A Note on Terminology
design, language, usability
December 26, 2003, 12:31 AM
As I was writing my post yesterday, I wanted to add a note on my usage of a few key terms here on this weblog, to clarify any possible confusions that might arise.
When I use the term design in reference to software systems, I am sometimes referring to interface design, i.e. the form and function of the portion of the software system that the user of the product sees and interacts with, and sometimes referring to internal design, the structure and specifications of the logical components that power the behavior of this interface. Although I recognize that this is generally a useful distinction to make, I tend to view these two disciplines as different aspects of the same activity. There is no question that the two influence one another. Because these two concepts are blurred in my mind, the word I use for them is fairly blurry as well. For the most part this is intentional.
Secondly, I tend to use the term usability rather more broadly than others do (or so I've found). It's convention in the field to make a distinction between "useful, usable, and desirable" as quality characteristics of software systems (and many other systems, as well). Useful refers to whether the software does what the user needs it to, usable refers to whether he can learn and remember how to use the software, and use it efficiently, and desirable refers to whether he wants to use the software. I often find this distinction silly, since there's so much overlap between these three concepts, and thus I tend to lump them all under the moniker "usable", i.e., to what extent can people use it? I've already argued that all software quality attributes reduce to usability. Now I feel like I need to argue that all usability attributes reduce to usability (which strikes me as a bit silly).
Finally, I have no good word to refer to the group of people who are members of the broad profession that aims to design products and systems for use by human beings. The general term for this discipline is often "human (or user) centered design" but the term "human-centered designer" often isn't seen as applying to user researchers, UI programmers, etc. who aren't necessarily designing an interface. But no other word will do. "Usability person" tends to get rejected by the designers. There are a plethora of more specific terms (interaction designer, information architect, interaction architect, UI designer, interface designer, user researcher, usability specialist, "gooey" (from Yahoo), user tester, usability engineer, usability professional, or generate your own) My friend Dana made up the word "usable" but nobody seems too fond of that either. So I might use any sufficiently general-sounding term to refer to the whole lot of them, until one emerges that everyone can agree on.
I don't believe that any of these issues are idiosyncratic to me, but rather reveal some inherent ambiguities in the way we talk about these things. However, this post should be taken as more of a clarification than a complaint. Ambiguity in terminology is confusing, but inevitable. Human language is riddled with ambiguity, which is part of what makes it such a beautiful and powerful thing. Some of the ambiguities I've pointed out above are (in my view) necessary, and I have faith that the others will sort themselves out in time.
Competing Definitions of Design
design, language, philosophy
November 18, 2003, 07:54 PM
Dan has a post up about Dick Buchanan's definition of design:
"Design is the human power to conceive, plan, and realize products that serve human beings in the accomplishment of any individual or collective purpose."
Just for fun, let's compare and contrast this with my own stab at defining design:
"a deliberate action prompted by an understanding of the state of the world intended to transform the world into an improved state."
First off, Dick's definition explicitly mentions humans, whereas mine leaves the actors undefined. I think for all practical intents and purposes this is an unimportant distinction.
Secondly, Dick's incorporates a three-stage process into the definition; to design you must "conceive, plan, and realize". Mine doesn't discuss any of the activities that make up design. Perhaps it should.
Next, his specifically mentions products as the output of design. I prefer leaving this undefined, for some design processes have as their output a refined policy or organizational structure, or perhaps a new process for performing some task. These things aren't products unless you define "product" much more broadly than most people do.
Next, my definition calls out "understanding the world" as an essential component of design, whereas Dick's does not. Perhaps this is too prescriptive, since lots of design gets done without any formal attempt to understand the context in which it must fit. I'd argue that designers still require some (possibly incorrect) understanding to move forward with, which often may be based solely on their own assumptions.
Finally, my definition specifies that design attempts to improve the world (in the mind of the designers, anyway), whereas Dick's gives as the aim of design "accomplishment of any individual or collective purpose".
