staff know what I do, but management know that I provide all the support that they need as far as them taking their decisions. I think the user staff (the technical staff) accept the technical support I either give or arrange for them to have.

Physical Work Environment

We hypothesised that spaces occupied by computers have a distinctive character which is sometimes at odds with the environments in which designers prefer to work. We wanted to find out how computers are located in the firm, whether or not computer workstations can be personalised, whether they support the culture of the designer, whether the narrow range of physical activities they support is a hindrance to design, whether or not computers are seen as cold and unfriendly, whether computers enhance communication or hinder it, and whether computers extend the work environment by allowing people to work from outside the office.

Locating the Equipment

We first asked the practitioners how they decide where to locate computer equipment. It appears that the decision is based either on the work practices of the firm, or on technical considerations. Those who answered from the point of view of work practices mentioned that their computer equipment is either located centrally, relative to each project, according to space availability, or on each desk. The technical considerations included lighting and glare, radiation, air conditioning and ready access to printers.

There is clearly a trend away from locating computer equipment centrally. One practitioner summed up the trend as follows.

We used to have all the terminals together. There are some pluses and minuses associated with that arrangement. For the last year we've had them in a distributed environment and there are some nice things to be gained from that.

According to another practitioner computers are located

where the people would be working as a matter of course. I have a personal preference for grouping work stations relatively close to each other so that if people have problems they can talk to each other, but by and large they're spread everywhere. With the latest technology there's no problem with lighting. You don't put them in front of a window, but with this sort of environment there's no problem at all.

Glare from natural light was a major issue for some.

It was a real problem because we designed this place to have excellent natural daylight for conventional drawing. That can be a real pain for a CAD screen because you get reflections everywhere.

Clearly flexibility in the arrangement of equipment is important. According to one practitioner the equipment

moves around according to the project needs. Sometimes we've had a workstation in a team of people working on drawing boards, at other times we've had all the workstations together working on a project.

We also asked if firms had drawing boards and computers at the same workstation. There was an even balance between those practitioners with both a computer and a drawing board and those without. Drawing boards seemed to fulfil two functions when located with computers: as layout tables, and for emergencies. But for some practitioners having both is an unnecessary expense. In one of the firms the architects had both drawing boards and computers, whereas the CAD operators only had computers. For some practitioners having both was not cost effective.

That is not cost effective. You either have a computer sitting there doing nothing or a drawing board doing nothing.

In order to discover to what extent practitioners using computers are able to personalise their work space we asked if computer workstations are generally shared or personal. The responses were evenly divided, but there is certainly a trend towards dedicated computers. Of course, secretaries, administrative staff and directors commonly had their own dedicated machines.

At the moment they're shared but we are trying to make it one per desk. However, anyone should be able to use anyone else's terminal (like a telephone) so in that sense they should continue to be shared.

The intention is that they're shared because we can't afford to give everybody a workstation. In practice what will happen is that if you use a terminal for four hours a day you'll get one on your desk. If you use it for less you'll go to a shared workstation. The workstations aren't just used for CAD. There are hundreds of programs running on them.

Computers for Design

We asked if computers are used for sketch design and design development or mainly drafting and modelling. The architects we interviewed revealed a broad range of uses including both sketch design and design development, drafting and modelling only, and design development, with one practitioner using the computer for sketching only. The engineers, on the other hand, were mainly interested in drafting and modelling. For one practitioner the use of the computer has reduced the amount of time spent with "butter paper" sketches.

We start on butter paper but quickly transfer in order to save time

Computers have clearly revealed and enhanced an important aspect of the design process-the fact that much sketch designing involves the modification of previous designs.

They are certainly used for sketch design and design development where editing is used. A lot of projects are modifications of previous projects. So if you're doing a multi storey building at the early stages and you want to do it quickly, you can often bring in a core or some form of a previous building, so you can build up a drawing very quickly from components.

It seems that the use of computers has also blurred the distinction between sketch design and drafting.

Drafting and modelling are part of sketch design development. Every job that we have has some computer development. We've done some jobs from sketch design and others from a later stage.

Where firms were using large, elaborate CAD systems that seemed to be an impediment to using the computer in sketch design, even though the system has sophisticated capabilities.

Regrettably, [the major application is] drafting and modelling. That's not the intention. The last people who convert to it are designers. They're lateral rather than logical thinkers, but our designers are now becoming skilled at CAD and they're inclined to sit down at a terminal and knock it out. The [name deleted] system is rather easy to use in 3D, but we've got a long way to go. You make the money in knocking out the documents so it's there that it is most cost effective.

For one practitioner the computer has introduced a new way of doing sketch design.

Concept designs tend to be on butter paper but they are transferred very quickly to CAD, some manipulation is done, then you spit it out again on the laser so that you're actually dealing in real dimensions, and then you work on butter paper again.

Computers Versus Drawing Boards

We then put it to the practitioners that perhaps computer workstations (for design and drafting) support a narrower range of physical activities than the traditional drawing board arrangement. People do not move about as much, they appear transfixed at the computer screen, they do not welcome intrusions into their work space, and they may even appear less sociable. We asked if the practitioners agreed with these propositions, and we asked for their observations of the way people work at computers as opposed to drawing boards. The responses were marginally in favour of our proposition. For one practitioner design is a solitary experience anyway.

The people who appear to be more sociable are the people who are detailing. Designers are actually more inclined to want to get away on their own and think.

I think they work in a similar sort of fashion. Some of them can be intense, some of them can be fairly easy going. It depends on the person. There is a certain amount of introspection with the terminal because you're sitting in one place and looking at it all the time, but I suppose the same thing can be said about a drafting board.

Considering computers as opposed to drafting boards: you do tend to get locked into place. Manual drafting is probably much better for exercise. You need to design the workstations around the individual rather than have the individual move from item to item.

According to some practitioners working at a computer is more intense that working at a drawing board.

