Click for next page ( 90


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 89
GENERAL DISCUSSION James Burrows National Bureau of Standards Dr.Licklider: Jim Burrows is going to moderate the general discussion session and he will be the first discusser also. Jim is head of the Institute for Computer Science and Technology at the National Bureau of Standards, so I won't blame him a bit if standards creep into the discussion. Mr. Burrows: Thank you very much, Lick. I think one of the problems that I have in being a speaker for Technology and Standards is that I am continually being asked to solve managerial, organizational and personnel problems with technology and I don't know how to do any of those things with technology. Somenow, that is a skill we haven't been able to automate yet. We, at the Institute, do have programs for standards in local area networks and computer based office systems, mostly having to do with extending the options you have in terms of being able to interconnect equipment and exchange media. We are not in the area of doing human factors experiments at the moment, although we hope we will get to that level sometime. Right now, we are mostly in hardware and in how to look at the use of office automation, but not actually doing experiments. One of the questions I would like to ask the speakers here is, would any standards have helped you do the jobs you have done thus far? Mr. Walsh: One of the requirements we recognized very early on working in an IBM main frame environment, was that we wanted to communicate initially in a largely job entry type format, but eventually we wanted to move to an on-line environment. It was key for our selection that we have a vendor who could offer us total transparent support in the communications mode. Our initial thrust was with DEC. In trying to put the communications together, their equipment didn't interface very well with our IBM environment. Wang did. 89

OCR for page 89
Also, in trying to communicate between our Vydex and our Wangs, we had total failure. A standard communications protocol, a bi-synchronous protocol, would have helped us considerably. I think it would have also afforded us an opportunity to deal with a number of other vendors. There are situations where, for instance, we might not have good Wang support, but we might have excellent Data Point support or excellent vydex or Xerox support. If a standard had existed, I would be more inclined to look at a number of vendors. Mr. Burrows: Thank you, John (Walsh). Does anyone else believe that standards might have helped you, or were you able to make your own standards? Most of the users appear reluctant to standardize across other vendors lines. You get hold of the manufacturers and ask them: "Why don't you standardize?" The answer is: "We don't want to sell inter- changeability. We sell everything we manufacture. Our production lines are flooding. Everything we put out the door, we can sell. Why should we change? We sell features that differentiate us, not make us the same. It is up to you to worry about whether you want standards and if you want standards, why don't you just buy mine all the time." That is the argument you get from them. So, the only people who are interested in standards are the users. Many manufacturers are not interested, at least not at this time. Further, most of us have a very tough time finding a place where we can get standards. The fastest way to get something done is like John (Walsh) said. He selected DEC and he asked DEC to do something for him. There is some nice leverage there, some of "I give a little and you give a little" that makes things happen. When you get in the environment of asking all of the manufacturers to do something standard, they don't know who you are going to buy from. They don't have any individual incentive. So the standards process is very slow. I don't expect to see universal standards unless the users need them and find some way to gather together and ask for them. Mr. Burrows: Could we now have general questions from the audience? Question: I am from the Department of Justice. In looking at the figures which divided up what people were doing in terms of tasks and how that was affected by office automation and how the amount of time people spent doing their work was affected, two things struck me. One is that you didn't include time for "goofing off" such as coffee break time or anything like that. You only had work time. I have found that omission is a big problem. If you don't include the amount of time that is lost during the day, the figures aren't really meaningful and as a result you don't know what you are comparing when you look at end results. 90

OCR for page 89
The second problem is that all of the workers, professional and clerical, as soon as they see you counting anything and conducting a study which quantifies their work, they resist the fact that you are just simply taking a hard look at changing the office system, the office structure or the way things are done, even if you make things easier. There is usually an hour, or two a day that is not accountable that the worker can alter to change his productive image. I wonder how do you approach that situation and after approaching it, what kind of results do you have, because you didn't show anything in the numbers. Mr. Walsh: Okay. There is another time factor that I think John Hogan would agree with that is probably five or six percent of a workers total time. I agree that any time you try to quantify what people do, they are going to respond and fill those hours up with work rather than say they are not working full time. When you do a sample, as we did, in a population of 2,000 and you do a fairly high sampling, let's say l0 percent of the managerial work force, and you do it across different functional groups, the chances are that your statistics are going to produce something that is fairly close to a semblance of how people work. In our case, I think that the sampling that we used was so large — that is a very large sample — that I think the results translated into accurate numbers. I was concerned when I did it that it would correlate and I didn't have a lot to go on other, than those three studies. But they did correlate pretty closely. Then three years later, I saw some of the Booz-Allen results that talked about a l9 percent opportunity regarding management. I know that the Department of the Army did similar work and I think they said the value was .between l6 and 22 percent. Therefore, you know, the real opportunity value is in there somewhere. So, I feel pretty comfortable with those numbers. John (Hogan) might want to add something. Mr. Hogan: One of the observations I might make is I think the numbers are not wildly wrong. I think there is a difference between accounting for the way people spend their time in a real sense and just calculating numbers in terms of mental arithmetic or the association of ideas and the persuasion of other people, things of that kind. As we move into the white collar area, the information worker's area, the perquisites that apply are so many more than in the other production areas where we have experience. Wherever we put MTM systems in, they have built in overhead to account for an amount of personal time. Everybody lives with those. 9l

OCR for page 89
But your employees or mine, professional level, some afternoon their car is supposed to be ready at the lunch hour. They had the brakes fixed and, alas, the car was not ready at noon, but it is going to be ready at 2 o'clock. Almost anyone in an office could be gone for a half an hour, go get the car and bring it back. That is a very different kind of a phenomenon than is presumed in the accounting for work and the way people spend their time. I think the standards are altogether different. I think standards have to apply to different kinds of products and their quality. To a degree the standards should originate from the people themselves, as they probably won't live with anything else. It is a complicated matter, but I think I know the ways in which to solve it. Mr. Walsh: I think one of the problems is a matter of definition. John (Hogan) talked about us working in a communications mode 60 to 70 percent of the time. That is probably true. It is a matter of definition whether you include telephone usage as part of the communications, or other shadow functions such as meeting in the elevator or talking informally. Is that a formal meeting or informal meeting. A whole lot of it is definition. I think if we had some definitions three years ago to work from, we would have been better off. I think John is now bringing a more professional, defined approach to these different tasks than we had as a frame of reference in l976. At the time there really were no standards. Mr. Burrows: I would like to ask Mr. Mertes how seriously do you take the possiblity of an electric power failure, particularly in Chicago and what should you do about it, if anything? Mr. Mertes: We take a power failure very seriously, particularly in Chicago. We happen to be fortunate enough that we haven't had a failure yet. We don't even have power backup on our computer center. We are currently tied into two generating plants. Mr. Burrows: Oh, you have two sets feeding your bank. Mr. Mertes: Yes. Correct. We do take seriously what will happen when the electronic mail system goes down. It is a new phenomenon when everyone in the bank could be angry at you simultaneously. Power loss is a very serious consideration and that is why we went to the two generating stations and are looking now at what kind of computer power backup we should provide on an ongoing basis. It becomes more serious as you get into the office function and are really interacting with people all day, every day. Question: I have two questions that bridge all four of you: however, I will address them to Mr. Walsh and Mr. Mertes. First, I was wondering what efforts you make to protect your corporate 92