Dick is certainly correct that definitions cannot provide closure on all the important philosophical issues in the field. But it's still interesting to hear what words people choose to define the terms they use everyday. It often unveils subtle, but important, differences in people's understanding of the meaning behind those terms.
Posted by James Spahr on November 18, 2003 at 08:43 PM
My favorite definition (just because of the double meaning and the beauty of it's simplicity) :
Design is a Good Idea
Posted by Rob on November 19, 2003 at 02:05 PM
Sounds like a good definition, in line with the "lively" poetic definitions Dan mentions in his post.
Dick's and my definitions were more philosophical, more precise, but also more boring :). Each is suited for different purposes, really.
Posted by haven on November 24, 2003 at 04:32 PM
When Dick uses the word "product" he is talking about it in the broader sense of the word. He's referring to products, as many of us over in the ID program often do, as artifacts (often confused with product), services, systems, environments, etc. The best way to think about it is product is that which is produced, and it does not necessarily need to be a tangible thing.
Perhaps that clears up that matter a little. As for the rest, we'll save that for a chat over a beer.
Posted by Rob on November 24, 2003 at 04:53 PM
Point conceded. I could quibble with the practice of redefining a standard term to mean something different that it's popular definition, since that invites confusion. But if I did so I'd start down a path of devoting my life to arguing with half the philosophers, engineers, and other skilled professionals that have developed a substantial amount of jargon in their field. Some things you just gotta accept!
Posted by Smriti Gupta on March 12, 2007 at 12:20 AM
To me, the best definition of design still says, "design is a problem solving
excercise. The intricasies like identofications of problems comes engraved in this.
Posted by Kenneth Lynn on September 04, 2007 at 02:57 PM
Design – "the conscious selection and deployment of resources to achieve a specified objective" – is the key activity of the human race. Everything, including survival, depends on the quality of our design ability. We are designing our future. The design of the built environment (Architecture)is of particular importance because of its inescapable effect on our lives. “We shape our buildings, thereafter they shape us.” (Churchill). Through the activity and appreciation of design we increase out understanding of our environment and of our place and potential within it. Vitruvius, writing 2000 years ago, identified three criteria essential in Architecture but applicable to all design - usefulness, sustainability and beauty. When we understand the significance of these three criteria we are in a position to make an objective judgement on the quality of design.
Email Rob:
Deos It Mttaer Waht Oredr the Ltteers in a Wrod Are?
language, psychology
October 13, 2003, 11:21 AM
Matt Davis, an actual researcher at Cambridge University, has an interesting page discussing the word scrambling and perception net-meme that was going around last month. The bottom line is that some of the claims in the email are at least partially true, but the central thesis (that we only read the first and last letters in a word) is false. Check out the page for the details, but note that it's designed very badly and you have to scroll past all the word-scrambling-in-multiple-languages stuff to get to the actual analysis.
Found via Andy.
Email Rob:
The Way We Really Speak
language, society & sociology
October 09, 2003, 11:06 PM
As part of our background research for CSCW, Dana and I are reading some social science articles on how people build common ground. "Common ground" is a term that refers to the customs, norms, and other background knowledge and experience that people in a group share that allows them to communicate efficiently (or communicate at all, for that matter). Common ground includes social norms as broad as "we will all speak English" to ones as narrow as "'that thing' refers to the sprocket wrench we were discussing earlier in our conversation". We feel that building common ground is an especially important area for us, since our particular problem, integrating usables into open-source software development, involves two groups that have a big problem with a lack of common ground.
Right now, I'm reading an article by Andrew Monk entitled "Common ground in electronically mediated communication: Clark's theory of language use". Monk's main point is that the ways we communicate with each other are much more socially-determined than you might think. To back up this claim, he describes Herbert Clark's theory of language use.
