markabel@uswat.uswest.com (Mark Abel) (01/18/91)
This is a followup to a discussion between Nick Szabo, Mark Shepherd (& later Alan Wexelblat). Let me say first that I strongly support Mark Shepherd's position that visual and auditory communication is a critical factor in interaction. This is based on 6 years of research in Multi-media Groupware systems at Xerox PARC and U S WEST Advanced Technologies. (Many other groups have been exploring these issues as well - e.g., Bell Labs, Bellcore, NTT, NEC, Olivetti, Xerox EuroPARC, etc., etc.). Social scientists claim that two types of communication are taking place in any interaction: 1. "information transfer," and 2. social or emotional info. For example, if two hackers are discussing the best way to code something, they not only exchange information (e.g., details of how a program works) but they also read each other and decide things like "Do I believe this person?" "Do I trust their opinion?" "Do they like me and/or believe me?" etc. As Alan Wexelblat said, the visual channel carries most of this socio-emotional information by allowing people to see gesture, posture, facial expression, eye-contact info, etc. Admittedly, some socio-emotional information is carried via other channels (e.g., through tone of voice, through "flames" in e-mail, etc.). Although one can coerce single channel interaction systems like e-mail into providing socio-emotional information (e.g., smileys :) ), a richer set of interaction media that includes two-way video, quality audio etc. is almost certainly much better to convey this sort of information. In short, why limit ourselves, especially when full multi-media communications solutions will be here in the 1990's? -Mark Abel U S WEST Advanced Technologies
shackelf@tlab2.cs.unc.edu (Douglas Shackelford) (01/19/91)
In article <14533@uswat.UUCP> markabel@uswat.uswest.com (Mark Abel) writes: Stuff deleted... > >Although one can coerce single channel interaction systems like e-mail into >providing socio-emotional information (e.g., smileys :) ), a richer >set of interaction media that includes two-way video, quality audio etc. >is almost certainly much better to convey this sort of information. In >short, why limit ourselves, especially when full multi-media communications >solutions will be here in the 1990's? > Something that has always been interesting to me is that the smiley ( :) ) is in some sense a humor equalizer. That is, everyone does the smiley in about the same way. It is hard to misunderstand a smiley, whereas, more complicated facial and speech behaviors can be easily misinterpreted. In some ways this is related to compact disc technology. By digitizing the sound, we can eliminate the background noise. I wonder if an advantage of the smiley is that it helps to reduce the noise that is inherent in any human conversation. --Doug Shackelford shackelf@cs.unc.edu "When I drink alone, I prefer to be by myself"
kevinc@cs.athabascau.ca (Kevin Crocker) (01/22/91)
szabo@crg5.UUCP (Nick Szabo) writes: >In article <14533@uswat.UUCP> markabel@uswat.uswest.com (Mark Abel) writes: >>This is a followup to a discussion between Nick Szabo, Mark Shepherd (& later >>Alan Wexelblat). >>Social scientists claim that two types of communication are taking place in >>any interaction: 1. "information transfer," and 2. social or emotional >>info. >This is a good distinction, and allows me to state my position >succinctly: category (2) is more harmful than helpful to technical >communications. >Allocating technical workgroup bandwidth for (2) will lead to a poorer >solution of that group's technical task. The best evidence is scientific >journals, which can communicate so much technical information >precisely because they are totally devoid of category (2) bandwidth. >Similarly, source code control systems lack category (2) content and have >proved very useful in coordinating the work of software engineering teams. Nick, I agree that for discussions of technical information the reductions of side band noise, in this case emotion, is beneficial. However, not all groupware discussions are technical in nature. At this university we use groupware on a very primitive level but little of it is used for technical discussions. My job is to design and develop courseware in my chosen field (Financial Management) for consumption by students. I work closely with an editor, an Istructional designer and a visual designer. Our discussions generally have little to do with the technical aspects of either the content (my area), the instructional paradigm (the ID's area), or the visual presentation (the Visual Designer's area). We assume, perhaps in error, that each of us are proficient in our field of study. What we use groupware for is to communicate ideas in a brainstorming session to determine new ways to apply already known concepts in an attempt to provide a learning experience that is more applicable to the learner. In this sense, we attempt to provide learning materials in a format that is suitable to the learners dominant learning mode with the addition of some standard presentation formats. In our courses we use print, graphics, video, audio, TV, radio, computers and any other media that we think would benefit the learner. Not that we use every media for every course but many of our courses are multimedia. In order to develop this type of material I need access to developmental resources to make this available. Therefore, having groupware capabilitites that permit multimedia communication is very important to me and our group. At the moment we are attempting to develop a multimedia workstation for a proposed M.B.A. degree. Kevin -- Kevin "auric" Crocker Athabasca University UUCP: ...!{alberta,ncc}!atha!kevinc Inet: kevinc@cs.AthabascaU.CA
szabo@crg5.UUCP (Nick Szabo) (01/22/91)
In article <616@aupair.cs.athabascau.ca> kevinc@cs.athabascau.ca (Kevin Crocker) writes:
[wants to try multimedia groupware to develop multimedia course]
Good example! Given that the product contains sound or video,
groupware that manipulates these media might be useful for the
product developers.