OCR for page 89
records. Second, I am troubled because you both appear to be looking at the office as it is currently structured and not re-examining job functions as Mr. Hammer suggested this morning. Mr. Walsh: I think the first question is an excellent question. We have for years been concerned about the protection of data in the main frame environment and we have very formalized procedures. We have shredders. We have very defined standards as to the maintenance of security in our environment. As we proliferate systems into the distributive environment, we are dealing more and more with users who are not familiar with the security and confidentiality factors, even when we give them procedures to archive and set up audit trails and do the right things. We are finding that even after we go back to a user with our security concerns four and five times, data is still left on the system that is accessible to many people through numerous terminals. Users will archive sometimes, but still leave open data on the system. I was shocked one Saturday when we went to test a new program — I happened to be on a Wang OIS system — to go through the index and see that all the salary information, all the bonus information, everyone of the directors and officers of the company. I am going to tell you the practicality of the real world. We went to that user and attempted to instruct him that this shouldn't be the case. We pointed out all of the deficiencies. We gave him a set of procedures. I am oriented towards formalization: giving procedures and then working with them from that. It is sort of a CYA, too. They really didn't listen to us. In fact, they were very aggressive about us looking into their data. Our internal auditing department looked at this same user a year later and found the same situation. It emerged as an issue on the president of the company's desk, much to my chagrin. But that is one small example of the problem that I am having with this. As we progress into a telecommunications mode with a lot of these systems, the question of encryption is becoming an issue. We are in a multi- national environment and a lot of people would like to know our marketing plans. So, the whole issue is very serious and very important. Mr. Mertes: First of all, to put availability in some perspective, we are centralized in operation. I run Europe and all of our branches out of our Chicago main data center. We have looked carefully at availability and continuity of operations because that is a big concern to us. We have backup files and have arrangements with multiple firms to support parts of our actions should we have a failure. 93

OCR for page 89
The thing that we are most disturbed with now and the one thing which will most likely cause us to move towards a second data center is the growing on-line environment. Just like electrical power outage, you really need two data centers running at about 70 percent capacity or more. We are laying plans to move in that direction. We aren't there yet. The thing that we found that isn't readily available — at least to my knowledge — is if you have an all on-line environment with everyone hooked up, is ease of switchability to the second environment and ability to just continue running. That is something we are trying to pursue further with AT&T and Bell labs. Now, a brief response to your other comment about studying not how people are operating today, but how they should operate. Our view is we believe that if we put in tools that will permit or provide the ability to work quicker, those personnel who are quality workers or the better managers will use those tools to enhance their productivity and change the way they do business and those that aren't willing to change will eventually fall further behind and a wider gap will exist between them and those that are willing to change. As for privacy, we have at least the same level, in fact, better privacy now with the on-line environment than we used to have. We have passwords which we control ourselves. We don't encrypt yet. One of the things that is more private now is the payroll file, which is a big concern. In the past we have had listings laying on people's desks. Now, the manager who has the people working for him is the only one with the password. He can look at his people and it doesn't go through the secretary's hands and all kinds of other handling that it used to do. So, what we are ending up with is that there is less likelihood of things lying around as they used to. We still have to tighten up the privacy, but I would view it as tighter now that it was previously. Mr. Walsh: Can I just make one more comment on your first issue? We recently went back and looked at all of the documentation that was considered to be confidential and, really, about l5 percent of the total should be confidential and the other 85 percent should not be, so there is the reverse issue as well. We also looked at what resource we spent shredding confidential material and it turned out to be two full time people. So, that is another critical issue. Mr. Hogan: There is a second aspect to that. I think it would be hard to find someone who would come out in favor of automating just what is there today. What I was describing in terms of task analysis and taking social temperature form only part of the diagnostics as we then move toward equipment specifications. The identification of tasks within processes, with the emphasis on improving the processes and fulfillment or contribution to mission, I think, is the name of the game. 94

OCR for page 89
I think work simplification and delegation and combinations, all the things that once upon a time were housed under methods and procedures are relevant here and I certainly subscribe to what Mike (Hammer) said. Eleanor (Wynn) also had some things to say about augmentation in human efforts that bear on this issue as well. Ms. Wynn: I did come up with an approach. I was kind of stumped for a minute. I think ray approach is an artifact of my being an anthropologist. I am actually not a sociologist or social psychologist and I spend a lot of time doing conversation analysis. That has given me an orientation of looking at both the surface impact, what people recognize about what they do; and the other levels, what they are doing, without realizing that that is what they are doing. So, when I look at the way people do jobs now, my purpose isn't to say this is what you want to automate, but I am saying that in producing the results that are produced in offices, there are two kinds of input. One is the explicit, procedural, kind of articulable model that you have and the other is the way things really happen in the unacknowledged processes. So, before you go to the kind of revamping that Mike (Hammer) is talking about, you may want to look at all of the goals that are already being met by the way things are done. Some of those may not be well understood or the processes may not be well understood. I also have something to say about privacy. I think there is an issue coming up in electronic mail with regard to privacy. The thing about electronic mail is that it is very conversational. It lends itself to spontaneous expression. There is a word coined already for that spontaneous expression when it gets really expres- sive. It is called "flaming". The difference between spontaneous electronic mail and spontaneous live expression is that with electronic mail you leave a record. That is a record that somebody else can make many copies of and distribute. I have questions about that. Conventions will probably develop to handle some of my concerns, but, there really is a privacy medium, broadcasting, communications kind of issue in how you are going to treat informal spontaneous communications that leave records. Mr. Allen: I am Frank Allen from Air Force. We have heard a lot today about the interfacing of office automation hardware and communications. Several speakers have alluded to the changing work habits of individuals. I guess, one of the closest ones who came up with this was Mr. Mertes in an offhand conversation as to the scheduling of workers time and the attitudes of people in their working environment. Possibly, even getting away from the nine to five type of atmosphere. In the few brief seconds that we have left, could someone address office automation along the lines of the psychological standpoint of possible changing work habits of people? 95

OCR for page 89
Mr. Mertes: I will comment. We have done a lot of talking about changing work habits tremendously. We have staggered hours where people can start at any hour they wanted, but where they adhere to the same hours all the time. However, the issue that we spent a lot of time discussing, especially considering location independence, is how about running a pilot with a commercial officer working out of his home in Denver, never coming into the bank other than to sales meetings. We have some documents assembled on that. Also how about a pilot on working out of your home: a mode of operation which is clearly demonstrated in a lot of sales or consultant organizations. The issues that we have discussed at length come around to how would your wife or husband like you at home all day. This is a real issue because it changes the style in which you are living on a day to day basis. Another one, that has come up in discussions with our young single people, is where are they going to meet all of those neat people if they stay at home? That seems to be a real issue, but only one of transition. When everyone in the singles complex is at home, there is no problem. In fact, then, when are they going to work? Another issue, that has had extensive discussion in our group, is if people start working in their home or in their own location have they picked the neighborhood in which to live based on spending their time with the people that are their neighbors or for other reasons. We have kicked around these issues and are going to run some pilots to investigate them over the next l8 months to two years. I think there are some real interesting opportunities, real difficult transition issues and some significant cost savings. So, I think the discussion has to continue. The other one is how do you know if the worker has two jobs or not. We really have found out how poorly we measure our staff. Seeing their bodies sitting at a desk makes us feel that they are working eight hours a day. Not seeing them and not measuring their attendance suggests that maybe they could be working for four firms at the same time, doing very little. Managers feel very bad about that. So, I have recommended that I be the first one to start working at home and that didn't fly, so we are going to start some other things. But those are the issues that we are confronted with. Ms. Wynn: I think the idea of working at home is an interesting one, but I don't think it will ever take the place of going to the office completely. The work that I have done has been sort of a microscopic examination of the kind of cognitive and information processes people apply to the work they are doing as they 96

OCR for page 89
are doing it. I am talking about what you may consider to be very low level JODS and there are an enormous amount of them. People constantly teach each other how to do their jobs. This is true for people who process orders and it is especially true for people who do more complex intellectual activities. So, that is one level, just remembering what your job is and continuing to learn about it and how to do it. This is something you get as a benefit of going to work every day. Another one is remembering what is important about your job. You know, my job is really created every day. If I go away for a vacation somewhere, then the world becomes the place where I have gone to and my work can sort of vanish as a reality. We don't realize this; because, in fact, we do go to work every day and we are there all the time, getting reinforced in the importance of the goals that our organization has and the value of our work to other people, the value of our performance and all of this kind of thing. But, in fact, I think that the socialization that takes place at the office — first of all learning and, second of all, maintaining the motivation of the job — is something that you can't replace by having people work at home on-line. I think that working at home is a good option for part of the time. The other thing is, of course, that the work place is the main organizational contact that people have outside of the home. Americans today don't have that much participation in social groups outside of their homes. Some of them do participate in voluntary organizations, but nothing to the extent of their participation with other people at work. I don't think that should be dropped out. A result, from an experiment of people working at home on-line, was that the participants had to be called into the office at periodic intervals because the electronic communications did not have much conversational range — all the features of conversation were not there. All you have is the words. Conversation has all kinds of contextualization and interpreting mechanisms going on to keep people geared to the fact that this is what this means and this is what that means and this is what I meant by what I just said. When you drop that out, the interpretations of what is going on can really start getting skewed. So, people who communicate exclusively electronically can actually start getting a little bit weird about what is going on, and they have to be called in and reminded of the good will of people that they have been communicating with at the office. So, for all of those reasons, although I would love to be able to work at home part time, I wouldn't like to work at home all of the time. But, it is great for the people who have to be at home all of the time anyway. That expands their association enormously. 97

OCR for page 89
Mr. Burrows: I have had the experience of working with an on-line message system twice. I mean, for production, not experimental purposes. It gets to be very frustrating when you want to clarify something with the human being you have been messaging with for the last l5 minutes only to discover that your phone is hooked up to the coupler and you cannot place a phone call. Now, what do you do. You must log out. Mr. Mertes: Not true. You need AT&T to get you a two button phone. I agree with that problem. In fact, they put one in my house so that I could make phone calls and operate my terminal simultaneously. Mr. Burrows: Was there one more question? Mr. Perry: Chris Perry from MITRE Corporation. Do you find that people get weird when they use just electronic communications? Mr. Mertes: How can I answer that? I guess in response to the comment, I don't think that is true because what happens — in fact, what we are finding with a lot of people using the electronic communication — is that they are freeing up time and having more personal interactions during a given work day because the mundane activities are getting handled through audio mail and electronic mail. It is not like a programmer sitting behind a terminal. It is like handling the mundane transactions that disappear and now what you have are more interpersonal meetings, talking about longer term kind of items. That was a by-product we did not expect, but it is turning out to be the case. Mr. Perry: I just want to comment. I really do think they get weird. We have experimented with three systems. For example, we are up on Hermes and we have been using the mail capability of the Source with different people up on these systems. One of the problems I have is when I initially started these projects, I intended to use them to give direction to people. These were a group of people who worked with me, l2 people, in one particular group. What I found actually was that they had me. They bombarded me with requests. It was just the reverse of what I thought. I am a sociologist. It is interesting that the people who are involving themselves with this integrative function in terms of bringing these things together in terms of large corporations, many of them are social scientists, not engineers or data processing or communications people. But I had a perception of one fellow who — I was on the terminal at l o'clock on a Saturday morning sending messages, now, that makes me weird — broke in between my sending messages and started communicating directly with me. That made me think that he was very weird and I haven't been able to get that out of my mind as to my perception of that individual. 98

OCR for page 89
So, it really has brought about a whole set of strange situations, based on the changing availability of time and the definition of the work day. It has just been very interesting, if that makes sense. Dr. Licklider: I am afraid I am going to have to do it. I conceded my summary time to this discussion because the discussion was lots better, but there is this matter of the bets and 5 o'clock and I have to protect my bank account. On the subject of working at home at the computer, my wife tells me, "I married you for better or worse, but not for lunch." So, I travel 30 minutes to get to my l8,000 bits-per-second connection with the computer instead of the measly l,200 I have at home. Okay, on behalf of the audience, I want to thank you all and the other speakers who aren't here. I think you were superb. You did just exactly what we wanted done and I am very happy with your performance. On behalf of the speakers, I want to thank you, the participants out there, who were just spectacular. You came on time. You had your coffee at the right time. This was very stimulating and very interesting. Now, on behalf of all of us I want to thank R.V. Mrozinski for putting this thing together. Rosy, you did a spectacular job and we appreciate it. Now, I want to give the meeting back to Lou Rader just before 5 o'clock. Dr. Rader: The meeting is now adjourned. 99

OCR for page 89