The classic model of human communication is very send/receive oriented. It goes something like this:
This model has been very useful, especially to communication engineers since it pretty much maps directly to the way two computers talk to one another. But the model breaks down when you examine how people really talk. The basic problem is that the model doesn't take into account the fact that the "code" people use to communicate isn't initially well-defined. With two computers communicating over a network, the protocols they use (such as TCP/IP, HTTP, etc) to exchange information have been precisely pre-defined by the programmers (or ginormous standards institutions). With humans, however, there are no such rigorously defined protocols. Sure, human languages are fairly well-defined, but the ways that people actually use these languages to communicate is not. The way people really speak is a process of negotating common ground. Monk gives an example in the paper:
Roger: Did you have oil in it
Al: Yeah, I-I mean I changed the oil, put new oil filters, r- completely redid the
oil system, had to put new gaskets on the oil pan to stop-stop the leak, and
then I put -and then-
Roger: That was a gas leak
Al: It was an oil leak buddy
Roger: Its a gas leak
Al: It's an oil leak!
Roger: on the number one jug
Al: It's an oil leak!
Roger: Outta where, the pan?
Al: Yeah
Roger: Oh you put a new gasket on it stopped leaking
Al: Uh huh
If you read over that conversation snippet (which really occurred "in the wild") and reflect on it, you'll notice that much of the meaning is being negotated rather than transmitted. The nutshell version is that conversation is much more like ballroom dancing than it is like sending and receiving computer messages. It's a process of speaking, receiving and processing feedback, clarifying, receiving more feedback, etc. In some sense, it's an iterative design process with really tight feedback loops.
Here's a picture of our revised understanding of language:
Again, what's really happening here is a development and negotation of common ground for the conversation. Unlike with computers, with humans the "protocol" is developed on the fly.
Monk also notes three types of common ground developed in conversations:
- Conversational conventions which are social norms that necessarily underlie all conversations (without them, we could not converse). E.g, "We will be as concise as possible" and "We will let the other person know when we don't understand something they said".
- Communal common ground are norms that spring from the surrounding culture and environment, such as "We will both speak English". These conventions relate most directly to online communities.
- Personal common ground is more specific to the particular conversation. These are norms developed between the two conversants previously or during the conversation, e.g. "We are going to lunch together tomorrow" or "The word 'Nish' refers to Newell-Simon Hall, the building on the southeast of campus with the green roof".
In essence, this theory is an interesting merge of classic cognitive, psycholinguistic approaches to understanding language and more sociological approaches, and appears to describe the ways we really use language much better than previous theories I'd learned while taking linguistics classes in undergrad.
More on common ground may be coming up soon (dunno if that's a teaser or a threat ;).
Posted by Rob on October 10, 2003 at 05:58 PM
It might be worth noting that Clark's theory extends the communication model of language rather than replacing it. You can think of all the arrows in the second diagram as all following the encoding-decoding process described in the first diagram.
Not sure I made that clear in the post...
Email Rob:
Bill Joy and Language as Architecture
language, software development
October 04, 2003, 03:25 PM
I'm reading an article on Fortune's web site interviewing Bill Joy about his recent departure from Sun Microsystems. He made an interesting comment while discussing Windows' architecture:
Also, Windows isn't well architected. There's a simple way to find out if an operating system has been well designed. When you get an error message, go to the help system and look up the exact words in that message to see if there was enough of a concept of an architecture that they have a consistent vocabulary to talk about what's broken.All you have to do is try it on a Mac and on a PC to see the difference. Apple took the time to come up with a concise vocabulary, but in Windows the designers of the help system used different terminology from the programmers. That reflects a lack of design discipline, which means that as the system grows, so does the ambiguity of the software itself. The result is a system encrusted with multiple layers of things that weren't really designed in so much as bolted on.
Whatever you may think of Windows, Joy's comments ring similar to my thoughts on architecture as a shared language as well as XP's focus on metaphors instead of architecture. The bottom line is, architecture is more than just boxes and arrows; it's even more than just interacting components. Architecture is the language the development team uses to communicate, and, as Joy points out, that language better be closely aligned to the language the rest of the organization and the external users of the software use to communicate as well. Certain organizations that claim expertise in this area (but shall remain nameless) could stand to learn this lesson a little better.