This is quite distinct from using sound and video to communicate emotions
between workgroup participants. Where the product involves emotions
(eg advertisements), groupware that communicates emotions might be useful.
Where the end product does not contain either emotions or sound/video
-- the large majority of technical work -- groupware that communicates
emotions, and for the most part sound/video capability, is not useful.
--
Nick Szabo szabo@sequent.com
Embrace Change... Keep the Values... Hold Dear the Laughter...pease@amarna.gsfc.nasa.gov (Pease) (01/23/91)
I recall attending a short course at UCLA in the 60's and one of the presenters had long hair and a beard. At that time the long hair & beard hippie look had not reached the east coast or the media. Guess what, I didn't absorb much or anything of what he had to say, all I could do was look at his weirdness. At that same gathering I had my first view of mini-skirts by some "daring" UCLA coeds, and do you think that helped in being able to study the material from the course? Wow, I'll never forget the experience. Anyway, in some cases the emotional content of such scenes certainly do distract from the purpose at hand. But maybe the emotional content was just as important as all that technical stuff, 8^). I would also like to make the point that there are some people whose preferred method of absorbing information are by sight and others by ear. That is some prefer to hear someone talk and others prefer to read the paper. Those who prefer to read take lots of notes when listening to a speaker compared to those who prefer to hear what is said. So, I think that for those that are "sound" people they would not find this type of written stuff very satisfactory, while those of us who are "sight" people can't see that audio is needed at all. Now to change the subject - About 10 years ago I took a graduate class in Small Group Processes (in the Psychology department). My theme paper was the use of computers in group processes. At that time networking was just beginning, and some researchers were investigating computer conferencing. Thus I indicated that computers would be used for recording individual's ideas, organizing them into inter- related subjects, and then accessing the combined organized thought of the group. I also indicated that some statistical techniques could be used for arriving at decisions based on an analysis of group members opinions (i.e., the "Delphi method"). I reported studies that showed that the Delphi method produced better decisions than was obtained by having one "person in charge" making the decision even after hearing or reading the same opinions that were given by the group members. In my paper I took a set of monographs entitled "Looking into Leadership" by Dr. Warren H. Schmidt of UCLA Busisness School in which he identified 13 principles of group behavior. I will present these principles below, but omit my analysis of how the use of computers/networking would relate to them. I will leave it to you to reflect on this. 1. Successful group productivity depends on the ability of the members to exchange ideas freely and clearly and to feel involved in the decisions and the processes of the group. 2. No group can be fully productive until its members are willing to assume responsibility to the group. 3. An effective group has a clear understanding of its purposes and goals. 4. An effective group is flexible in selecting its procedures as it works toward its goals. 5. An effective group has achieved a high degree of communication and understanding among its members. 6. An effective group is able to initiate and carry out effective decision making, carefully considering minority viewpoints, and securing the committment of all the members to important decisions. 7. An effective group achieves an appropriate balance between group productivity and the satisfaction of individual needs. 8. An effective group provides a sharing of leadership responsibilities by group members - so that all members are concerned about contributing ideas, elaborating and clarifying the ideas of others, giving opinions, testing the feasibility of potential decisions, and in other ways helping the group to work on its task and maintain itself as an effective working group. 9. An effective group has a high degree of cohesiveness (attractiveness to its members) but not to the point of stifleing individual freedom. 10. An effective group makes intelligent use of the differing abilities and interests of its members. 11. An effective group is not dominated by its leader or by any of its members. 12. An effective group can be objective about reviewing its own processes; it can face its problems and adjust to needed modifications in its operation. 