My first conscious memory of thinking about “gradients” was in college biology. Gradients of chemicals make a difference in terms of cell differentiation, our professor explained.
For the past few days, I’ve been walking the garden on the lookout for “gradients.” As an experiment, I’m going to break up the discussion of gradients into sub-categories.
Visually, all the various gradients look similar. I am categorizing them by the cause of the gradient.
So, first, we have “Perspective Gradient” … as we look at identical or nearly identical items at different distances, they will obviously appear to have different sizes and different distances between those items.
Second, we have “Light Gradient.” For example, a series of spheres or cylinders lit by a single source, the three-dimensional shape will result in a gradient of light and dark from most angles.
Third, there are often “Color Gradients” for both animals and plants.
Fourth, there are “Growth Gradients.” Color Gradients, are often a specific example of Growth Gradients. However, there are other types of Growth Gradients. For example, in many plants new growth is at the distal end and is smaller. Old growth is more central or more toward the roots; it is older, and is larger.
Growth Gradients not apply to many plants and animals. They often apply to organizations as well. Imagine a multi-national distributor with field offices in half the countries in the world. If they move into a new country, at first, there will only be a few
Conversely, Color Gradients can be quite intentionally added by the artist! Here’s a detail from a John Lennon lithograph. Note that “Boundaries” and “Gradients” reinforce each other. The Gradient makes the Boundary Stronger and vice versa.
There are also “Force Gradients” that appear in response to water, wind, gravity, magnetism, etc. One could argue that “Growth Gradients” are due to forces as well. Historically, gradients can arise from evolutionary pressures. These, in turn, are instantiated, in many cases, by chemical gradients of hormones that control growth. However, there are also many non-living cases of gradients.
Finally, there are “Intentional Gradients” that are put in by artists, architects, landscape gardeners, designers, and UX designers.
Although the examples above are visual, there are also gradients in music. There are gradients in activity.
Gradients are also present in narrative. Think of the rising passion between two lovers or the increasing tension and suspense in a mystery or horror film scene.
Are there gradients in User Interfaces? One example that springs to mind was in the “Audio Distribution System” and “Olympic Message System.” These were audio messaging systems from IBM that allowed subscribers to the service to leave an audio message to a person or distribution list “on a person’s phone.” (Of course, the message was really on a server and played upon demand.) The UI’s only visual elements were an ordinary touch tone key pad circa 1980 and possibly a small booklet or template with the various command options. Primarily, the user was meant to hear Prompts for actions.
The gradient part came in the fact that the user could speed up the audio more and more as they became more and more familiar with it. That variable speed applied to prompts and also to messages. A typical user would experience slow messages, then faster messages, then, still faster messages, and so on.
In another IBM project, we built a “Personalized Education System” that would allow IT personnel to build a “Custom Course” for on-line learning that could vary continuously in length and difficulty. However, while each individual user could choose length and difficulty, they typically did not experience a gradient in that regard. By analogy, some insects may prefer to chew on new growth while others may like more mature leaves. An individual insect only experiences one “size & tenderness” yet the plant taken as a whole provides a gradient. Similarly, if your fellow citizens ever get there act together enough for you to have a barbecue party again, you might offer up a range of options on how done you cook the burgers. In this case, we might say the BBQ service offers gradations though an individual burger is not a gradient. This isn’t strictly true. A well-done burger is fairly consistent throughout but a rare burger may be much more rare in the middle and charred on the edges. So, for that choice, there is a gradient of experience.
Compared to the natural environment around me, which is positively thick with gradients, I don’t see (or feel) many gradients in any of the UI’s I interact with (games aside). The fisheye lens provides one way of introducing gradient into the UI. It produces, a kind of “Perspective Gradient.”
It would be possible to put gradients into many of the visual elements of a UI. Would it make the interface look more like nature? Would it be more beautiful? Would it be distracting?
——————-
Here’s a poem that illustrates a gradient of love.
Complex organizations typically have many levels of organizations. In life, there are cells which often have symmetry. The human body has large-scale bilateral symmetry (two eyes, two ears, two arms, two legs, two sides to the brain, etc.). If we look at our hands, we see a thumb and four fingers. Each finger, and the thumb, each have roughly bilateral symmetry. Symmetries such as these are visible.
There are also often functional symmetries in living systems as well. We breathe in. We breath out. Our nervous system has excitatory circuits and inhibitory circuits. Most complex organisms reproduce via two sexes. Many animals have cycles of sleep and wakefulness. Often birds, and other animals have migrations tied to the cycle of the seasons. Trees are able to draw water up and move water down.
The structural symmetries and the functional symmetries are often connected. When a person runs, for example, they use one leg and then the other. In fact, as your left leg goes forward, typically so does your right arm. As you move your right leg forward, you move you left arm forward. As you breathe in, your rib cage expands on both sides symmetrically. As you breathe out, your rib cage contracts on both sides symmetrically.
Human organizations often have functional symmetries as well as structural symmetries. For example, most organizations have an “onboarding process” as well as a “termination process.” Our physical artifacts often exhibit local symmetries and these are often related to our physical and behavioral symmetries. A long boat, for instance, allows for multiple rowers to row in unison. The boat and oars are symmetrical and so is the rowing of the boat.
Human artifacts of many scales may exhibit local symmetries for two reasons. First, since most natural living things exhibit local symmetries, copying that may strike us a more beautiful. In addition, if one learns the skills necessary to make the left half of a canoe, one already knows a lot about how to make the right half, provided the canoe has bilateral symmetry. The same is true of an oar, a cane, a bowl, etc. It also makes it easier for the user of the artifact.
As a user of an artifact such a chair, for example, you come to expect that the right arm rest and the left arm rest will be at the same height and be equally hard. If the arm rests are at different heights, you will, I believe, be more likely to bang your elbow when shifting position or reaching for something.
One could, no doubt, adapt fairly quickly to a chair which had two different kinds of arm rests, but imagine a keyboard for your computer in which every key was a different size and shape. Or, imagine a piano keyboard in which all the keys were at different spacings, and sizes.
Local symmetries also offer another advantage. Systems with local symmetries are easier to repair or maintain. Imagine how much more expensive it would be to maintain a piano, for example, if a repair shop had to keep 88 different sizes of keys! In a similar fashion, imagine a programmer decided it would be fun to program every pull-down menu with completely different algorithms. When the next release of the application required changes or additions, it would make understanding and modifying the code much more difficult.
Why does local symmetry, as opposed to merely, symmetry, make sense? Let’s go back to the chair example. It might make sense for a family to have three chairs for three different sized people; e.g., a papa bear chair, a mama bear chair, and a baby bear chair. Or, think of tee-shirts that come in Small, Medium, Large, and Extra Large. It makes no sense to make every tee-shirt the same size. People differ a lot in their size. But for mass manufacturing, it does make sense that the left and right sleeves match for any particular size. Similarly, it doesn’t matter that the arms on papa bear’s chair are of a different size than the arms on baby bear’s chair. It doesn’t matter if the teeth on my comb are different from the teeth on your comb. It doesn’t matter if the tires on my Saab match the tires on your Range Rover. And, generally, this is true for most artifacts, organizational structures, architectural features, and anatomical features. Local symmetry almost always has four advantages: 1) it adds to beauty. 2) it’s easier to create 3) it’s easier for the user and 4) it’s easier to maintain or repair.
Design always carries messages. Design which solves our problem or beautifies our world or resonates and echoes the properties of natural beauty carries the messages: “I care. I am a bit like you. I am making this for you and others like it. It shows the beauty that I see is a lot like the beauty (and usefulness) that you see. You are not alone. We are in this together. You and I both belong to “Team Humanity.”
Good Shape. Positive Space. These seem to be highly related concepts. Most of the pictures I used as illustrations of Positive Space could equally well illustrate Good Shape. A Positive Space is often composed of multiple good shapes arranged in a “balanced way.”
In order to illustrate “Good Shape,” I decided to take some pictures to illustrate “Bad Shape.” Since I have surrounded myself mainly with nature or with artifacts that were intentionally designed to be aesthetically pleasing, it turned out to be more difficult than I imagined at first. The other issue is that when I take photographs, I (like most people) “frame” the picture so that “Good Shape” is clear. I began to intentionally brake that habitual way of framing with pictures of flowers that overlap. The natural beauty of flowers is so rich that, to me, even a quasi-random framing still results in a beautiful picture.
This turns out not to be the case with artifacts however.
Although some of the elements of this seem to have “Good Shape,” the overall composition lacks a coherent center. It seems to me to be an artificial “kludge” of colors, textures, and shapes and there is no overall “Good Shape.”
By re-arranging the elements into this form (above), to me, it seems to get closer to “Good Shape” but the purple “tennis ball” seems ill-proportioned and out of place. In “real tennis” there is a physical reason for the ball to be fuzzy but here, the fuzziness of the ball seems added on and arbitrary.
Here too (above), the overall composition lacks “Good Shape.” The large semi-random dark blob in the middle has no real “Center.” The elements around the edges do not cohere or reflect off of each other. None of the elements around the edges are complete enough to have “Good Shape” except possibly the links of the small chain.
Here is another example of quasi-random elements brought together into a single picture. Though to me, both the wood and the stone are in and of themselves beautiful, there is nothing about the composition that integrates into a whole. There is no center, no echoes, no alternating repetition, no levels of scale, and despite the nice texture of the stone and wood, the overall effect, to me, is lacking in “Natural Order” and in beauty.
By way of contrast, the flower here seems to have a good overall shape reinforced by a Strong Center, Alternating Repetition, and each of the petals themselves has Good Shape. Around the flower, to me, is a “Positive Space.” Note that one of the petals seems slightly out of place. To me, however, this does not spoil the “Good Shape.”
The same can be said of these flowers.
It isn’t necessary to have overall “simplicity” in order to have “Good Shape.” These flowers are not “simple” in shape, but they still have “Good Shape.”
The bird on a branch shown above seems to me to have “Good Shape” even though it is (in outline) lacking symmetry. Though lacking symmetry in profile, the shape strikes me as “balanced.” I don’t perceive the bird as having to exert a lot of energy to keep from “falling off the branch.” The bird’s shape, like the shapes of most animals and plants is the result of 4.5 billion years of evolutionary history.
Many people view some creatures as “enemies” and therefore “creepy” or “yucky” or perhaps even as dangerous. If you can set that aside and simply look though, I think you can see that “Good Shape” characterizes all sorts of animals (as well as plants).
At a more abstract level, it even seems that natural animal behavior often shows beauty. That also seems characteristic of skilled human behavior. The Olympics show plenty of examples. It is hard to imagine an “ugly” baseball swing resulting in a home run. The runners, jumpers, throwers, gymnasts, tennis players, and so on produce “Good Shape” in three spatial dimensions and one temporal dimension.
I play tennis several times a week. I’m not averse to winning a point even if the ball hits my frame, skids off to the net cord and drops over for a winner. But — I never feel as pleased by such outsized luck as when I hit a good shot that results from proper choice, preparation, and execution (and a history of practice and reflection). I never think of a lucky mis-hit as “beautiful.”
The concept of “Good Shape” even applies to more complex human organizations and endeavors. Racism, for instance, to me is truly ugly. It is not based on truth, but on a kludge of lies masquerading as truth. White Supremacy for instance, basically says, “We’re white! That makes us superior! In fact, we’re so superior, that we insist every aspect of our society be tilted so we can be guaranteed a win!”
Huh?
Bad logic is never beautiful.
The “Good Shape” exhibited by most plants and animals reflects both a long evolutionary balance of forces (phylogeny) and the development of a particular organism (ontogeny). When biology goes awry, as in cancer, the result is not beauty. The “logic” of cancer is that billions of years of evolution that allows our bodies to work harmoniously is put aside for short-sighted egotism. A random cell says, in effect, “I’m no longer going to do my part for the body as a whole. Instead, I’m going to grab all I can for myself and grow in an unrestricted way!”
Such a cell, by making that decision, signs its own death warrant. Either the body destroys the cancer and therefore the cancer dies — or, the cancer prevails and in that case it destroys the life of the organism it was supposed to be a part of — and then it dies anyway! It’s an “ugly” decision both because it is disproportionate and because it is illogical. It aims to be self-serving, but it is not even self-serving. It guarantees its own death and kills more life along the way.
It’s the same, to me, with any dysfunctional human organization. If democracy is perverted by some people demanding that decisions be made so that they can have their way regardless of how it impacts the overall tree of life and the overall health of a city, state, or nation, it lacks “Good Shape” and instead becomes a distorted kludge lacking symmetry, lacking balance, lacking life. It is no accident that “Gerrymandered” districts do not have “Good Shape.”
In extreme cases, not surprisingly, dictatorships have resulted in a war on truth, beauty, and life itself. It’s no “accident” that Hitler, Stalin, and Mao were each responsible for millions of human deaths. Their activities also show utter disregard for the beauty of the natural environment and the plants and animals of the planet.
When it comes to User Experience, the most obvious way to apply “Good Shape” is to the visual elements of the interface — icons, windows, menus, tool bars etc. However, I think it also applies to the design process. Are the needs of all stakeholders taken into consideration?
Do the various disciplines involved in design respect each other? Do people tell the truth and show trust? Does the company treat its users with respect and tell the truth in its various communications?
If a company provides “free games” or “free services” that are purportedly for entertainment or education but the real purpose is to gather data to sell to advertisers or foreign powers trying to use the data to destroy democracy, it would seem to me that business model itself — the raison d’êtrefor the entire project is an “ugly shape” regardless of how pretty the icons are.
What do you think? Do the concepts of natural beauty apply to the design of organizations and even governments?
(This is the fifth in a series of 15 blogs that explore Christopher Alexander’s “Fifteen Properties” of good design, natural beauty, living spaces, etc.).
Life is a dance on the edge between chaos, on the one hand, and sameness & repetition on the other hand. Repetition without variation or change is not life. On the other hand, endless random variation without any pattern or principle is also not life. Planet Earth is sometimes said to be a “Goldilocks Planet” — not too hot and not too cold — to support life. Most of the life forms on planet earth themselves exhibit a “positive shape” — a shape that wants to expand and fill out — and yet, there are limits because of circulation, gravity, communication etc. so that growth is typically not unrestricted. The tension between trying to grow and having to stick close enough to form creates positive space.
Life may not always be “a bowl of cherries” but nonetheless, a bowl of cherries does illustrate the property of “positive space.” The bowl connotes living things trying to expand right up to the boundaries possible. A bowl that is filled with cherries and even slightly over-filled but without inducing a fear of spillage, to me, connotes life being vigorous and healthy. To me, a single cherry sitting all alone in a large bowl seems to exhibit positive space much less — at least at the scale of the bowl.
When I look at an individual cherry, I also see positive space. The cherry is not a perfect sphere, featureless and smooth. If it were, to me, it would be a much less positive space. The fact that there is a stem and a noticeable indentation where the stem goes provides a kind of center. So too does the slight asymmetry of the cherry. It appears as though it might have developed as two halves that tried to circle a space at the same time from two different directions – clockwise and counterclockwise. They met and merged to form the almost spherical cherry.
The cherry is not the only natural food that has this kind of positive space. Think of corn on the cob. Each kernel seems to have tried to grow as large as it can. It pushes up against and into other kernels. Each kernel seems therefore to be “bursting with life.”
Consider a cucumber. It’s roughly cylindrical. But “roughly” is important here. To me, it would seem much less an example of “Positive Space” if it were literally a cylinder. It tapers at the ends and it bulges at the middle. That seems more “positive.”
A person who doesn’t understand that life is primarily a cooperative endeavor among all the parts of the vast tree of life, but instead sees the world only as a zero-sum game, it must seem as though the positive living space of a cucumber makes the space around the cucumber less positive and less living. But this is not true! The space around the cucumber’s slightly rotund shape is much more positive and living than the space around a perfect cylinder.
Think of a tree trunk compared with a telephone pole. They are both made of wood and they are both approximately cylindrical. But a tree trunk feels much more alive to us than a telephone pole. I don’t think this is simply because we “know” a tree is alive. It looks more alive and if you allow yourself to “tune in” to it, it feels more alive. And, importantly, not only does the space that the tree inhabits seem more living. So too does the space around it! A long line of telephone poles may exhibit “Levels of Scale” and “Alternating Repetition” and such a picture may make us feel some life to it.
But a single telephone pole does nothing to stir life in me. In that sense, it’s nothing like a tree. It’s flat and it kills the space on either side as well. The columns in the Parthenon are roughly cylindrical and obviously not literally alive. But they, unlike a telephone pole, have positive space and create positive space around them. Life is not fundamentally a zero-sum game. The bee does not “steal” the flowers pollen and thus damage the flower. The bee helps the flower. The flower helps the bee. Positive space in an object does not steal from or detract from the space around the object. It enhances it!
Similarly, the “competition” for space among the kernels of corn on a cob of corn shows that competition can be a beautiful thing. I like to see the kernels pushing against each other. All of them seem living and positive. Imagine instead a cob of corn in which one kernel had grown like a giant spider of orange cancer and took up 98% of the cob. Aside from this one ultra-greedy cell, there were only a few widely spaced and shriveled up kernels. Would you pick that from the pile of corn? Would want to slather it with butter and bite into it? I sure wouldn’t! And to me, that is exactly what cruelty, unfairness, and greed look like. It might seem positive — after all, the dictator is trying to “grow.” That is true but he’s rigged the game. Competition is no longer beautiful when the game is rigged. Competition is no longer beautiful when your tennis opponent slices off your leg before the match starts or pays off the chair umpire to call everything you hit out.
I don’t think there’s anything particularly special about cucumbers or cherries or corn. I just happened to have been washing vegetables before writing this. These radishes also exhibit positive space. And, although their shape is not the same as a cherry or a cucumber, the radishes also enhance the space around them. And, so does the avocado. An avocado that is just past ideal ripeness begins to flatten irregularly and the “pebbles” of the skin also flatten making a less positive space. The pebbles of a perfectly ripe avocado are themselves tiny positive spaces that add to the overall impression of vigor.
Flowers provide numerous examples of positive space. And, so far as I can tell, the positive space of flowers invariably also enhances the positivity of the space around them.
Animals also generally exhibit positive space. Their shapes, after all, come about through a struggle among different tradeoffs that have been “learned” by each species over four and a half billion years. But whether you look at the leg of a turtle, the leg of a turkey, or the leg of a polar bear, you will see positive space. The shapes are quite different because their histories are quite different. But in every case, the shape is determined by the history and each of those histories is the same length — 4.5 billion years. The fact that each of these shapes comes about through the dance of trying to survive and thrive in all their various habitats makes every one of them beautiful in its own right — even those whom most of us don’t particularly like to interact with such as rattlesnakes or lions. Even if we don’t wish to come face to face with a lion or grizzly bear we do recognize their beauty. And, I would claim that as those animals walk through any landscape, they will not only exhibit their own positive shape; they will also enhance the positivity of the shape around where they are walking. Think how many times photographers and painters have put an animal in the foreground of a landscape! Of course, doing so makes the scene immediately more living, more interesting, and more beautiful.
What does this have to do with the user’s experience on a computing device? I would have to say that most interfaces on most devices do not do a lot to bring the interface “alive” including having positive space. The tool bar icons on this version of my Mac are very slightly “bulging” like a living cell (or kernel of corn). Even more subtly, the windows have slightly rounded corners.
It’s an entirely different story with many games though. It seems to me that some of the game artists are among the best artists in the world. They have created many very beautiful games. Could they be more beautiful still if game designers consciously thought about using Christopher Alexander’s “Properties” in general, and “Positive Space” in particular? I think so, but for many games, such as Horizon Zero Dawn, many of the elements are copied so closely from nature that aspects such as “Positive Space” are mostly already taken care of by evolutionary bias toward the beautiful. (Yes, I do think there is such a bias, but it will have to be the subject of a different post.)
What about other applications and systems? Do you think you’ve seen good examples of “Positive Space” in such systems? How could you see incorporating “Positive Space” in UX?
(This is the fourth post in a series of 15 which aim to examine the fifteen properties of natural order postulated by Christopher Alexander. I wonder whether the property of “Alternating Repetition” could be more frequently incorporated into User Experience and what the pros and cons might be).
4. Alternating Repetition
In nature, whether we look at clouds, waves, mountain ranges, forests, plants, the leaves of plants, or almost any animal, we see alternating repetition. Such natural repetition is almost never precise and complete replication. Just as every reproductive act of life introduces an element of recombination as well as the possibility of mutation, so too the repetitions we see in life are not mathematical. We indeed see patterns repeated in a line or in a circle, but the instances in living things show variation. Similarly, human artifacts made with the hands; e.g., weaving or poetry or rock walls or tennis strokes show repetition but with slight and subtle variations.
If you look at modern industrial society, you see many examples of exact repetition. It’s true that if you looked at 10,000 examples of the same brand and model pen, or notebook, or scissors, or streak knives under a microscope, you could likely find small differences. To the naked eye, however, they would appear identical. Seeing 10,000 pebbles, leaves, or waves you would always see slight to moderate differences. To me, it makes a difference in terms of beauty. Handmade items and human actions naturally show more variation than things produced by machines.
The various “elements” of a User Interface generally have the feel of something made by machine. The size of all the application icons on the tool bar for instance are the same. They are all almost the same exact shape and size. On my tool bar the icons for Finder, Launchpad, Mail, Safari, Chrome, FaceTime, Messenger, Maps, Photos, Contacts, Calendar, Reminders, Notes, System Preferences, Keynote, Pages, Preview, and News are the same. However, the Kindle icon is slightly larger and more square. The Zoom icon is very slightly rounder and the icon for Facebook is circular. In other words, there is some slight variation, but mainly the icons are the same in outline, but different inside. But the set of icons on the tool bar has no sense of “harmony.”
Note that the tool bar is both abnormally homogeneous in terms of positioning and size but completely disconnected in terms of the images which differ in a random disharmony seldom exhibited in nature.
Instead, each icon seems to scream out its individuality without regard to its neighbors. (Which reminds me of things like masking, vaccinations, telling the truth, etc.). Anyway, the layout of the toolbar, to me, it’s reminiscent of driving down the highway and seeing a sequence of billboards, each advertising some random product or service in no particular order. Often, each billboard is the same size but they are independent of each other. The one exception I can think of were the “Burma Shave” signs along the highway. Each sign was a small rectangle with several words on it. About a quarter mile down the road, another sign would appear. They were meant to be seen in a specific order and together, they made a kind of “story” which usually rhymed. Typically, the signs touted the joys of the product — “Burma Shave” — or, provided public service announcements about safe driving.
In the early days of computing, putting in “variation” would have been insanely wasteful of space and processing time. This is no longer true. Yet, the “look and feel” of User Interfaces remains, for the most part, quite “mechanical” or “mechanistic” rather than “natural.” Sometimes modern computer games include naturalistic looking variations in how repeated plants, rocks, clouds, etc. are represented. Look at the palm trees and you will see variations among the trees that are reminiscent of the real palm trees I see every day. In addition, each tree contains a number of fronds that repeat a generic theme; yet, each front is somewhat different. The fronds themselves each contain alternating repetition, just as do real palm tree fronds.
We might consider whether it would make sense to put more natural looking alternating repetition into the more utilitarian aspects of user experience. For example, could it be both more beautiful and more useful to allow more variation in the way files are represented visually? Will Hill, James Hollan, Dave Wrobleski, & Tim McCandless suggested (in a CHI ’92 paper) that physical documents such as manuals naturally show wear in places where they are most used. They suggested that the visual representation of computer documents might be altered slightly to show which documents have been used or edited more and also that such cues within a document might also be useful.
Hill, William; Hollan, James; Wrobleski, Dave; McCandless, Tim; “Edit Wear and Read Wear,” Human factors in computing systems: Striking a balance. Proceedings of the 9th annual conference. SIGCHI ’92 Proceedings (Monterey, CA), Addison-Wesley, 1992, pp. 3-9.
One of the few cookbooks I use with some regularity is entitled, Fat Free and Flavor Full. The recipe I’ve used many times is the black bean salad. If I open the physical book, it naturally falls open to that particular recipe. “Open Recent” I find a very useful features of the Pages application, but I don’t see a way to look at files based on overall frequency of use. It could be the case that there are some applications for which you would want specific files to be easily found depending on the season or the date. The authors of “Edit Wear and Read Wear” were mainly talking about the utility of encoding information about a file into its visual representation. There might well be aesthetic reasons to do so as well.
The wooden blinds shown below exhibit alternating repetition in several ways. First, there is the matter of perspective. Some variation almost always presents itself simply by virtue of the fact that I typically view the blinds at an angle and some are closer to me than others. This means that the size, and the retinal shape are slightly different. Second, there is variation caused by slight changes in the angle of the slats relative to vertical. Third, each slat is made of wood. The wood itself shows rings (or more commonly a planar projection of the concentric cylinders.
The wooden blinds, because they are illuminated from the outside by tree-filtered sunlight also show further variation; however I don’t think that is typically “alternating repetition.” Perhaps not coincidentally, although I find it a little interesting, I can’t say I find that source of variation to be beautiful to me. The other three all add to the aesthetics, at least to me.
Is it feasible to introduce alternating repetition with slight variation into User Interfaces? Would it be desirable? What negative side-effects might arise? Do you even agree with Christopher Alexander that Alternating Repetition is an aspect of natural beauty and beautiful design? One thing that occurs to me is that if some element of the UI is more variable in outline etc., it may mean that the actions upon that object must deal with that variability. If designed greenfield, that might not be too difficult. But if, for example, the code for dragging and dropping was written under a presumption of zero variability, that might be problematic. What do you think?
——————-
Discussion of Alternating Repetition with additional visual examples can be found here:
Here are a few thoughts about “Boundaries” and how they apply in User Experience.
I decided to gift a copy of of Volume One on The Nature of Order to my daughter earlier today. I logged on to Amazon and looked at my address book. I am aware that she moved fairly recently. So, I was scrolling through my earlier text conversations with her to see whether she had told me of her new address. I couldn’t find a text about her new address so I texted her to get the new address.
Suddenly, a popup window appeared from SIRI. It had her new address. I hadn’t said anything aloud. I thought of SIRI as a voice-activated service on my iPhone. It was disconcerting to have it “notice” my text message and then suggest an answer (which turned out to be correct).
Last week, after physical therapy, my therapist & I began to discuss the time for my next appointment. I pulled up the calendar application on my iPhone and went to a particular day and began to type in her name. After the first two characters were typed in, the “type-ahead” function suggested three possible “completions” the first of which was the time we had been orally discussing (which was not a common time nor the time of any of my recent appointments with her). It also filled in her complete name and the purpose of the appointment, but that was more understandable.
One sense of “Boundaries” in User Experience connects with a notion of “boundaries” that is much discussed in contemporary mental health. We are advised to “establish boundaries” with co-workers, family, friends, and strangers. We don’t necessarily want to share personal information with everyone or let everyone touch us in any way they choose to. If intimate details are shared in a recovery group or group therapy, it is generally agreed that such details will not be shared with others.
We sometimes extend the idea of informational boundaries to written materials as well. If, for instance, we keep a personal diary, we do not expect other people to “search for it” or to read it. In this story, I relied on the expectation that someone would read a paper I “accidentally” dropped on the sidewalk. But she was so protective of my privacy that she wouldn’t even glance at my paper.
On the other hand, if we write and publish an autobiography, then we can expect that other people will feel justified in discussing the contents. To me, it would seem odd for an author to feel “violated” if people start talking about the contents of their autobiography (or their blog).
When it comes to modern interactions with computer software however, the boundaries are invisible — and sometimes non-existent. It can feel as though I write a private diary on paper; lock it up in a safe immediately; and then — without any sign that the safe has been broken into, I suddenly find details of my personal life revealed!
There appear to be boundaries between applications, and certainly between devices but these boundaries may be illusory. I find that troubling and confusing. I think the first application of “Boundaries” as a property of UX is that apparent boundaries should be real. There may be exceptions for exceptional circumstances; e.g., the police may get a search warrant to search your house if there is reasonable suspicion that you have committed a crime.
When a social media site analyzes your reactions, relationships, and word usage to determine what to try to sell you and what type of approach is most likely to succeed, that does not strike me as a reasonable response to an “emergency.” As most readers know by now, such information is not only used to try to sell you more stuff; it was also used to manipulate public opinion; for example, to convince some US voters to stay away from the polls on election day in 2016; to convince voters in the UK to vote for Brexit; to convince people not to get vaccinated.
Living things do have boundaries. Breaching those boundaries is typically something to be avoided. We call such breaches by names like “bites”, “wounds”, “diseases”, “gunshots”, “parasites.” Living cells typically have a cell wall. Within the cell are tinier organelles such as the mitochondria. The mitochondria have boundaries. The nucleus of a cell has a boundary. Within the nucleus, the nucleolus has a boundary. Larger structures often have boundaries. Motor neurons have a myelin sheath which allows neural impulses to travel faster. Almost our entire body is covered in layers of skin which for a boundary.
The formation of boundaries does not stop with our physical body. Organizations of humans — nuclear families, clans, nation-states, counties, cities, townships, teams, corporations — they have boundaries. A bank, for instance, might have a safe for the money, but the building itself also functions as a boundary — not an impermeable boundary — customers are allowed to come in during banking hours. There are also legal boundaries. If you have “an account” at a bank, you will be allowed to do things that non-customers cannot. Similarly, if you are an employee of a company or a member of a sports team, you will be allowed to do things and go places that you couldn’t if you were outside that boundary.
All boundaries are semi-permeable. Boundaries change over time. A thorn tears your skin. Your boundary is broken. If you’re not careful, bacteria can get in and cause an infection. Your white blood cells destroy the invading bacteria. Your body heals. If the cut was bad enough, you may get a scar and the scar is now part of your “boundary.” It isn’t only at the level of the body that changes occur. Your social boundaries change too.
You get married. You get divorced. You are born. You join a team. You quit the team. You sell your house and other people buy it. Now, you are no longer allowed to come into the house without an invitation. Meanwhile, you buy another house. You have acquired new boundaries. Or, perhaps, you have no home. You are homeless and your boundaries are not so secure.
Most of our possessions have clearly defined boundaries. Your hammer is separate from your saw which is separate from your drill. They come from an earlier time and the “boundary” of such objects are determined by their shape. More recently, such tools (and nearly everything else!) Is packaged in bubble wrap which forms an additional boundary. This makes it harder for people to hide one under their clothes and walk out without paying. Such packaging has the added advantage that it will require time and energy on your part before you can actually start using the tool for its intended purpose. Not only that — such packaging helps pollute our world beyond the pollution required by “old style” tools.
Once you have separated your new tool purchase from its packaging, if you have any energy left, you can saw a board, or drill a hole, or hammer a nail. But you do not expect (not yet, at least), the saw to “communicate” behind the scenes with your drill. Or with you. You’d be surprised if it piped up and said, “Gee, Gene, you just sawed a board. Now, you have taken up the drill. Would you like suggestions on how to build a dog house?” (That’s what Clippy would do).
Clippy tried to be helpful. But it didn’t really have enough information about my tasks, goals, and context to actually be helpful. But today’s behind-the-scenes information sharing with dark forces is not trying to be helpful. It’s trying to get you to change your behavior for someone else’s benefit — and you don’t even know who those someones are.
Notice that if you buy a house (which typically comes with doors and keys), you can lock the house and the default is that it keeps everyone else out — except those you’ve given a key to or those who have rung your doorbell or otherwise asked for permission to come in. Typically, if a rare visitor comes to your house, you make arrangements for a time and a place. The piano tuner comes to tune your piano. You might let them use your bathroom or even offer them something to drink. But you don’t expect the piano tuner to redecorate your study or to spend the night uninvited.
That’s kind of what does happen in the electronic world though. In many cases, you cannot visit a website or use an application unless you give permission for the “guest” to rifle through the choices you make. Just to be clear, these “choices” are not only explicit choices; your “choices” can include how long you linger over a particular message or video clip. In many cases, you have not just given a key to a specific vendor, application, or website — in many cases, you have also given them rights, essentially, to make as many copies of your front door key as they care to make and hand them out to whomever they like.
These are missing boundaries, not so much in the user interface design, but in the socio-technical context in which we use our technology.
In the physical world in which we evolved, invasion of privacy typically involved symmetry. If I can see your eyes, you can see mine. Conversely, if I can’t see anyone, chances are that they can’t see me. Of course, this isn’t literally true. A tiger’s camouflaging stripes may mean that they can see the gazelles even though the gazelles cannot see them. The astounding eyesight of the eagle allows them to see a mouse on the ground and start their deadly dive before the mouse can see the eagle.
In the electronic world, it isn’t genetically coded asymmetries of information that allows other people to invade your boundaries — in many cases without your permission or even knowledge. It is an asymmetry that comes from money and time. You don’t have anything like the fortune that rich companies have. They can hire experts at subverting your boundaries. They can hire an entirely different set of experts to convince you that it’s all okay. They can afford to hire still other experts to defend themselves in a court of law should you seek redress for any particularly unethical behavior. They can afford to hire politicians as well in order to make laws to protect their unfettered access to your data. You typically cannot afford to hire politicians to protect your right to privacy.
You probably don’t have 10,000 to 100,000 people working for you. Companies not only have the money to spy on you. They also have to time to collect and analyze your behavior & make sense of it. You don’t. Perhaps, every once in a while, you take the time to wade through a “privacy policy.” In most cases, since experts were hired to make the text as incomprehensible as possible, you likely didn’t see much value in reading the document.
The Nature of Order is about aesthetics, not ethics. And, this post was meant to be about aesthetics, not ethics.
Life includes differences in sensory capability. And life includes camouflage. Generally, however, when you get to the end a cliff and step off, you have a pretty good idea what’s going to happen. The boundary is visible to you, to a bison, to a mouse, to a lemming, to an eagle.
When we walk through the woods in northeastern USA where I lived for many years, we run the risk of being attacked by a deer tick. The dear tick makes a hole in you and starts sucking your blood (oh, while they’re at it, they may inject a large does of Lyme disease bacteria into your blood stream). You don’t notice it, because the deer tick is “kind enough” to administer a local anesthetic so you don’t feel any pain from this invasion of your person; this breaking of boundaries. It’s a one-sided breach. The deer tick is well aware of the invasion. It’s the whole point! But you do not perceive the breach. At least, I didn’t. Twice. Thankfully, I don’t seem to have any long-lasting effects though I have several friends who do.
A one-sided boundary breach, doesn’t seem “aesthetic” to me. Nor does it seem “truthful.” The little orange deer tick, is, in a very real sense, lying to me. It uses its narcotic to tell me, “No worries! There’s no wound here! There’s no deer tick sucking your blood. There’s no deer tick injecting a serious disease into your blood. No, no. All is well!” It seems the opposite of beauty and the opposite of truth.
I suppose if I had been born a deer tick, I might view things differently.
This is the second posting in a series of fifteen which examine Christopher Alexander’s “Fifteen Properties” of natural beauty and suggests how these properties might apply to user experience design.
2. “Strong centers” is probably one of the most overlooked properties of design in UX/HCI. Often, what exists for the user, from their perspective, is a “sprawl” of functions, tool bars, and icons with no obvious overall or subsidiary organization. A better design would allow the user to quickly find a “home base” from which, it would be obvious where to find subsidiary home bases. There is some sense in which hyperbolic trees, fisheye lenses, and home pages partly begin to address this issue.
Instead of “strong centers”, the impression I often get in looking at applications for word processing, organizing photos, searching, or dealing with settings is that the designers are given or generate a long list of functions to be supported. Which ones are related to which though? Which ones are central? In many cases, UX practitioners give users (or, more often, potential users), a set of cards with one function each and ask the users to sort these into piles. I am not against such studies, but they are unlikely to lead to a coherent design with a strong center. The users are not, in most cases, professional designers. In many cases, an application is supposed to support many different specific actions. For example, I use word processors to write essays, poems, and fiction. I also use a word processor to proofread something, to re-organize ideas, to “jot down” a bunch of ideas, or to write an outline. These are very different tasks, at least to me. If asked sort cards, I would do it differently depending on which type of task and which type of material I’m thinking about.
As I type this, I glance at the “Pages” tool bar which includes: Pages, File, Edit, Insert, Format, Arrange, View, Share, Window, and Help. None of these seems like “home” to the task of writing an essay. I know from experience that if I want to write any kind of material, I must go to the “File” menu even though, as best I can recall, my initial impression of this label was that it would be something to do after I was “done” with the tasks of composing and proofreading. The toolbar gives no impression of their being any “center” at all, let alone a “strong center.”
In my native language (English), I read from top to bottom, left to right. In that sense, the Apple Icon is first and the “Pages” item is next and it is in bold print. That could be considered a subtle clue that it’s the “most important.” In a way, the items on the “Pages” menu are “meta-items.” In that sense, I suppose you could argue that they are “important” — though as a writer, none of the items seem that important. In fact, if we get right down to it, nothing in Pages really seems designed to support the actual writing process. And, I’m not trying to single out Pages because it’s the only one or the worst one. Lack of a “strong center” seems true of nearly all applications.
Different people use different processes for writing — and I myself use different processes for different types of writing, so perhaps trying to organize the features and functions so that there is a “strong center”reflective of the “strong center” of the task of writing is just not feasible. I am certainly not advocating for resurrecting “Clippy.”
Ideally, it should be possible for users to “know where the action is” upon entering an application or a web page.
In another interpretation, “strong centers” refers more to underlying architecture and points to the need for a core of functionality that transcends a specific release or even a specific application. A good underlying architecture will communicate this essential center (related to central purpose or style) to the user.
All too often, the processes involved in developing an application or system themselves have no “strong center.” If the development process is itself a hodgepodge political process of accommodating to a portfolio of features and functions that are advocated for by diverse and uncoordinated stakeholders, then, what one has are a long, unorganized list artificially shoved into menus and sub-menus and toolbars. It should not be surprising then, that the user finds it difficult to know where to begin when first encountering an application — even if the user knows exactly what they want to accomplish.
Compare and contrast most menu structures and user interfaces with the “strong centers” that are extremely common in life forms. Here are some examples of butterflies.
The central axis includes the head, the thorax and the abdomen. These are in a line in the strong center. Typically, they are colored differently from the wings. The bilateral symmetry of the wings as well as the overall shape reinforces the strong center. Wing patterns, and even the antennae and legs lead the eye back to the strong center.
Most people think of butterflies as beautiful — and I agree. When asked, however, most people will say they are “brightly colored.” Some are; some are not. But “bright colors” don’t necessarily give rise to beauty!
Strong Centers doesn’t just apply to butterflies. Look at most birds, fish, mammals, insects and you will see how the symmetries and smaller centers reflect and strengthen the major center. Indeed, Strong Centers are not limited to the animal kingdom. The trunk of a tree provides a strong center. Each branch is itself a center and the connection of the branches to the trunk reinforces the strength of the central trunk.
Even single cells often exhibit strong centers (and many of the other 15 properties, by the way).
It would love to be able to provide you with a process or checklist or formula so you could design user experiences with “strong centers.” I cannot really do that. Nor can anyone else. If you keep it in mind, even at the back of your mind, you may see opportunities to help make that happen with regard to whatever you’re working on. I’m curious to hear your thoughts about “Strong Centers.”
———————
Some useful links to more information, discussion, and examples relevant to “strong centers” or to the fifteen properties.
This is an introduction to a series of blog posts on the “Nature of Beauty.”
Christopher Alexander was an architect and city planner. In his MIT dissertation, Alexander took a very mathematical approach to design. In our studies at IBM Research on the “Psychology of Design” I first ran across that work (Notes on the Synthesis of Form). Later in his career, however, he took a quite different approach to design. With an international team, he visited many different parts of the world to see what “worked” in terms of architecture and city planning. The results were documented in the form of a “Pattern Language.” In this sense, a “Pattern” is the named solution to a recurring problem. A “Pattern Language” is a connected lattice of Patterns that together, “cover” a field.
Others (including me) have emulated his approach for other fields such as pedagogy, organizational change, object-oriented programming, software development processes, and human-computer interaction. A few years ago, I suggested such Patterns for collaboration and teamwork. Here’s a link to the introduction of that effort. Here’s a link to the index of those Patterns.
Still later in life, Christopher Alexander embarked on a project called The Nature of Order. This work is documented in a series of four books. In the first book, he proposes fifteen properties of good form in nature — and in beautiful artifacts. Explication and example will be needed to appreciate what these properties mean. In this post, I list the fifteen and will attempt to explain the first one with respect to UX and Human Computer Interaction.
OLYMPUS DIGITAL CAMERA
First, you might be wondering what relevance these fifteen properties of nature might have to interface design. After all, can’t the designer just test out their ideas empirically until a UX design is shown to be usable, learnable, and perhaps enjoyable as well? Well, sure. Ideally, every possibility could be explored and tested empirically.
But how many possibilities are there? Without any guiding principles, there are not only more possibilities than can be tested by you. There are more possibilities than there are atoms in the universe. Imagine a very simple interface on a small mobile device. Let’s say there are only ten screens in your whole application. The iPhone 12, for instance, has nearly 3 million pixels. 24-bit color allows over 16 million colors per pixel. If you literally tested out every possible arrangement, this would mean 16 million to the 3 millionth power for each of the ten screens!! The number of atoms in the universe is estimated to be between 10**78 and 10**82. Obviously, this is far less than 16,000,000**3,000,000 !!
Of course, I’m not suggesting that anyone would attempt a pixel by pixel test of an interface, but the general point remains: you need some way to limit testing to reasonable alternatives. The notion of using these fifteen properties is not that they dictate a particular design nor that you don’t need to do any empirical testing. Rather, the fifteen properties could be used to help guide design. The properties could be thought of as reducing the search space.
Here are the fifteen properties:
Levels of Scale
Strong Centers
Boundaries
Alternating Repetition
Positive Space
Good Shape
Local Symmetries
Deep Interlock and Ambiguity
Contrast
Gradients
Roughness
Echoes
The Void
Simplicity and Inner Calm
Not-separateness
Perhaps the names themselves might resonate with your own sense of aesthetics for design and composition, but let’s review them one by one.
When it comes to natural beauty, a few moments reflection may provide you with many examples. Christopher Alexander claims this property is also present in traditional art and architecture across many cultures. As you see something such as, e.g., the Taj Mahal or the Parthenon in the distance, you see a beautiful shape. As you approach it, you will see more and more levels of scale. By contrast, many modern buildings are largely featureless between the overall shape and the texture of the building material.
If your design has multiple levels of scale, it will be easier for your user to orient themselves; to know “where they are” in the application and therefore easier to take appropriate action. It’s something to keep in mind with respect to your design — whether hardware, software, documentation, or a building.
How could you use or see “Levels of Scale” as a desirable property of what you are doing right now?
———————————-
Some references to Pattern Languages in HCI.
Pan, Y., Roedl, D., Blevis, E. and Thomas, J. (2012), Re-conceptualizing Fashion in Sustainable HCI. Designing Interactive Systems conference. New Castle, UK, June 2012.
Thomas, J. C. (2012). Patterns for emergent global intelligence. In Creativity and Rationale: Enhancing Human Experience By Design J. Carroll (Ed.), New York: Springer.
Thomas, J. (2012). Edging Toward Sustainability. CHI Workshop Position Paper for Simple Sustainable Living. CHI 2012, Austin, Texas.
Thomas, J. (2012), Enhancing Collective Intelligence by Enhancing Social Roles and Diversity. CSCW Workshop Position Paper for Collective Intelligence and Community Discourse and Action. CSCW 2012, Bellvue, WA.
Thomas, J. (2011), Toward a pattern language for socializing technology for seniors. Workshop position paper accepted for CSCW 2011 workshop: Socializing technology among seniors in China, Hangzhou, China, March 19-23.
Thomas, J. (2011). Toward a Socio-Technical Pattern Language for Social Systems in China and the World. Workshop position paper accepted for CSCW 2011 workshop: Designing social and collaborative systems for China. Hangzhou, China, March 19-23.
Thomas, J. (2011). Toward a Socio-Technical Pattern Language for Social Media and International Development. Workshop position paper accepted for CSCW 2011 workshop: Social media for development, Hangzhou, China, March 19-23.
Bonanni, L., Busse, D. Thomas, J., Bevis, E., Turpeinen, M. & Jardin, N. (2011). Visible, actionalble, sustainable: Sustainable interactin design in professional domains. Workshop accepted for CHI 2011. Vancouver, B.C., May 7-12.
Thomas, J. (2011). Focus on Ego as Universe and Everyday Sustainability. Workshop position paper accepted for CHI 2011 workshop: Everyday practice and sustainable HCI: Understanding and learning from cultures of (un)sustainability. Vancouver, B.C., May 7-12.
Thomas, J. C. (2018), Building common ground in a wildly webbed world: a pattern language approach. Journal of Information, Communication and Ethics in Society, 16 (3), 338-350.
I worked at IBM, all told, about 28 years. During that time, management put more and more pressure on us to make our work “relevant” to the business. In fact, the pressure was always there, even from the beginning. Over the years, however, we were “encouraged” to shorten evermore the time gap between doing the research and having the results of that research impact the bottom line. This was not an IBM-only phenomenon.
I was a researcher, not a politician, but it seemed to me that at the same time researchers in industrial labs were put under pressure to produce results that could be seen in terms of share price (and therefore payouts to executives in terms of stock options), academia was also experiencing more and more pressure to publish more studies more quickly — and to make sure “intellectual property” was protected to make sure the university could monetize your work. This was about the same time that, at least in America, increasing productivity and the wealth that sprung from that increased productivity stopped being shared with the workers.
In the late 1970’s, the “Behavioral Sciences” group began to study the “psychology of design.” For the first few months, this was an extremely pleasurable & productive group, due mainly to my colleagues. Over the next few blogs, I’ll focus on some specific techniques and methods that you may find useful in your own work.
In this short story though, I want to focus instead on some broader issues relevant to “technology transfer”, “leadership” and “management.” Even if you are or aspire to be an expert in UX or HCI or design, I assure you that these broader issues will impact you, your work and your career. I wouldn’t suggest becoming obsessed with them, but being aware of their potential impact could help you in your own work and career.
It is telling that, almost invariably, whenever I told someone inside IBM (or, for that matter, outside IBM) that I was studying the “psychology of design,” people responded by asking, “the design of what?” So, I would explain that we were interested in the generic processes of design and how to improve them. I would explain that we were interested in understanding, predicting, and controlling these processes to enable them to be more effective. I would explain that we could apply these findings to any kind of design: software design, hardware design, organizational design, and (see last post about IBM) communication design. I would explain that design was a quintessentially human activity. I would also explain that design was an incredibly leveraged activity to improve.
Looking back on it, I still think all these things are true. I also see that I missed the “signal” people were giving me that, while I thought of design as something that could be studied as a process, that most people did not think of it that way. To them, it was never the “psychology of design,” but only the design of something.
Don’t get me wrong. I agree that somewhat different skills are involved in designing a great advertising campaign, a great building, and a great application. I agree that different communities of practice treat various common issues differently. I still think it’s worth studying commonalities across domains. For one thing, we may find an excellent way of generating ideas, say, that the advertising community of practice uses that neither architects nor applications developers had ever tried. Or, vice versa.
My own academic background was in “Experimental Psychology.” We were forever doing experiments that we believed were about psychological processes that were thought to be invariant regardless of the domain. It was an axiom of our whole enterprise that studying memory for any one thing shed light on how we remember every other thing. Similar studies looked at decision making or problem solving or multi-tasking. We came to understand that there were some interesting exceptions to being able to separate content from process. For instance, it is much easier to multi-task a spatial task and a verbal task than it is to multi-task two independent spatial tasks or two independent verbal tasks.
We used a spectrum of techniques to study “design” from laboratory studies of toy problems, to observing people doing real-world design problems while thinking aloud. After about 3-4 months of very productive work, we were told that we had to make our work relevant to software development. That should be the focus of our work. We were told that this command came from higher-ups in IBM. That might have been true, or perhaps partly true.
It might also be relevant that someone in our management chain might have been the recipient of a grant from ONR which was specifically focused on software development. So far as I can tell, nothing had been done on that grant. So, our past, present, and future work could have been co-opted to be “results” done under the auspices of the ONR grant.
In any case, regardless of the “reasons,” the group began to focus specifically on software design. In one study, we used IBM software experts as subjects. Each person was given information that was geared toward a specific transformation that occurred in software development. One person was presented with the description of a “situation” that included a number of “issues” and they were asked to write a requirements document. In real life, I would hope that this would be done in a dialogue (and, indeed, in other studies, we recorded such dialogues). Absent such dialogues, what we found was that different software experts — all from IBM research — and all given the same documentation about a set of problems generated vastly different problem statements and overall approaches.
In other parts of the study, other experts were variously given requirements documents and asked to do an overall, high level system design, or given a high level design and asked to design an algorithm, or given an algorithm and asked to code a section. There was always diversity but the initial showed the greatest diversity. The initial stage is also the one that can cause the most expensive errors. If you begin with a faulty set of requirements — a misreading about how to even go about the problem — then, the overall project is almost certain to incur schedule slip, cost overruns, or outright failure.
While the vital importance of the initial stages of design is true in software development, I would argue that it is likely also true for advertising campaigns, building designs — and even true for the design of research programs. We designed our research agenda under the assumption that we had a long time; that we were studying design processes independently of specific communities of practice or the nature of the problems people were attempting to address. We assumed that there was no “hidden agenda.” Although we believed we would eventually need to show some relevance to IBM business, we had no idea, when we began, that only relevance to software design would be “counted.”
—————-
Some of our studies on the “Psychology of Design.”
Carroll, J. and Thomas, J.C. (1982). Metaphor and the cognitive representation of computer systems. IEEE Transactions on Man, Systems, and Cybernetics., SMC-12 (2), pp. 107-116.
Thomas, J.C. and Carroll, J. (1981). Human factors in communication. IBM Systems Journal, 20 (2), pp. 237-263.
Thomas, J.C. (1980). The computer as an active communication medium. Invited paper, Association for Computational Linguistics, Philadelphia, June 1980. Proceedings of the 18th Annual Meeting of the Association for Computational Linguistics., pp. 83-86.
Malhotra, A., Thomas, J.C. and Miller, L. (1980). Cognitive processes in design. International Journal of Man-Machine Studies, 12, pp. 119-140.
Carroll, J., Thomas, J.C. and Malhotra, A. (1980). Presentation and representation in design problem solving. British Journal of Psychology/,71 (1), pp. 143-155.
Carroll, J., Thomas, J.C. and Malhotra, A. (1979). A clinical-experimental analysis of design problem solving. Design Studies, 1 (2), pp. 84-92.
Thomas, J.C. (1978). A design-interpretation analysis of natural English. International Journal of Man-Machine Studies, 10, pp. 651-668.
Thomas, J.C. and Carroll, J. (1978). The psychological study of design. Design Studies,1 (1), pp. 5-11.
Miller, L.A. and Thomas, J.C. (1977). Behavioral issues in the use of interactive systems: Part I. General issues. International Journal of Man-Machine Studies, 9 (5), pp. 509-536.
——————————
Blog posts about the importance of solving the “right” problem.
As I mentioned recently, when I first arrived at IBM Research in the early 1970’s I began to work on Query By Example and other schemes to make it easier for non-programmers to interact productively but flexibly with computers. Some of the lessons learned have to do, not with my own work, but with the work of others.
At that time, some researchers labelled their work as “The Psychology of Programming.” There were many debates — and some studies — about structures, syntax, which language was better than others, etc. There were also lengthy discussions about the process that one should use for software development. Doing any kind of “controlled” experiment on large scale code development is prohibitively expensive. It is rare that a company is willing to have two independent teams build the same piece of software in order to learn which of two methods is “better.”
Of course, one such comparison would likely not prove much. It may be that one of the two teams had a “super-programmer” or an extremely good manager. Perhaps, flu broke out in one of the two teams. You would really need to study many more than two teams to properly and empirically study the impact of language, or syntax, or process. Doing reasonable-sized experiments would be far too costly and impractical. Our lab and others did do laboratory tasks in order to test one syntax variant against another and so on. The problems were generally quite small in order for the study to be practical. So, the applicability to real-world development projects was questionable.
Walston & Felix (See below), on the other hand, were able to find data on a fair number of real-world projects and rather than try to control for languages, processes, etc., they did a multiple regression analysis based on what languages, methods, etc. real projects used and what the important predictors were of actual productivity.
Personally, I learned two lessons from their study.
Lessons Learned: #1 — Sometimes, when it comes to what matters in the real world, controlled laboratory experiments have to give way to other methods such as studying “natural experiments.” Despite the many issues with trying to interpret such findings, no-one will pay for massive controlled experiments that parametrically vary programming methods, programming languages, etc. while controlling for quality of management, experience, complexity of task, etc. Multiple regression studies and in-depth case studies; ethnographic studies; interviews: all of these can provide useful input.
OLYMPUS DIGITAL CAMERA
Lessons Learned: #2 — The impact of the variables that our community of people were looking at in terms of syntax, structure, etc. were dwarfed by the impact of organizational variables. For example, Walston & Felix found that the complexity of the interface between the developers and the customers was extremely important.
DeMarco & Lister (See below) claim that, based on their decades of experience as consultants to the software development process, projects almost never fail for technical reasons; when they do fail, it’s almost always for organizational and management reasons.
That conclusion dovetails with my experience. Many decades later, working for IBM in “knowledge management,” it was amazing how many companies wanted us to “solve” their knowledge management issues by building them a “system” for knowledge sharing.
But…
Management at the company would not provide:
Incentives to share knowledge
Space to share knowledge
Time to share knowledge
Or, commit any personnel to gathering, vetting, organizing, and promoting the knowledge repository.
So — knowledge sharing was something they were simply supposed to do on top of everything else they were doing.
They did not want a computer system, IMHO; they wanted a magic system.
These experiences were part of my motivation for attempting to catalog “best practices” in collaboration and teamwork in the form of a Pattern Language. Christopher Alexander and his colleagues looked at “what worked” in various parts of the world when it came to architecture and city planning. What they did for architecture and city planning, I want to do for collaboration.
Naturally, merely creating a catalog is not sufficient. I need to have people who will read it, understand it, modify and improve it, and then promulgate it via actual use. For now, it’s free. Comments and critiques are always welcome.
—————-
C. E. Walston and C. P. Felix, “A method of Programming Measurement and Estimation,” IBM Systems Journal, vol. 16, no. 1, pp. 54–73, 1977.
petersironwood 