Email Rob:
Power and Language in Interaction Design
language, usability
August 04, 2003, 10:59 PM
I found an Ask Tog column today via Dan's weblog proposing "Interaction Architect" as a new name... er... "brand" for interface designers as well as the development of a corresponding professional association to elevate this discipline to a more respected status in the computing industry. I always thought interaction designers / user experience professionals / whatever they want to be called fit rather nicely with the user testing people under the umbrella of the Usability Professionals' Association, but no matter. Tog's proposal is interesting in its own right, but it lead me down an entirely separate avenue of thought.
In the past, I've argued that terms like "architect" speak towards similar patterns that appear in many disciplines. Tog's post, however, provides some indirect contradictory evidence for this assertion; he wants to use the term "architect" primarily because it's a "power word", i.e. its a term that is associated with a certain amount of respect in the political climate of software development teams. I could see someone arguing that this implies there is no such common meaning; terms like "architect" just get co-opted by certain groups of people to further their own political purposes. (To be clear, I'm not claiming Tog is taking this position.)
There is some merit to this argument; language gets used for many purposes by us naked apes. One of those purposes is, in fact, political posturing; politicians have to master this method of using language if they want to be effective at their jobs (just read some Machiavelli to see what I mean). But this argument doesn't tell the whole story. Tog uses the words "architect" and "designer" not only because they further his cause, but because they make sense in the context. Tog explains why he sees "Interaction Architects" as leading the development of software systems in a similar fashion to the way building architects lead the construction of houses. If there weren't underlying commonalities in the meanings of the terms, no one would be willing to accept "interaction architect" as a moniker. After all, Tog didn't choose "Interaction Executive" as his name of choice, even though "executive" is certainly a powerful title. "Interaction Executive" is laughable in this context precisely because it makes no sense.
Words are powerful to humans, and get put to a wide variety of ends. But ultimately they must follow the rules and norms of human languages; they must be grounded in common concepts and must make sense to those who hear them.
Posted by Lyle, Lyle, Croc O' Lyle on August 06, 2003 at 02:41 PM
Hi Rob,
I responded to Tog in "An Open Letter to Tog"
http://crocolyle.blogspot.com/.
In short, I think he's got it all wrong.
I also point to some related resources and articles on titles in the user experience area.
Lyle
User Experience Architect
Posted by Rob on August 08, 2003 at 09:32 PM
Hi Lyle,
I didn't bring this up in my post, since I wanted to focus on other things, but I'd tend to agree with your main premise: why do we need yet another professional organization for people in the interface design trade? Why can't "interaction architects" just join UPA?
However, I was talking to a guy who has reviewed for UPA (the conference) in the past, and he said it tends to be pretty focused on usability testing. If your paper isn't testing-related, then its a good chance it won't get in. So at least in its current instantiation, UPA might actually be too "tester-oriented" for our friends in interaction design.
I believe this just points to a need for change in UPA, however. If interaction designers infiltrated UPA instead of forming their own organization, then more funds and energy would get funnelled into one place and UPA could be a stronger advocate for the usability trade as a whole, for testers, designers, and those renaissance men and women who can do both. I'd much rather see our field unified behind the banner of human-centered design than splintered into subcultures that never communicate even though they share so much in common.
Email Rob:
Corporations are Plural Nouns
language, society & sociology
May 15, 2003, 11:30 PM
From reading K5, I've noticed an interesting difference between Standard American English (SAE) and Standard British English (SBE) that I believe runs deeper than a mere syntactic curiousity. It's this: in SAE, the names of corporations are considered singular nouns. In SBE, they are considered plural nouns. So whereas an American would say "Microsoft has come out with a new version of Office." a Brit would say "Microsoft have come out with a new version of Office." Upon reflection, I'm of the strong opinion that the Brits have this one right.