Yes, computers are more intense because you are dealing with something that is very powerful. You can lock into it as an intellectual task. On the drawing board a lot of things can get quite dull, but I think people lock in on drawing boards when they're designing. There's also another pressure in that people on the computers are trying to prove how good the machine is and how much better they can do the job. So they tend to work harder. In years to come when everybody has a computer screen this will change.

Yes, computers requires more concentration. You don't like being interrupted as much. You're far more productive if you're not interrupted. Computers seem to gobble up time much more than the drawing board ever did.

CAD also requires you to think about other things than when at a drawing board.

I think you need more attention because you're not just drawing a line, you're thinking about how long this line has to be, and you're thinking about the X and Y coordinates.

One practitioner did not see any substantial difference between working at a drawing board and working at a computer in terms of interaction.

They [the computer operators] are very interactive with the people around them. They do get up and walk around. A computer is really the same thing as a drawing board.

One practitioner interpreted the relative inaction of computer operators as being more relaxed.

I think with computers people are more relaxed. If they need to get close to a particular aspect of the drawing they can just zoom in instead of having to physically lean over or squint.

For one practitioner there are other factors that contribute to the degree of interaction between people in an office.

The difference in the way people work is more to do with the physical arrangement and layout (lighting, etc.) than anything else.

Some practitioners proposed solutions to the problem of the apparent self absorption of the computer operator.

They [computer operators] should be interacting with the decision making of the project so that shouldn't be happening. There should be dialogue between the decision makers, be they technical decisions or design decisions, happening through their day that keep everyone in the office communicating.

One solution to the apparent stress of working at a computer is to take frequent breaks.

You do feel physically cramped when you're working at the screen, I think you need to take a break very regularly. I guess we do so many different tasks that it's generally not a problem with us, although on occasions when I have to get out a drawing you know it by the end of the day.

One practitioner had experimented with a better ergonomic design for a computer workstation.

One thing that I implemented with our GDS systems was to have on one side or the other of the workstation an adjustable sloping reference desk, you could put the drawing that you were referring to on the sloping surface so that you only had to turn your head and it was there, but they were quite expensive and we couldn't afford to fit out the office like that.

We also suggested to the practitioners that environments populated by computers may be cold and hard compared with the more casual human environments centred around drawing boards. We asked if they agreed and, if so, if anything is being done about this in the firm. Most thought that computer environments are not inherently cold and hard. The image of computer environments as being cold and hard is more to do with office layout. Some of those who agreed with our proposition said that it is due to insufficient attention being paid to the design of the workspace.

It's because nobody pays any attention to creating spaces that are user friendly for the people and the equipment.

One firm even kept drawing boards to keep the work environment friendly.

That's the reason that we haven't gotten rid of the boards. You come to work for fun as well as work.

According to one practitioner computers are caught up in a game of image making by the firm, and it need not be so.

It's an image thing. A lot of firms want to push that high tech stainless steel and brick image, and computers can be part of that. Some of the big firms invite the clients around to see that all the technology is there and that tells them something.

One firm attempted to overcome the problem of the sterility of computers by better spatial integration, though there are problems.

We attempt to integrate the work stations into the general office space, rather than have a separate section. The big problem there is the very different lighting levels required by each. The problem is almost insoluble because you don't want bright lights near a computer screen.

Computers and Communications

We suggested to the practitioners that CAD and IT provide unprecedented opportunities for collaborative design activity: collaborative design at a distance; communication within the office through electronic message and memo systems; shared work areas on the computer screen; electronic communications with consultants; all electronic fax processing; electronic mail; electronic conferencing; and multimedia. We asked the practitioners if they agreed with these propositions. The majority agreed. But most of those who agreed also added that the potential is largely unrealised in design practice at present. The major beneficiaries appear to be large organisations who have consultants in-house.

There is no doubt that faxes make it much easier to get hard copy from one place to another but the use of electronic mail in most architectural offices is non-existent. To make it work properly you would have to have a large organisation with a high percentage of workstations to employees or it's self defeating. The exchange of information between various disciplines (eg. the architect and the engineer) is almost a non-event because nobody has the same system, even if they have the same system they don't have the same hardware or the same disc or tape drive.

Some practitioners appeared unaware of what was possible, thinking that comprehensive electronic communications were expensive and still some way off in the future. We expect that these comments would be revised in the light of the publicity given more recently to the feasibility of teleworking, the use of Internet, and the "information superhighway."

We asked specifically if practitioners thought that electronic communications were adequately exploited in the firm at present. The majority thought not. Only one firm reported using software that enables designers in different locations to work on the same drawing in real time-synchronous computer-supported collaborative work.

We're using Timbuktu [collaborative software for the Macintosh] to work with an office in Newcastle.

Only a few of the practices reported using electronic mail.

Some of us use electronic mail but it's limited to those people who have a screen on their desk.

For one practitioner security was a problem with electronic communication, though they had overcome this.

We don't use outside consultants so we don't have to worry so much about security, although obviously you have directory protection. There's no way a mechanical engineer or draftsman can alter an architect's drawing.

One practitioner took communications to include the whole matter of presentation. In this they included multimedia presentation as part of the communication process.

You have to communicate in the realm of concepts and ideas competently, and the architects hang around waiting for someone to pay them to do a colour perspective or a ray traced model of their design before the client ever really gets to see anything of what it's about. Architectural drawings are the limit of their domain. The truth is that there's so much media around that you can muster to give full expression to your ideas and these decision makers are subjected to communication like that all the time now, everybody gets presentations on videotape. If you're not competent in those communications fields you're going to be stuck in a sort of paper technological field.

But it appears that there is a trend towards greater use of electronic communications and the integration of this with CAD.

We aren't set up so that we can share all our files across the network. We're still doing a bit of disc swapping. That's a matter of time.

We asked if computers were helping or hindering the communication process. Computers are generally seen to be helping with external communications, data exchange, etc. (with reservations), but are not necessarily helping with the interpersonal communications believed to be important in design. According to one practitioner computers are revolutionising communications.

Helping! They've revolutionised it!

For some computing tends to be alienating.

It is hindering [communication] in the sense that the nature of the computer screen and the person transfixed there is very isolating and little communication is chosen to be undertaken by the person or by others-an alienating thing in some senses.

It has increased efficiency, yet centralisation and the specialised nature of using computers cuts down on communication.

Email is helping with communications externally but tends to reduce the level of informal face to face communication within the office.

On the other hand one practitioner did not think the computer is making any difference to communications, in the sense that we do not necessarily exchange more information.

I don't think they're making the slightest difference. You can exchange more information when you want to exchange it, but whether it's affecting the frequency of the exchange of information I don't know.

Teleworking

We asked if members of the firm are involved in teleworking-whether they have data links between home and work, and whether they are able to do some work at home because of this. Only a very few of the practitioners made use of such a facility-6 out of the 29 interviewed. Some reported that it is possible but not exploited as it should be. According to two practitioners:

Almost everyone has a modem.

Yes, our CAD manager has a modem connection to a PC at home. Yes, also I just transport my Macintosh.

Clearly there is a trend towards the practice of working from home, though there are impediments.

In a government organisation it takes a lot of change in policy. I can see a lot of opportunities for people to work at home.

Some work at home but are not linked to work via modem. Some use fax.

People do work on their PCs at home and bring it in on disc. Very few people access our network via modems.

Not modem link. A couple of people have faxes. Architecture is a face to face industry.

We asked the practitioners what they thought of the common projection that it will be necessary to come to work only occasionally. We also asked if it would apply to all levels of staff in the firm. It appears that this is already happening for a very small minority of architects, and a majority of others see it as a positive possibility. But some of the practitioners were highly sceptical of the feasibility of such a proposal.

That happens for some people now. I know at one architectural office they have almost as many people working at home as they do in the office. They're in and out. Some people want to be at work all the time with other people. It gives choice. It should apply to all levels of staff. I don't see why not.

It appears though that there are some technical problems to be overcome at the moment.

It's workable if one can get the whole of the information system working properly. It's not easy or fast enough to send a drawing through a modem at the moment. You'd have to come into work maybe three hours a day.

Some strongly disagreed with the feasibility of teleworking.

A lot of nonsense. Business doesn't work like that. Sometimes people go home just so they're not interrupted, but to say that you can run a major job by yourself from home is a lot of nonsense.

The main reservation was that you need the face to face contact with people-for the benefit of the client.

You still need that personal contact, and your clients and consultants want to see you in the office. If a person was only doing word processing or data processing they could probably work from home.

Some practitioners thought that face to face contact is also essential for supervision, management and the learning process.

It is possible, yet I question the perceptual and social aspects of that. It is a question of management confidence. It would not apply to junior staff because it is a management issue of monitoring function, etc. Also the education side of working in a firm is important. Control is the issue. It depends on how the practice runs. It would need specific guidelines.

It is almost impossible to supervise anybody at a distance, it's even more difficult if you're the owner of a company to be able to trust every single one of the people that you've got working for you.

But perhaps there are changes.

It's essential that you come to work occasionally. Face to face communication is so important. Maybe it's not so important for young people.

Teleworking is more feasible for individuals whose work tasks are well defined and relatively independent of other staff.

You could easily have say, a renderer who has his equipment at home. You could just send him the model and he would never have to come near the office. I can see it happening for some people (some consultants) but I don't see it happening as quickly as the media would have us believe.

This is certainly the case with design firms that specialise in software development.

A lot of people in our industry (software development) do tend to do that sort of thing. Coming to work's a habit. I know one person who lives in Adelaide but works for a Sydney company. He communicates by modem and comes to Sydney occasionally.

Bearing in mind these developments, we put it to the practitioners that the introduction of networked personal computers may be one of the most significant developments of IT in the workplace. There was general agreement with this proposition. We also suggested that this was in part because PCs enhance the autonomy and sense of self worth of the individual user. Only a minority agreed with this second proposition. Those who supported the second proposition provided the following evidence.

One feels more relevant with an expensive piece of equipment than with a drawing board, but not in the future when computers become commonplace.

The sense of self worth is because information equals power, until such time that enough people have that information or access to it. Then the power ceases.

If you are able to achieve more in a day it does enhance your sense of self worth.

Trends and Research Issues

There are clearly many developments in computer technology in train at the moment. We put several trends pertaining to CAD to the practitioners and asked how they rated their importance: increasing photorealism, better animation, virtual reality, improved information exchange standards, expert systems, tools for remote collaborative design, hypermedia and multimedia, and presentation techniques that simulate manual drawing. By far the most important of the trends we put to practitioners was the development of improved information exchange standards. According to one practitioner:

There's currently no way of transferring 3D. We've just had a job where it would have been very nice to get the engineer's 3D.

One practitioner was sceptical about our ability to realise improvements in this area, stating that

the moment that you try to create standards is the moment that people try to change them, but there would be a lot to gain.

The second category of trends that the practitioners favoured as important included increased photorealism for presentations. According to one practitioner so much of design practice comes down to presentation.

Presentation is half the sale.

Most practitioners were equally supportive of the trend towards improved animation techniques.

Expert systems were also seen as important. These are computer systems that simulate some aspect of professional expertise, and do so in a way that is amenable to updating and change-to the "knowledge base." These systems exploit computer programming techniques developed in artificial intelligence, and frequently make use of codified rules rather than programmed procedures. According to one practitioner expert systems are applicable because design practice is information intensive.

Information exchange means by default that you have access to a lot of information. Manipulating it is a province of expert systems.

Expert systems haven't taken off anywhere near where people thought they would, but that's partly because you need an environment rich in data.

However, there was some scepticism about what could be achieved with expert systems.

An architect makes a snap decision that would require an expert system to sort through hundreds of variables, and the data base has to be up to date. An architect might not make the correct decision but he makes it instantaneously taking all sorts of abstract concepts into account.

For some practitioners there is a lot of unrealised potential in the idea of expert systems.

I think a lot of the stuff which has been done is nonsense, but we'll get there eventually.

Potentially important, but I haven't seen anyone particularly take advantage of it.

The trend towards tools for remote collaborative design was also regarded as important, though there was little elaboration by the practitioners-perhaps as few practitioners knew of the developments in this area. Hypermedia and multimedia also rated as important, again, without elaboration. It was apparent that at the time of the survey little was known about these applications, though they sounded interesting to the practitioners when they were explained to them.

Virtual reality rated slightly less in importance. A number commented that it was a long way down the track and that there was a lot of media hype about the phenomenon.

Datagloves-I think that sort of input device is the way that we should be going for design work. One of my basic problems is with the sort of input devices that we have, and I think that something like a 3D mouse, something that allows the architect to gesture, as if modelling with his hand, is incredibly exciting.

We also mentioned the trend towards miniaturisation (laptops, palmtops etc.) Slightly fewer practitioners saw this as an important trend, though there was still a majority of practitioners who saw it as important.

Taking the machine with you is important

The smaller the better. I don't go anywhere without the laptop now

I can't see them getting much smaller than a keyboard allows them to get. Then you've got a screen. I want to see the thing. Reduction in weight is handy. It would be good if the screen could fold up.

We also suggested the trend towards presentation techniques that simulate manual drawing. The majority of practitioners considered this an area of little interest. According to one practitioner "we have that already."

It's interesting what it does for client's perceptions of drawings. I laughed at it to begin with. I thought it was a real con. The same with hand lettering. But there's a point to it. Clients are much more inclined to have some input into drawings that look rough, and interestingly enough the lettering style that looks like hand lettering stands out from a machine drawing.

Experimentation and In-House Development

We also put it to the practitioners that there are now computer tools that enable experimentation with new ways of working and designing. We asked if this experimentation actually takes place in practice, and if so, in what way. Opinion was evenly divided. In several cases the experimentation was seen to be taking place, but as being ineffective, or as taking place in only a small number of practices.

I think the whole advent of CAD and PCs is itself such a tool, and that has changed the way that we work. About five percent of people in practice are [experimenting], they are constantly assessing new products.

Some of the examples provided by practitioners included the use of spreadsheets.

You can turn Excel [a spreadsheet program] into a design tool. Instantly you have a tool for exploring options with plot ratios.

Another practitioner was involved in the development of a custom built database system.

I am trying to establish a database system which will read off our drawing information which is relevant to my superiors in terms of occupancy, floor ratios, costs, etc. I would have to write an intermediate package due to the incompatibility of the drawing program and the database.

According to one practitioner there is no time for experimentation.

People don't have time to experiment in practice.

For others, the tools were considered too cumbersome.

There are very few tools that allow easy experimentation. For simple things like area calculations you can.

For one practitioner the computer encouraged experimentation with new methods of working.

Tackling a design brief by computer is different to traditional methods. Traditionally you had to follow a step by step process. Now you can do the steps out of sequence. You can jump straight five steps ahead and do a preliminary pass. It doesn't waste your time and it has a bearing on step 2.

We asked if practitioners saw any in-house development of customised computer tools. The majority of practitioners said that they did see in-house development. Much of this seems to be in the area of programming spreadsheets, setting up menus, libraries and macros. Reasons given for customisation included the specialised requirements of each office and the fact that vendors don't understand the needs of the designer. The comment was made that customisation may be intensive when a system is first introduced but that you reach a plateau where customisation becomes minimal. Some of the practitioners drew attention to some of the pitfalls of customisation: the expense, the expertise required, and the fact that idiosyncratic programs may be developed that people don't know how to use. For some, customisation is an important activity in their practice.

We're doing it all the time. We probably have about a thousand macros: from things that automatically generate drawing files and put them together to form the final plot drawing, to programs that will actually trace a line on the screen and put down spot heights at predetermined intervals in order to produce good quality 3D images (from contour plans).

There are traps to be avoided in customisation. The main problem is that in general the only person who knows how to use the customised tool is the person who developed the customisation. That can pose serious problems when that person leaves the firm.

You can't expect that your drafting system will do everything you want it to do so there has to be development. It's got to be carefully used because everyone can be going off and developing these programs and no one else can use them.

Customisation is also a particular skill.

We don't do a lot though because most of the people here are professionals, and they just want to use it as a tool, you've got to be a bit of an enthusiast to start writing macros.

It is not something practitioners particularly want to do. Customisation is undertaken to overcome the shortfall in what is provided by vendors.

Some, but mainly because the software industry is ignorant of the needs of architectural practice.

Some think that customisation is becoming less necessary as vendors become more in touch with the market.

Generally the third party vendors are providing better interface services.

Some practitioners disagree with spending a lot of effort on customisation.

I see serious in-house customisation as a waste of time. That is where you lose the expertise when people move out. I'm a firm believer that you buy proprietary tools. There are thousands of hours of R and D in those packages that you can buy for $5,000. I'm a firm believer in waiting for the proprietary product.

It generally ties you to that one individual or the one firm. You don't have the flexibility of working with general software packages and finding the best one to use and the best way to use it.

Understanding Design

How are these trends affecting our conception of design? We asked if the practitioners thought that the use of computers leads to an emphasis on form and appearance (due to computer graphics, modelling and rendering), and was it a good thing or not. The majority of practitioners said that they did not think that the computer led to an emphasis on form and appearance. Those that thought that it did saw this as a positive thing-making public ideas about the way that a proposal would look, ideas that would formerly have been the private domain of the designer. The majority of engineers believed that computers did lead to an emphasis on form and appearance, adding that they thought this to be a good thing. One practitioner remarked that nothing had really changed in this regard due to computers.

This hasn't really changed, you've always had perspective artists.

For some the emphasis upon form is a good thing and has meant that the client can see what the building looks like.

It helps the client to see what the building would look like at an earlier stage, but the client doesn't get dazzled by it.

computers enhance our understanding of design? We asked if practitioners thought that the computer helps our understanding of the mind of the designer. The majority of the practitioners thought not. Some practitioners reported that there was a conflict between designing and computer use. For some, the computer appears to have revealed the "illogicality" of the designer.

It's certainly formalised a lot of things. It's made you realise that designers aren't logical, that they can make two and two add up to anything that you like. So you have a conflict.

We also asked if the practitioners thought that it is now (or will be) possible to put design expertise into a computer. The majority of the practitioners said "yes." But most qualified their response by limiting such "expertise" to "technical trivia," such as door numbering and code checking.

Good offices that are well organised have an office procedures manual. Why can't you embed that, and you get the office procedure?

We could probably come halfway, but I do not think we can replace a designer with a computer. They will eventually be very helpful in the decision-making process. Computers are good at number crunching and storing information, and the extent of its use will be as a code checker.

Not in the sense of the qualitative human experience and feeling side of the design process, and the art side of architecture. The quantitative knowledge side of things, yes.

We asked what is the role of the computer in design, now and projected. Views varied from more extensive use of the computer throughout the design process (simulation, briefing, data bases) to providing an entirely new medium for design. Some practitioners were careful to point out that the computer would never automate the conceptual phases of design. Some thought computers would have a coordinating function, others that computers would allow designers to do certain things more easily (such as enhanced visualisation-multiple views and the projection of certain types of shapes). The responses acknowledged that creative uses are being made of the technology now. Most recognised that appropriate interfaces to make computers more amenable to design are on the way.

It's expanding substantially. Now we have the body of expertise to do it. Hardware, databases, simulation, optimisation was done in separate little programs in universities. Now that can be brought into the office environment.

Computers are great for solving problems a little bit at a time.

CAD allows you to better visualise and test your designs in three dimensions.

Some shapes are really difficult to draw on the drawing board so you don't even think of doing it , whereas on the computer it's not that difficult.

It gives an opportunity to take a look at design ideas and to see that in juxtaposition those design ideas work. Often somebody comes up with a concept and the size that he or she initially thought would work won't work. Complex geometric shapes don't necessarily work when they all come together. It's a visualisation tool of the reality that has been in the mind.

A means of documenting the design and exploring options. When I say documenting the design I mean full documentation, not just drawings. Briefing information, data used over the life of the building. The information used in design will have a longer life over the life of the artefact.

Predictions

We asked what the practitioners recalled were the major predictions about computing in design practice made ten years ago. It is clear that different practitioners were listening to different predictions: predictions of replacement (paperless office, no drawing boards, reduced staff numbers, word processing), design impact (3D modelling, system building, expert systems), and work practices (24 hour shifts, automated drafting, automated design). Most common were predictions relating to changes in work practices. A number of practitioners added that few of the predictions have been achieved in all areas, though most are there in some form or another. What were the major predictions ten years ago?

The greatest myth was the paperless office.

That a computer would do everything. That you would push the button at one end and it will produce a building at the other. A computer can't do everything. It's limited by the person who is operating it.

The people projecting the future always make it sound a lot closer than it is. Individuals have reached the predictions but not across the board. It's an ideal. Most people have to compromise.

There were predictions that offices were to run 24 hour shifts in order to make use of the capital equipment that they needed to have. There were predictions that there would be a lot of expert systems used to check designs and compliance with codes and so forth. I have not seen either of those happen.

Computers would make all buildings look similar. It would standardise, simplify and get things done quicker. A systems building process.

The documentation side is the only thing which has come true. The tools around for documentation now are excellent. As far as the design side goes they're still making the same promises and predictions as they did ten years ago.

We also asked what were the major predictions regarding computers in design. Again there was a great diversity of predictions; mostly that something new would become available (there would be expert systems, virtual reality, automated design systems), but also that the potential impact had been underestimated (that computers would have little impact on design). In all cases across the range of predictions, the practitioners thought that the major predictions had failed to eventuate. On the other hand, some practitioners thought that the predictions made ten years ago were not ambitious enough.

I think every prediction will one day come about. You tend to underestimate the predictions.

People didn't believe that it would make much of an impact at all. They wouldn't have believed that computers would have gone this far.

Ten years ago, nobody thought of computers as a design tool but just as a mechanical aid to production. I do not think anybody saw it as anything that a designer would ever touch. That anticipation of ten years ago has been proved incorrect.

There are clearly some predictions that have not come about.

Computers for showing the way that things would look, hinting at virtual reality, and we're not there yet.

That the designer would be designing and documenting straight onto it. That doesn't happen.

According to one practitioner there is one prediction that has fortunately not come to pass.

That they would stultify and inhibit design, but I do not think they have.

We asked specifically which of these predictions appeared to have been born out in practice. One practitioner said very few.

No very few, 1990 was the water shed between drawings and info data bases. Expert systems have yet to be taken up. There have been predictions about simulation and optimisation which haven't come about because the building model (which has to be the glue) doesn't exist.

It's been a lot harder [than predicted] for the individual sitting at a terminal to do. It requires a very high level of concentration. Your mind has to work at three times the speed to keep up with what the computer is telling you.

Electronic mail could eventually come about so that you don't need paper but it's a long way off yet.

Partly, some have happened faster and others haven't. The price and power of the computers has come down. The general practitioner is usually 10 years behind the cutting edge. The field is segmented, everyone is on different blocks [different specialisations within CAD].

Some, most of the short-term predictions are fairly accurate-a bit dubious about the long term. There is a lot of science fiction in there.

We asked if the practitioners had any major predictions for the future. There were two kinds of response-predicting more of the same and predicting a Gestalt shift. The common technological changes predicted were of increased speed and computational power, with further improvements in interface design. There were a range of changes to work practices predicted: more remote collaboration, computers everywhere, and improved integration of design, construction and material supply. The most common prediction was that virtual reality (in the form of improved animation and graphics techniques) would be the next major development. There was some mention of intelligent expert systems and electronic data models. Some practitioners focussed mainly upon the technical changes.

The speed of the boxes is going to make everything possible. Whether people are going to be prepared to invest the money in the software to make it all happen is another matter.

The trend will continue as it has gone in the last ten years and the systems will get simpler, and we will eventually eliminate the drawing board.

Speed and realism, virtual reality, will far exceed our present expectations. Whether that's a good thing I don't know.

Its being predicted for us. The front end people are doing it and the rest of us are just following along. You've got the manufacturers rationing the technology. Canon's A3 colour copier has been around for 5 or 6 years. They're not going to release it until they squeeze as much as possible from the black and white technology or until some one comes up with a competing product. I don't believe you should be leading edge. I think you should sit back and let someone else waste their money on being leading edge.

According to some practitioners there will also be structural changes to the industry.

Small firms will become just as powerful and able to compete with the large firms, and the role of the CAD operator will diminish.

Yes, I see a move that projects can be done in a more distributed way where the people working on the project are distributed geographically and the actual offices involved are much smaller. The team make-up will change from project to project rather than being the same in-house team for every project.

Problems Requiring Research

What are the problems that should be receiving research attention at the moment? We asked what (if anything) were the problems with computers in practice today. Many of the practitioners mentioned cost. Others mentioned standardisation, integration, speed and the ongoing need to upgrade.

Speed is always a problem. We're always quicker in thinking than the computers. Once you start in computers it's a never ending process of upgrading and updating.

People are saying "I'd love a computer that would do what I'm thinking." It's not a problem. It's just a wish. Standardisation is a problem that's difficult to solve.

Expertise is a bit of a problem. Budget is always a problem-being able to afford what you want.

The problem we have is that all the software packages function in isolation from each other. Our biggest limitations are software limitations. Ideally, we would build a package by taking modules from other packages. We cannot find software packages which fit exactly our needs because they are rigidly constructed and distinct from each other, which makes it difficult to mix and match to build something which fits our needs.

The complexity of the systems. More people would use them if they were easier to learn.

We asked what were the problems and issues ten years ago. Many of the issues raised in the previous question were repeated, such as cost and speed. Some also added the issues of getting suitably trained personnel, the difficulty in getting the machine to do anything at all, and people not understanding what computers were capable of. Other problems included the basic question of whether or not to get involved with computers, and if so, what system to use, adjusting to change as the technologies developed, and adapting oneself to the new technology.

We then presented several research topic areas-problems-to the practitioners and asked for their opinion as to which ones were the most worthy of research effort. Of the 15 topic areas we suggested one met with resounding support. This was management. CAD and CAD services need to be managed better.

Yes, I think most CAD systems aren't managed. The classic case is the [name deleted] system. It's put on someone's desk, they spend two weeks playing around with it, and the boss will say "OK they're trained now," when actually they're producing half the work that they would have manually.

They just use it as a drafting tool. They don't see its potential in redirecting the way that we approach documentation. We'll tend to produce buildings more like the way that we produce motor cars. We'll have standard designs and variations on those standards, and architects will produce the systems.

A close second was a category of problems, the first of which pertained to integration. The problem is better integration between systems and the incorporation of data and communications standards.

This is where it's all going-getting my system to talk to the consulting engineer's systems to talk to the client's real estate agent's system, his property manager... It's all about communication, communication protocols and data protocols. This is where the excitement is.

The way different packages communicate with each other is appalling. Data exchange standards should be a lot clearer and universal.

Of equal importance is the effective exploitation of information for automation. The problems with design relate to the transfer of information down the line to the production/construction process. Computers make this a lot easier and more effective. We need accurate databases, modelling techniques, and automated design checking as well as better information standards.

Yes, the construction industry wouldn't have a clue about what this stuff is all about. They're so reactionary. The impact on the built environment is going to be enormous. It's going to come through integrated databases. The design professional will leave the contractor behind in this regard. It's a bit like quantity surveyors. They haven't been too interested in automated quantities because a) they want to check them and b) they're not going to take any notice of input an architect has done unless they can see it and measure it. The amount of time that an architect would have to spend inputting the data over and above what he needs for his graphic production costs more than the quantity surveyor's fee. CAD systems can count but how do you measure the footings of the building and apply a different rate to one that is 600mm deep to one that is 1200mm deep? The architect has to explain it all. It's a myth this full 3D model in which everyone whips in and takes data out of it, but a disintegrated model is quite feasible and I think that's the way we're going.

Of slightly lower priority, it would appear that we need to research better design oriented computer systems. Computer support for designers is poor. We need better systems created with the needs of the designer or professional in mind.

To date it's been documentation oriented. Designers for various reasons haven't used it, and they haven't had the tools.

The problem is that computers require accuracy. You can give a computer inaccurate information but it will interpret that inaccurate information accurately, and designers can't think that way. My observation of designers is that it's very hard to make a computer assist them in a vague way. You can have a great idea and say to the computer "Does it fit into the space I've got?" but it's not going to give you the good idea to start with.

Its not so much the computer. Its the human side which needs to make the adjustments.

The problem is that not even the designers know what they want. So how can people give you what you do not know. The designers need to be educated in the computer field so that they become more aware of their needs.

There was equal agreement on the need to use computers to address the increasing complexity of design. Design is complex and difficult. It is becoming even more so lately. We need computers to assist in this complex task. It is interesting the phenomena that the practitioners identified as being the source of the complexity: increased expectations, new modes of working, and the increased complexity of ordinances and regulations. Firms tend to conceive of these issues as being extraneous to design.

I do not think that design has become more complex than it ever was, but I do see that computers allow us, by taking care of some of the complex tasks, to concentrate on other ones.

Very important-not because design is becoming more complex but because more is expected of us.

We also suggested that perhaps designers have to cope with more information now. It is impossible to keep up unless we have information processing aids in the form of computerised catalogues, databases and electronic communications. The practitioners also saw this as equally important.

The situation isn't much different than in the past. Most manufacturers use computers and have realised that they can now document on computer, and they've realised that people can take that data digitally. Managing that data is an important task. CSIRO [Australian Commonwealth Scientific and Industrial Research Organisation] has a role in developing an environment that makes the development of these products compatible across industry.

I do not think that we are barraged with more information than in the past, but the opportunity is there to access information and it is just not being presented. I think it is insane that you cannot dial up a database and find out all the different types of glass available on the market.

Of slightly less importance was an attitudinal "problem" amongst practitioners. We suggested that perhaps the potential of computing in design practice is not fully realised due to conservatism, ignorance and unwarranted scepticism. One practitioner thought that it was an important issue but one that time, rather than research, would solve.

Of slightly less importance was the problem that there is a major split in the design professionals between those who embrace the technology and those who shun it. This division is serious and needs to be broken down.

Very Important-in the future you will not be able to design without a computer in the same way that you cannot design without a drawing board.

I think that division exists, but the people that shun it, that's their problem, and I am not interested in helping them solve it.

We also suggested that perhaps little is understood about the role of technology in society and our relationship with it, and this needs research. There was less enthusiasm for this proposition, but still the majority of practitioners supported it.

A lot of architects don't like the technology of their practice (even drawing boards). I think we need an understanding of the role of technology, and in the past architects have been pretty bad at catching up.

This is changing, society is much more aware. They all have TVs and home computers.

Very important-a lot of the time we are looking at computers through very old ways of looking at them, and if we can re-assess our relationship to them we are going to be able to make more satisfying use of them.

Amongst those who did not support the proposition that research is needed into our relationship with technology, one practitioner suggested that there is such a degree of awareness of computers that our relationship with technology is no longer an issue.

Not important-I do not think that is applicable any more. People are becoming more aware of what computers do and how they contribute to society. Our children need to be more involved with computers as they go through their education. They [computers] will have to become more economical for schools to afford them. As the children grow up, computers will become part of life.

Practitioners were more evenly divided on some of the other research areas we put to them. We suggested that perhaps the problems of computing are mainly technical. We need better hardware, software and systems. Those who disagreed with the importance of the area stated that it is going to happen anyway.

The hardware and software we have at the moment is very good and improving all the time. It is adequate for today's standards. We have come a long way in a short time.

The practitioners were also divided on the importance of making computing economical. We suggested that perhaps the problems with computing are mainly economical. How do we make it pay?

No, this is the big argument but it shouldn't be. It's economical if you're using it properly, but unfortunately 90% of people using computers are not making it pay and they don't even realise it. They've got CAD because they've got to have CAD because of the client or, if they're an engineer, the architects insist on them having it.

Only a minority agreed with the importance of the final four problems we listed. We suggested that perhaps CAD and CAD services need to be marketed better. The pace of change is a problem for practitioners. IT is inflicting changes on practice at a rapid rate. There is little chance to take stock, assess the situation and ask if we really want to go where the technology is leading.

It's not that it is taking us somewhere that we may not want to go, but rather that we are travelling so fast that we are missing a lot of opportunities. I think we could be doing a lot more, but we do not have the time to sit back and think through all the opportunities and realise them.

We also suggested that perhaps the ideological entanglement of technology with rationalism needs to be unravelled. This problem met with some confused responses, but in general was not regarded as important by the majority of practitioners interviewed. Amongst those who thought it important, one practitioner gave a detailed response.

Very important-the reason that I think that is important, is that for a lot of lay people the idea of a computer is completely tied up with the idea of cold, hard-headed rationalism, and this actually prevents a lot of people from accepting or using the technology because they would consider accepting computers to be the same as accepting a cold, hard-headed, rationalistic approach to life. If we could break down the concurrence in people's mind of computing and rationalism, a lot of interesting things could happen with computers in practice.

We also suggested that perhaps little is understood about the design process, and computers help our understanding. The majority did not see this problem as important.

There is room to understand the design process better but the computer will not contribute to our understanding of the design process. It is only a tool we use in the design process.

I think that a lot is known about the design process, and I don't think computers will help.

Of all the research problems we posed, none were regarded as unimportant by an overwhelming majority of practitioners, and many problems seemed to divide the interviewees. It could be argued that the ordering given above should be reversed. Those problems for which the respondents were divided could be precisely those issues most in need of research. It could well be that half of the CAD community are unaware of the problems seen by the other half, and need to be persuaded of their importance.

We asked if the practitioners had seen any changes in what were regarded as important research issues ten years ago compared with now. We also asked to what they attributed the change. Most respondents considered the issues to have changed in emphasis rather than in kind.

I thought it would slow down, but every year there seems to be a new problem that needs addressing. Building models will give us a platform that design systems can benefit from.

Then we had to be evangelists, now we just do it.

Ten years ago it was still relatively new. It scared people. That has changed markedly because of the widespread publicity/education, ease of use and speed. Everything contributes to it.

You have the same problems as you did ten years ago. You're always limited to your budget restraint, finding the right type of people for the office and for a particular type of project. For a hospital project you have people who are specialised health planners but that doesn't necessarily mean that they are specialised at using computers.

The issue ten years ago was whether you wanted to do it or not. The issue today is how. The issue in ten years time will be those have and those that haven't. It's just the evolution over twenty years of going from a craft based office to a production line machine based office.

The whole ethos of computer use ten years ago was based on questions of centralised management of a big capital intensive resource and limiting access and so forth. Now all the questions have to deal with a sort of personal relationship between a particular user and a particular computer and how that is enhancing their work or otherwise. In addition there are the social issues of how this change in work practice allows us to work in radically different ways. The most interesting question for me is how computerisation is going to help me work on projects in other countries whilst being in Sydney. How is it going to help me work with a team of people spread all over the place, the people that I want to work with and not just those who happen to be there? So I am interested in the way it is structuring my work life by increasing these opportunities. I am interested not only in the computing side but also the communication side.

The issues now are related to presentation techniques and design related activities. Ten years ago, it was how do we make them do what we did traditionally [as working drawings]. The change is due to enhanced skills and more powerful machines which can handle the technology.

Significantly different. The perceived limitations of computers have not been proven. The facility, speed and high level of complexity that it can deal with, and the increased acceptance of screen received information have dispelled these perceptions. This change has mainly come about by people using computers.

There has always been a concern about the effects of computers on society, which is still around. As the influence of the computer increases, the need to understand it's effects becomes greater. It has become much more commonplace now.

When we asked for further comments some practitioners took the opportunity to specify the kind of research they would like to see happening in this domain: specifically research that would help with coordination of IT, and the presentation but not the automation of design. Some practitioners also volunteered opinions on university research and how universities and practice could work together.

There are a lot of strengths out there in practice that are not present in academia, and at the same time there are a lot of things going on within the academic world of computing which are very interesting and would be helpful to practitioners. All the philosophical implications of computing that are being explored and discussed, and the radically superior way in which the academic world has set up its own computing networks and maintains these international communications, would in the private world be a radical innovation. What I do not see happening is any real kind of interfusion between these two domains of expertise.

Any kind of research program that, rather than maintaining an arms length relationship between the academic world and the world of practice, and that actually promotes a cross-cultural transfer of knowledge, I would think is very important.

Conclusion

The influence of computers on design practice is substantial. Computers are implicated in major changes to work practices. The work tasks of the practitioner are changing as a new equilibrium has been struck between professional tasks and those of draftspersons, middle management and secretaries. Some firms see operating a CAD system as a professional skill. It also appears that computers are used not only for administration and CAD drafting, but also in providing support during the design process. The environments in which computers reside appear to be different than those occupied by drawing boards. It is also apparent that projects in which CAD is used are managed in subtly different ways to convention projects. CAD has changed notions of where accuracy and precision enter into the design process. Computers also affect the dynamics of communications within the office. They are also affecting the way practitioners communicate with clients and consultants. There is an interest in teleworking and other experiments in spatial mobility amongst practitioners, though this change is only just starting to be felt.

The profile of firms is changing and some practitioners believe the viability of small firms is enhanced by computerisation. Some firms have also cast themselves in a totally different light due to computerisation. Some have diversified into new areas, such as multimedia production, or moved into the role of custodians of databases. Practitioners seem more resilient to the pressure of changes now than they were ten years ago. There is more folk wisdom about now for coping with problems, better support networks, and the equipment and software is easier to use and more powerful. Computer components and software are more modular, interchangeable, and the individual components are less expensive. Even though a whole system may be a major expense for the firm, the individual components are not. Acquisitions can be made incrementally and on a trial basis. This has made a major difference to how IT is appropriated in practice.

Practitioners who use CAD are very interested in the potential of computers and what the future holds. Computers are not used impartially, but, as the responses to our questions about trends indicated, are also caught up in hopes and expectations for a bright and prosperous future of better integrated computer systems and environments more attuned to the needs of designers.

Acknowledgments

This work is supported by an ARC Small Grant. The following firms assisted in the survey: Ancher Mortlock Woolley, Australian construction Services, Cadpac Solutions (division of BCM Systems), cMacd Systems, Crone & Associates, Direct Applications, EGO Design, Gutteridge Haskins and Davey, H.O. Woodhouse and Danks, Havens, Kirkwood and Meertens, Hawker de Havilland, Hydranautics Australia, Jackson Teece Chesterman Willis, John Sparks and Associates, Lawrence Nield and Partners Australia, Malone Buchan Ednie-Brown, Network Design Management, Ove Arup and Partners, Pacific Architecture, State Projects (PWD), Strategic Technology Group, Westfield Projects, Westpac Banking Corporation (Property Division), Whitelaw and Chrystal, Woods Bagot, Woolacotts Consulting Engineers. The following also assisted in the survey: David Marchant and John Rollo.

References

Bjørn-Andersen, N. 1983. "The changing roles of secretaries and clerks," in New Office Technology: Human and Organisational Aspects, Otway, H.J. and Peltu, M. (eds), Frances Pinter, London, pp. 120-137.

Coyne, R.D. 1989. "The implications of computer use in architecture Schools," ADTRA: Journal of the Architectural Design Teaching and Research Association, Vol. 1, No. 2, pp. 5-12.

Crozier, M. 1983. "Implications for the organisation," in New Office Technology: Human and Organisational Aspects, Otway, H.J. and Peltu, M. (eds), Frances Pinter, London, pp. 86-101.

Cuff, D. (1991) Architecture: The Story of Practice, MIT Press, Cambridge, MA.

Danziger, J.N. 1985. "Social science and the social impacts of computer technology," Social Science Quarterly, Vol. 66, pp. 3-21.

Forester, T. (ed) 1989. Computers in the Human Context: Information, Technology, Productivity and People, Blackwell, Oxford.

Gotlieb, C.C. and Borodin, A. (1973. Social Issues in Computing, Academic Press, London.

Gutman, R. (1988) Architectural Practice: A Critical View, Princeton Architectural Press, New Jersey.

Kling, R. and Iacono, S. 1988. "The mobilization of support for computerization: the role of computerization movements," Social Problems, Vol. 35, No. 3.

Long, R.J. 1987. New Office Automation Technology: Human and Managerial Implications, Croom Helm, London.

Marien, M. 1989. "IT: you ain't seen nothing yet," in Computers in the Human Context: Information, Technology, Productivity and People, Forester, T. (ed), Blackwell, Oxford, pp. 41-47.

Mumford, E. 1983. "Successful systems design," in New Office Technology: Human and Organisational Aspects, Otway, H.J. and Peltu, M. (eds), Frances Pinter, London, pp. 68-85.

Olley, H. (1992) 1990 Profile of the Architectural Profession, Royal Australian Institute of Architects: National Division, Canberra.

Otway, H.J. and Peltu, M. (eds), 1983. New Office Technology: Human and Organisational Aspects, Frances Pinter, London.

Radford, A.D. and Stevens, G. 1987. CADD Made Easy, McGraw Hill, New York.

Stevens, G. 1989. "The social construction of computer-aided architectural design," ADTRA: Journal of the Architectural Design Teaching and Research Association, Vol. 1, No. 2, pp. 51-56.

Wainright, J. and Francis, A. 1984. Office Automation, Organisation and the Nature of Work, Gower, Aldershot, Hampshire.