13. An effective group maintains balance between emotional and rational behavior, channeling emotionality into productive group effort. In my conclusion, I noted that Managers tend to overlook the human aspect of the group process all to often and superficially. They focus mainly on productivity, efficiency, costs, meeting deadlines, and quality of results; and that for these reasons computers will be used in group process for the benefits they will bring to these areas. The human aspects of group interaction will be diminished in the computerized system. However, there are some aspect which us humans would find desirable, such as greater self direction, ease of contributing to the group product, more flexibility in the times one works on the group tasks, and less travel or time constraint in getting everyone together to interact; but, for many of us there is a joy which comes from the human interaction of sharing ideas over coffee which would be sadly missed if all work were done over a computer network. Phil Pease My witty disclaimer - everything I perceive, through either sensory or extrasensory means, has been filtered to such an extent that you had better not attempt to attribute anything I say to anyone else.
lippin@oreo.berkeley.edu (The Apathist) (01/23/91)
Recently szabo@crg5.UUCP (Nick Szabo) wrote: >Where the end product does not contain either emotions or sound/video >-- the large majority of technical work -- groupware that communicates >emotions, and for the most part sound/video capability, is not useful. I design software which contains little emotion, sound, or video. Nevertheless, when I'm considering new features or new ways of doing things, I talk to people face-to-face, rather than by email. I want their emotional reaction to the changes, so that I can design a product that will please people. Wearing another hat, I'm a mathematician. In this role, I attempt to produce theorems and proofs. I'm interested in my colleagues' emotional reactions to these as well. Why? Because a good theorem isn't just true -- there are scads of true theorems just lying about -- it also has to be either interesting or useful. Or both, for a really good theorem. If it's neither interesting nor useful, there's no way it'll get published. This brings me to a secondary product: communication. I want the papers and courses I put together to be clear and easy to follow. Emotional feedback is essential in getting the presentation right. I greatly prefer one-on-one tutoring to classroom lecturing for this reason. Similarly, it's generally accepted that small classes are more effective than large lectures. And getting the reactions of a test reader is a very good way to find the confusing spots in a paper. (And a sense of confusion is *very* difficult to gauge through email.) When considering needs like these for emotional communication in technical fields, I'm hard pressed to find technical fields that wouldn't benefit from working it better into their groupware. --Tom Lippincott lippin@math.berkeley.edu "I enjoy working with humans, and have stimulating relationships with them." --HAL
kenn@intrbas.uucp (Kenneth G. Goutal) (01/24/91)
Heh heh... You *think* smileys are hard to misinterpret! But is this true? I had unquestioningly assumed so until you brought it up. Now I am not so sure. It seems easy to believe that, by reducing the number of 'facial expressions' to a very small number, the writer will be able to 'assume an expression' that will be universally understood. I see at least two problems with this: (1) What if there is no icon that accurately represents how I feel? If this happens, then no matter which one I pick, it will be intrinsically in error by some amount, right at the outset, no matter who interprets it. (2) There really is no basis for believing that the emoticons are universally understood. Take the smiley -- ":-)". (Ignore for the moment that I use the "nasal dialect"!) Suppose I say: This will cause the end of civilization! :-) Does this mean that I am happy that civilization will end? Does it mean that I am happy that "This" will be cause thereof? Does it mean I'm happy about some side effect (e.g. no more styrofoam cups)? Does it mean that I'm just in a good mood, and perhaps I'm joking, but I'm not a winking sort of person? I'm afraid that this is not a very imaginitive example, and does not do justice to the sort of different slants that readers can impose on emoticons depending on their own background, their own mood, etc. This is not to say that emoticons are not an improvement over natural facial expressions and body language -- perhaps they are. But I think that there is ample room for miscommunication in them, and we should not fall into the trap of believing that the use thereof will cause emotional miscommunication to disappear. -- Kenn Goutal ...!linus!intrbas!kenn ...!uunet!intrbas!kenn