To Americans, a corporation is usually viewed as a singular entity with singular goals, desires, and motives. Moreover, it's frequently conceptualized as an entity "separate" from the people who make it up, even when it comes to unethical behavior. Here's a classic example: say I know that a car I own has bad brakes and is unsafe to drive. Yet I sell you this car anyway without telling you about the bad brakes. On the way home from your purchase you get into an accident as a result of brake failure and die. Due to my negligence, I can and should be held criminally accountable for your death, right? Yet in the 1970s this exact scenario played out with Ford Motor Company; due to a design flaw their initial Pinto line of cars were prone to burst into flames during an accident. A Center for Auto Safety article reports:
In 1977, Mark Dowie of Mother Jones Magazine, using documents in the Center files, published an article reporting the dangers of the fuel tank design, and cited internal Ford Motor Company documents that proved that Ford knew of the weakness in the fuel tank before the vehicle was placed on the market but that a cost/benefit study was done which suggested that it would be "cheaper" for Ford to pay liability for burn deaths and injuries rather than modify the fuel tank to prevent the fires in the first place. Dowie showed that Ford owned a patent on a better designed gas tank at that time, but that cost and styling considerations ruled out any changes in the gas tank design of the Pinto.
Ford was sued for millions and eventually issued a recall. But no legal action was brought against the executives who made the decision. "Ford the corporation" was the culprit, said society, not them.
This, of course, is a myth. Corporations don't really exist; they are just convenient abstractions for dealing with groups of people who come together for a common business purpose. Everyone knows this intellectually, but they talk, act, and reason as if the abstraction were a real entity, even in situations as dire as manslaughter due to gross negligence.
Although I'm not claiming that our use of corporation names as singular nouns is wholly responsible for this fallacy, I think it contributes. The way we talk and the way we think are inseparably intertwined, some have even gone so far as to suggest they are the same thing. Singular corporate names hide the fact that corporations are just a whole bunch of flawed human beings with conflicting goals and values; a fact that is important not to forget. So to help see the people behind the "corporation" abstraction, I'm making a conscious effort to incorporate plural corporate names into my common discourse, for I'm convinced America have the worse of the two conventions on this one.
Email Rob:
Defining "Design"
design, language, philosophy
April 29, 2003, 11:53 PM
I was ruminating on the nature of design last night: what design is, what it means to design something, what the place of design in the world is, and so forth. I've especially been pondering the question of to what extent I'm a designer; on the one hand I've always felt that design was a very "artsy" notion that was mysterious to a not-very-artsy person such as myself. On the other hand, I've been called a software designer for awhile, and I've always had a suspicion that "design" ran through much more of what I do than just that; perhaps through almost everything I do. So I dreamed up the following definition that seems to work well for me, and I thought I'd share.
Design is: "a deliberate action prompted by an understanding of the state of the world intended to transform the world into an improved state."
Allow me to now dissect this definition in excruciatingly painful pedantic detail.
"a deliberate action" - This phrase says two things. First off, design to me is a verb, not a noun; it is something we do. Although you can also speak of "a design", this strikes me as a derivative sense of the term. "A design" is just something that has been designed; it has no meaning except in its relation to the verb sense since two arbitrary things that are called "designs" may have nothing in common except that they were both designed. Thus I consider design proper to be an act, not a thing. Secondly, this phrase says design must be intentional; only conscious beings can design and they cannot design by accident.
"prompted by an understanding of the state of the world" - This phrase also says two things. By saying "prompted" I'm implying that design has a source, you don't decide to design "out of the blue". And that source, according to the rest of the phrase, involves some conception of the way the world works. Thus, all design is performed in the context of some kind of framework, the "understanding" you have about the world. This doesn't imply, of course, that your understanding is correct; lots of design gets done based on completely wrong understandings. So in a sense, the main point here is that you will start your design with initial preconceptions, the only question is how accurate these preconceptions are.