stewarte@sco.COM (Dr. Luther's Assistant) (01/25/91)
Even net.pundits were baffled when shackelf@tlab2.cs.unc.edu (Douglas Shackelford) wrote: >Something that has always been interesting to me is that the >smiley ( :) ) is in some sense a humor equalizer. That is, >everyone does the smiley in about the same way. It is hard to >misunderstand a smiley, whereas, more complicated facial and >speech behaviors can be easily misinterpreted. Yes, but this is precisely because the smiley contains less information -- and as a result of that, people have to shoehorn a variety of meanings into that one symbol. The distinctions between good-natured leg-pulling, sarcasm, and poking fun at someone can't be made at all with the smiley, wheras with those more complicated behaviors there is some hope that they will be correctly interpreted. In short, I think they're less likely to be misinterpreted because they don't lend themselves to much interpretation at all. -- Stewart -- "The old stereotypes must be lost that peace and knowledge and love are soft." -- Blastmaster KRS-ONE /* uunet!sco!stewarte -or- stewarte@sco.COM -or- Stewart Evans */
kevinc@cs.athabascau.ca (Kevin Crocker) (01/25/91)
szabo@crg5.UUCP (Nick Szabo) writes: >In article <616@aupair.cs.athabascau.ca> kevinc@cs.athabascau.ca (Kevin Crocker) writes: >[wants to try multimedia groupware to develop multimedia course] >Good example! Given that the product contains sound or video, >groupware that manipulates these media might be useful for the >product developers. >This is quite distinct from using sound and video to communicate emotions >between workgroup participants. Where the product involves emotions >(eg advertisements), groupware that communicates emotions might be useful. >Where the end product does not contain either emotions or sound/video >-- the large majority of technical work -- groupware that communicates >emotions, and for the most part sound/video capability, is not useful. Sometimes it is difficult to separate the media from the message. Especially in this day and age of media management (sic!) that is exercised by a wide variety of sources but most specially the media experts like advertisers. How much of TV advertising is devoted to a single clear message and how much of it is devoted to manipulating the emotive aspects of perception? In attempting to develop materials in a multimedia environment there are cues and informational content in all aspects of both the message and the media container or channel(s). Whether the information that exists within the message and channel is relevant and useful is of course debateable -- what is not debateable, at least to me, is the rather arbitrary decision to exise part of the message or informational content. As the receiver (decoder), I would prefer to exercise control (and perhaps build or develo filters) of the informational content that I receive. As a sender (encoder), I should be cognizant of what I compose, how I compose it, and how I provide for the enrichment of the receiver, as well as providing the capability for receivers to filter the message as they see fit. Now, when it comes to groupware, much communication takes place in side bands whether it was intended or not. For communication of pure technical information where the focus is on encoding and decoding a pure signal with 100% accuracy other side bands of communication will introduce signal loss and dropout. For other types of communication, side band information may or may not be beneficial. Kevin -- Kevin "auric" Crocker Athabasca University UUCP: ...!{alberta,ncc}!atha!kevinc Inet: kevinc@cs.AthabascaU.CA
kevinc@cs.athabascau.ca (Kevin Crocker) (01/25/91)
lippin@oreo.berkeley.edu (The Apathist) writes: >more effective than large lectures. And getting the reactions of a >test reader is a very good way to find the confusing spots in a paper. >(And a sense of confusion is *very* difficult to gauge through email.) One way that I encourage the groupware concept is when I ask several people to test or proof material that I have developed. The way that this is accomplished is to get these people together and get them to go through the material while I am present. This is not using computers as a facilitator of the groupware process. The interactions that take place are both numerous and complex. One thing that I would like computer groupware to capture is this richness of interaction and information. At the present time I feel that the computer environment needs to start at providing a method of transmitting the various signal sources such as audio and video. Just some more $.25 worth. Kevin -- Kevin "auric" Crocker Athabasca University UUCP: ...!{alberta,ncc}!atha!kevinc Inet: kevinc@cs.AthabascaU.CA
szabo@crg5.UUCP (Nick Szabo) (01/29/91)
In article <632@aupair.cs.athabascau.ca> kevinc@cs.athabascau.ca (Kevin Crocker) writes: > >...there are cues and informational content in all >aspects of both the message and the media container or channel(s). >Whether the information that exists within the message and channel is >relevant and useful is of course debateable -- what is not debateable, >at least to me, is the rather arbitrary decision to exise part of the >message or informational content. We are explicitly designing groupware. We _must_ make decisions to include or leave out certain media, and to use those media in certain ways (whether by software constraint or usage convention). Hopefully these decisions are not "arbitrary", but they must be made. >As the receiver (decoder), I would >prefer to exercise control (and perhaps build or develo filters) of the >informational content that I receive. That's a good idea. Any groupware that allows the senders to force-feed the receivers will not be something I want to receive. The receiver wants to be able to change channels, grep for favorite subjects, clip 'n save the good stuff, filter out noise, and do it at his/her convenience not the sender's or other receivers'. >As a sender (encoder), I should >be cognizant of what I compose, how I compose it, and how I provide for >the enrichment of the receiver, as well as providing the capability for >receivers to filter the message as they see fit. Hah! Senders do not not want receivers to be able to filter anything. They want the reciever to see everything, as originally intended: they don't want to be colorized, edited, quoted out of context, skipped over or worst of all put in "kill files". Sender vs. receiver is a major source of conflict in groupware. >.... -- Nick Szabo szabo@sequent.com Embrace Change... Keep the Values... Hold Dear the Laughter...
kevinc@cs.athabascau.ca (Kevin Crocker) (02/01/91)
szabo@crg5.UUCP (Nick Szabo) writes: Kevin>In article <632@aupair.cs.athabascau.ca> kevinc@cs.athabascau.ca (Kevin Crocker) writes: Kevin>at least to me, is the rather arbitrary decision to exise part of the Kevin>message or informational content. Nick>We are explicitly designing groupware. We _must_ make decisions to Nick>include or leave out certain media, and to use those media in certain Nick>ways (whether by software constraint or usage convention). Hopefully Nick>these decisions are not "arbitrary", but they must be made. Nick, I wasn't trying to imply that these decisions were being made without regard to the eventual users for that is what this discussion is all about. I agree that the design of systems and software require the application of constraints. What we need to remember is that often important information is contained in channels that are not traditional (reading and contextual mapping). Sound does contain significant information but even more information can be conveyed through the use of visual representations (pictures/graphics/motion etc). The old truism that a picture is worth a thousand words is one that governs the communication process for many people. I will azlso admit that good writing can convey a richness and depth that can never be conveyed through a visual representation, just as an operetta can convey information that can never be captured by a picture or description of what happened. When we design groupware I feel that we must not limit the ability of the sender and receiver -- especially in this day and age where the bandwidth that is becoming available is growing at an ever increasing rate. Groupware is about "groups" of ""people"" communicating. We should provide for as many different channels of communication as possible. we don't have to implement everything that is designed into the system but it is much harder to retrofit a totally new channel of information into something that it was not designed to handle than to design for as much as possible and not implement until it is needed -- even given that what we designed will be hopelessly out of date by the time we implement. I'm going to step out on a limb here: CAVEAT -- I am not a Microsoft fanatic by any means, I use it because I have to! DOS was not designed to handle graphics and sound etc. It has taken Microsoft almost 10 years to incorporate a primitive level of graphic support into it. Loading GRAFTBLE.EXE to be able to get on screen the high order characters is incredibly poor. DO WE WANT TO DESIGN SOMETHING NOW THAT WILL TAKE TEN YEARS BEFORE IT CAN INCORPORATE EVEN WHAT WE ALREADY KNOW ABOUT vis-a-vis communication???? Kevin>As a sender (encoder), I should Kevin>be cognizant of what I compose, how I compose it, and how I provide for Kevin>the enrichment of the receiver, as well as providing the capability for Kevin>receivers to filter the message as they see fit. Nick>Hah! Senders do not not want receivers to be able to filter anything. Nick>They want the reciever to see everything, as originally intended: they Nick>don't want to be colorized, edited, quoted out of context, skipped over Nick>or worst of all put in "kill files". Nick, I'm appalled at this type of attitude. You're talking like a technician that has never communicated with anything except a machine. Communication occurs in many ways beyond electronic transfer of bit streams. We're not just talking about news/notes/conferences/etc, we are talking about interaction in at least one-to-one, one-to-many, and many-to-one relationships. The mere fact that you indicated "kill" files supports the ability of the receiver to filter messages as they see fit. The question remains -- should we build this kind of capability into the groupware or not? The ability to communicate intelligently connotes many things but at the most basic level it connotes that we understand and are aware of the level and degree of control we have over both the message and the channel. The intent of communication is twofold -- to provide for the transfer of information and to allow both the sender and receiver the ability to filter the information. Kevin -- Kevin "auric" Crocker Athabasca University UUCP: ...!{alberta,ncc}!atha!kevinc Inet: kevinc@cs.AthabascaU.CA
drd@siia.mv.com (David Dick) (02/02/91)
In <633@aupair.cs.athabascau.ca> kevinc@cs.athabascau.ca (Kevin Crocker) writes: ..description of groupware "in person" elided.. > One thing >that I would like computer groupware to capture is this richness of >interaction and information. At the present time I feel that the >computer environment needs to start at providing a method of >transmitting the various signal sources such as audio and video. I think the management of the interaction is more important, IMHO. As humans in the real world we are +-------------------------------------- continuously bombarded by sensory | As to the complexity of the world: input--the world is a complicated | I recently attended a conference place. | on virtual reality and cyberspace, | and one of the speakers noted that, Our great skill in navigating it | in his experience, he appreciated is that our perceptual systems do | reality much more after spending considerable filtering below the | time in a virtual world. It's level of consciousness. | consistent, doesn't have video | glitches, and there's always something Analogously, I think groupware | behind every door! needs to give us a way to manage +--------------------------------------- the complexity. Why go to all the expense and trouble just to recreate the chaos and inefficiency of most meetings? David Dick Software Innovations, Inc. [the Software Moving Company (sm)]
kevinc@cs.athabascau.ca (Kevin Crocker) (02/05/91)
kenn@intrbas.uucp (Kenneth G. Goutal) writes: >I think (though I could be wrong) that what you benefit from in one-to-one >face-to-face interaction vs e-mail is not *emotional* feedback, or even >the presence of feedback, but *immediacy* of feedback. In a typical A very interesting and vital observation. Here at Athabasca University (we are almost entirely a distance education institution) we have extensive experience with non-immediate communication with groups and to date the groupware metaphor has largely been conducted with paper/mail transfers. What we have found is that there is a significant amount of evidence (in our scenario) that indicates that the most critical factor in the success (don't ask how we define this one) of students is the immediacy of feedback. Thus we have instituted a hot-line telephone system, electronic mail access that guarantees responses within 24 hours to supplement our regular telephone system. We are now investigating the application of the groupware concept to the instructional design as opposed to the instructional delivery of university level education. In this light we are experimenting with computer aided communication through e-mail, file sharing, computer conferencing, file transfer, etc. Most of these experiments have become defacto operations systems rather than experimental pilot projects. However, we would like to go much further into audio and video transmission on the development side. Thus the reason for my interest in this dicussion. We have just received a major project ($1.5 Million) to develop a Instructional design system. WE are also about to invest heavily in a system wide document management system, although the emphasis will definately be broader than "document". Groupware is a vital part of what we do here and we need to get a handle on the technological alternatives. Kevin -- Kevin "auric" Crocker Athabasca University UUCP: ...!{alberta,ncc}!atha!kevinc Inet: kevinc@cs.AthabascaU.CA