"intended to transform the world" - Design is an inherently normative action. By designing you are changing the world, moving it in a direction it otherwise would not have gone. Thus you are going against the current, and, as "intended" implies, you may not succeed in the ways you thought. Whenever you try to transform the world, you're taking a gamble that the world won't outsmart you.
"into an improved state." - Improved state according to whom? Well, the designer of course. This final phrase inserts something tricky into our definition since different people will probably have different notions of what an "improved state of the world" is. Yet rational designers will act to bring about their own conception of "improved", which is worth keeping in mind if you're hiring one to help design something for you. In a sense there is no designing for someone else; the designer ultimately works only for herself and her visions of improving the world.
Currently, I think this definition neatly captures all the factors of the "design" concept that I find critical. But I admit I haven't read much by people who've thought about this at a much deeper level than I have. Comments and suggestions are very welcome as always.
Email Rob:
This Is the Title of the Entry, Which Also Appears In the Entry Itself
funny, language
April 03, 2003, 04:34 PM
This is what happens when you let mathematicians write literature, Lewis Carroll notwithstanding. If I ever have kids, I'm going to read this to them as a bedtime story. I still find it fall-down funny, but those of you out there who are less nerdy than I might just shake your heads and say "Rob!" in an exasperated, what-am-I-gonna-do-with-him sort of tone. But that's ok, because I'm getting a kick out of envisioning you all doing it :-D.
This is the title of the entry, which also appears in the entry itself :).
Email Rob:
Levels of Knowing
language, teaching & learning
March 30, 2003, 01:36 PM
I was talking to my mom today about her knowledge of genetics and how she feels she has a deeper understanding of the subject than the average person but not nearly as much as scientists who understand the various theories and how the math fits together. This got me thinking on how we humans have different levels of understanding about a given subject or body of knowledge.
I hypothesize four levels of understanding:
At level 0, you have no knowledge of the subject at all. In the case of genetics, you either aren't aware that we possess DNA that is partially responsible for determining our physical and mental makeups or you may have heard this truth but know nothing about how it works or how it applies to real life.
At level 1, you have learned a few factoids and propositions that are accepted in the field of knowledge in question. You probably have some sense of how these propositions relate to reality. In the case of genetics, you probably know about DNA, are aware that it passes from parents to child, and may have some concept of how dominant and recessive genes combine to determine traits.
At level 2, you understand many of the major idioms in the field and are capable of acquiring and interpreting new knowledge on your own through books, articles, and discussion. You are able to articulate your own theories on the subject and argue intelligently for them. In genetics, this would be the level of understanding possessed by someone who had taken several courses or obtained a degree in biology or who had read widely enough to gain a similar level of knowledge.
At level 3, you have acquired a rule-based framework of knowledge about the field. You are able to apply this framework to new facts and propositions and analyze them with respect to the rest of the knowledge you've acquired that makes up your framework. You are capable of understanding the difference between propositions that are trivially false and those that are supported by past data and the difference between proposed facts that fit the framework and are therefore likely true and those that do not and thus require more evidence. In the case of genetics, this would be the level of understanding of a researcher who works in this field.
This approach to understanding "understanding" looks very linguistic to me, even somewhat Wittgensteinian. I've been reading an article applying Wittgenstein to software design so maybe that's partially what prompted this (more on that later, perhaps). Note how similar these levels of understanding are to the levels you need to go through to acquire a new language (i.e. level 1: understanding a set of phrases, level 2: obtaining a richer vocabulary and sense of grammer, and level 3: full rule-based system of knowledge).
I wish I knew more about how people learn and internalize knowledge. If anyone knows of any related material, please do pass it on.
Posted by Andy Carrillo on October 24, 2006 at 10:03 PM
A huge turning point in my life was when I realized that I didn't know squat! That's when the real learning begins...
Good post!
Andy
Email Rob:
Email Rob: