Creativity, Communication, and Teaching
Apr. 2nd, 2012 01:08 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
infryqFollowing Olin faculty on Twitter is fun because you get access to a slow but manageable trickle of interesting undergrads, one of whom is @browley20, who wrote the following:
http://brettrowley.blogspot.com/2012/03/importance-of-language.html
... which is awesome, and which I will proceed to agree with and rebut simultaneously, eventually fading into a bit of a rant about approaching creativity from an analytical mindset.
Some primers:
One of the things that sticks in my mind about Olin -- a particular way we were trying to respond to industry and build better engineers -- was the emphasis on skilled communication, particularly oral. We were told (and this is changing, now, as other schools catch up) that most engineers don't get much presentation experience. This has pretty much borne out, if the abysmal presentations I've sat through from non-Oliners is any evidence. We Olin students, on the other hand, were giving at least one project presentation per class per semester, plus innumerable less-formal speaking tasks (CORe, CCO, HB, endless committees, FWOP, ExpressO, Expo, peer teaching, ...). So it strikes me that either Olin's been doing less in terms of speaking tasks as it matures, or perhaps restricting the range (5 minutes is better than nothing, but long-form speaking is a completely different beast), or perhaps they're doing just as much as they always were and browley just wants more because he's got higher standards. That's allowed. Particularly given Olin's spirit of continual improvement.
One of the other things that sticks about Olin is how well it actually mirrored the utter lack of specificity in both problem specs and feedback that can be found in the world post-college. It would, indeed, be great if clients would bother to tell you exactly what they wanted, or the relative importance of different features and requirements, in the way that (some) professors lay out beautifully constructed and communicated rubrics for course assignments. A client's goal is not to make your life easy; a client's goal is to get what they want, with as little of their own effort as possible, and with a chasm of ignorance about what expert knowledge is necessary to define the problem they've posed and give it the proper scope. Generally, if they knew what they wanted, they wouldn't need you.
So I argue this: Olin has several conflicting goals here. To develop good technical skills and confidence, students need well-communicated instructions. To make students part of the "whatever happened to good communication skills" solution and not part of the problem, students need to somehow acquire good communication skills of their own. To operate competently in a real-world environment, students need to learn to translate poorly-communicated instructions into usable information.
The last one is hard. There is no universal mapping from Vague Handwaveytown to the Land of Perfect Clarity. Users don't know what they want. Clients don't know what they want. There isn't a right answer. There are infinite Good answers, and equally many Not As Good answers; it all depends on whether you're thinking critically, can explain yourself, and believe in what you're doing.
It's really hard to teach students to think critically, to have enough background self-awareness to explain their actions and decisions, and to believe in what they're doing, when they're checking ticky boxes on a well-thought-out rubric.
Some topics can operate in both modes. Many engineering projects can be specced to within an inch of their lives, and have solutions that can be weighed against one another sensically, but still have room for an open-ended approach. Matchstick bridges. Rubber-band powered model airplanes. Electro-chemical-mechanical Rube Goldberg devices.
Some topics only really operate in rubric mode. Anything with one right answer that doesn't care how you got there. It's a very simple rubric.
Some topics only really operate in Handwaveytown. Most of design is like this. Which means if you're expecting well-thought-out rubrics in UOCD, then UOCD is a nightmare.
One thing I think Olin (and the entire design community) can do better is to prepare its students for this necessary uncertainty. Nobody can tell you exactly what would be bolder for a particular project, but there are creativity exercises that stretch your mind into the very corners and edges of things and can give you some semblance of practice and familiarity with the feeling. They feel utterly ridiculous until they become as familiar as the analytical tools already in an engineer's toolbox. It's hard to gain that level of familiarity in a single semester. For my master's coursework I took a close duplicate of UOCD (football-shaped creative process, team projects). I followed it with an independent study that first aimed to "fix" the course and the teaching methods, but turned into a general study of teaching and of creativity and how traditional teaching methods and research can be applied and transformed to be effective in creative disciplines. The course resulted in a general model of teaching for creativity, and a conference paper. The ethnography I did that summer -- after UOCD at Olin, and after my master's course, and steeped in all of this creative process research -- was the first time stuff started to sink in. I had a chance to fail fast, to take risks with interviews, and I did. Some of them were fun wastes of time and some of them were dreadful slogs yielding crucial results and some were both fun and useful and they definitely didn't happen in that order. But I'm pretty sure I could never have done interviews like that back in UOCD. It's like Taylor series*. It doesn't stick the first time.
* (Insert your Ultimately Vanquished Nemesis Repeating Topic here)
I think my point (that I've somewhat lost) is that the fuzziness of UOCD, and creative process coursework everywhere, is significantly less about communication problems than one might think coming from traditional analytically-focused coursework. Creative communication is different. The tools you use to write a patent are fundamentally different from the tools you use to bring your team into a superposition of user empathy and inspiration. The communication problems that *are* present in design coursework tend to be in the opposite direction -- nailing down a requirement that should be optional; explicitly expressing a concept that, if left unexpressed, would result in greater creativity, perhaps including a crucial variation that would otherwise have been ignored. Too much specification can lock you in to local minima.
The challenge in bringing creative coursework to analytical minds is in motivating students. Students are accustomed to being motivated by instrumental value: they need to pass the course to graduate, so they will do what is necessary to pass (or get whatever grade to maintain their GPA). So long as the knowledge you want students to attain is what you're measuring in order to give them a grade, this works. So long as the knowledge can be measured, this works. So long as students understand what is being measured, this works. And so long as students believe that it is physically possible to do something that will be measured as success, this works; and so long as students believe that it is within their personal power to do so, this works. In this way, most traditional coursework uses grading so that a student's objective expectation of success and their understanding of their own efficacy leverages instrumental value to create motivation [citation available].
In creative disciplines, this technique breaks down because of the difficulty of measuring creative knowledge. In my research we came to the conclusion that the way out is to discard instrumental value (get the grade!) in favor of intrinsic value: pride in quality of work. There can be no objective measure of the quality of a design, particularly at the high levels that Olin students and MHCI students are capable of. How can anyone judge whether you understand your users "enough," or whether more contact time would be worthwhile? The analytical approach would simply declare some minimum number of interviews, personae, scenarios, and competitor analyses and call it done, but these numbers -- beyond perhaps being nonzero -- have little correlation with understanding. It is much simpler if students have a desire to connect with their users, to create a quality product, to exhaust the space and determine for themselves the proper scope. With intrinsic value it doesn't matter what is being measured, or how, or whether it's possible, so the fact that these factors are poorly defined turns out to be unimportant. The hard part is getting people to believe that they can make their own decisions, without making those decisions resistant to feedback. Then "enough" can be described and defended by the students themselves, while still being revised over time. In a way, this speaks exactly to browley's "whatever happened to communication" point, since students' success depends primarily on their ability to convince others they know what they're talking about. Use a little rhetoric. I promise you, professors are susceptible.
But back up for a moment: giving students an intrinsic value of the material is not easy. It is so much simpler for busy students to fall back on "what do I need to do to pass". You can play games with quantity -- given two ceramics classes, the first graded solely on a final project, the second graded on the total weight of work created over the semester, the latter will generate a higher-quality piece for the end-of-semester showcase [citation available]. You can play games with nontraditional rubrics -- "You must generate at least 1 fantastic failure" -- or nontraditional tasks -- "Research a local bank and come up with a detailed plan to steal something from it" -- but then you run aground of students' belief that success is possible. Most of the designers I have spoken with stand by improv games and brainstorming exercises. These feel like hazing when you first do them, but feel like fun once the embarrassment wears off. I suspect this is a big part of why UOCD feels dumb or frustrating. You don't know what you're doing, the activities seem to have no point, you don't know what you're supposed to be getting out of them. As far as I've been able to find out, designers don't know either. They just know that they help. In some nebulous way, they help. When we do a silly sentence-completing improv game for 5 minutes to warm up, the brainstorm that was the point of the meeting is easier to get rolling, and creates a greater variety of results.
Creative people have to push the boundaries of society's comfort zone. If we knew what it felt like, someone would've been there before, and it wouldn't be outside our comfort zone. There's no way to tell someone how to get there, aside from having them try over and over again, each time a little harder, each time in a different direction. What's hard enough? You tell me. I don't know. You're the first person who will ever have been there.
It's like trying to argue with a Zen monk. Sometimes you have to just say "Yes" to everything with joy and enthusiasm and see what happens.
http://brettrowley.blogspot.com/2012/03/importance-of-language.html
... which is awesome, and which I will proceed to agree with and rebut simultaneously, eventually fading into a bit of a rant about approaching creativity from an analytical mindset.
Some primers:
- UOCD is User-Oriented Collaborative Design. It started as Olin students' first contact with the creative process (needfinding, prototyping, evaluation, iteration), is team-based, and uses a combination of design reviews and journaling to effect evaluation.
- Olin is a tiny, tiny college (68 were in my graduating class) that only offers engineering degrees. It strives for curriculum innovation and to ground students in entrepreneurship, humanities, philanthropy, and design as well as technical topics. Graduates tend towards project management, startups, or graduate school: anywhere requiring interdisciplinary prowess.
- Early on, Olin students had a huge hand in the design and running of the school. This has tapered off since the inaugural class graduated in 2006, but since our 5-yr reunion last year there's been some rekindling of the urge to college-build among current students.
- Since Olin I have dropped out of a robotics PhD program, worked for a few different labs at CMU, and am slowly working through CMU's MHCI (Human-Computer Interaction) program part-time, where I've been focusing on design, creativity, and education.
One of the things that sticks in my mind about Olin -- a particular way we were trying to respond to industry and build better engineers -- was the emphasis on skilled communication, particularly oral. We were told (and this is changing, now, as other schools catch up) that most engineers don't get much presentation experience. This has pretty much borne out, if the abysmal presentations I've sat through from non-Oliners is any evidence. We Olin students, on the other hand, were giving at least one project presentation per class per semester, plus innumerable less-formal speaking tasks (CORe, CCO, HB, endless committees, FWOP, ExpressO, Expo, peer teaching, ...). So it strikes me that either Olin's been doing less in terms of speaking tasks as it matures, or perhaps restricting the range (5 minutes is better than nothing, but long-form speaking is a completely different beast), or perhaps they're doing just as much as they always were and browley just wants more because he's got higher standards. That's allowed. Particularly given Olin's spirit of continual improvement.
One of the other things that sticks about Olin is how well it actually mirrored the utter lack of specificity in both problem specs and feedback that can be found in the world post-college. It would, indeed, be great if clients would bother to tell you exactly what they wanted, or the relative importance of different features and requirements, in the way that (some) professors lay out beautifully constructed and communicated rubrics for course assignments. A client's goal is not to make your life easy; a client's goal is to get what they want, with as little of their own effort as possible, and with a chasm of ignorance about what expert knowledge is necessary to define the problem they've posed and give it the proper scope. Generally, if they knew what they wanted, they wouldn't need you.
So I argue this: Olin has several conflicting goals here. To develop good technical skills and confidence, students need well-communicated instructions. To make students part of the "whatever happened to good communication skills" solution and not part of the problem, students need to somehow acquire good communication skills of their own. To operate competently in a real-world environment, students need to learn to translate poorly-communicated instructions into usable information.
The last one is hard. There is no universal mapping from Vague Handwaveytown to the Land of Perfect Clarity. Users don't know what they want. Clients don't know what they want. There isn't a right answer. There are infinite Good answers, and equally many Not As Good answers; it all depends on whether you're thinking critically, can explain yourself, and believe in what you're doing.
It's really hard to teach students to think critically, to have enough background self-awareness to explain their actions and decisions, and to believe in what they're doing, when they're checking ticky boxes on a well-thought-out rubric.
Some topics can operate in both modes. Many engineering projects can be specced to within an inch of their lives, and have solutions that can be weighed against one another sensically, but still have room for an open-ended approach. Matchstick bridges. Rubber-band powered model airplanes. Electro-chemical-mechanical Rube Goldberg devices.
Some topics only really operate in rubric mode. Anything with one right answer that doesn't care how you got there. It's a very simple rubric.
Some topics only really operate in Handwaveytown. Most of design is like this. Which means if you're expecting well-thought-out rubrics in UOCD, then UOCD is a nightmare.
One thing I think Olin (and the entire design community) can do better is to prepare its students for this necessary uncertainty. Nobody can tell you exactly what would be bolder for a particular project, but there are creativity exercises that stretch your mind into the very corners and edges of things and can give you some semblance of practice and familiarity with the feeling. They feel utterly ridiculous until they become as familiar as the analytical tools already in an engineer's toolbox. It's hard to gain that level of familiarity in a single semester. For my master's coursework I took a close duplicate of UOCD (football-shaped creative process, team projects). I followed it with an independent study that first aimed to "fix" the course and the teaching methods, but turned into a general study of teaching and of creativity and how traditional teaching methods and research can be applied and transformed to be effective in creative disciplines. The course resulted in a general model of teaching for creativity, and a conference paper. The ethnography I did that summer -- after UOCD at Olin, and after my master's course, and steeped in all of this creative process research -- was the first time stuff started to sink in. I had a chance to fail fast, to take risks with interviews, and I did. Some of them were fun wastes of time and some of them were dreadful slogs yielding crucial results and some were both fun and useful and they definitely didn't happen in that order. But I'm pretty sure I could never have done interviews like that back in UOCD. It's like Taylor series*. It doesn't stick the first time.
* (Insert your Ultimately Vanquished Nemesis Repeating Topic here)
I think my point (that I've somewhat lost) is that the fuzziness of UOCD, and creative process coursework everywhere, is significantly less about communication problems than one might think coming from traditional analytically-focused coursework. Creative communication is different. The tools you use to write a patent are fundamentally different from the tools you use to bring your team into a superposition of user empathy and inspiration. The communication problems that *are* present in design coursework tend to be in the opposite direction -- nailing down a requirement that should be optional; explicitly expressing a concept that, if left unexpressed, would result in greater creativity, perhaps including a crucial variation that would otherwise have been ignored. Too much specification can lock you in to local minima.
The challenge in bringing creative coursework to analytical minds is in motivating students. Students are accustomed to being motivated by instrumental value: they need to pass the course to graduate, so they will do what is necessary to pass (or get whatever grade to maintain their GPA). So long as the knowledge you want students to attain is what you're measuring in order to give them a grade, this works. So long as the knowledge can be measured, this works. So long as students understand what is being measured, this works. And so long as students believe that it is physically possible to do something that will be measured as success, this works; and so long as students believe that it is within their personal power to do so, this works. In this way, most traditional coursework uses grading so that a student's objective expectation of success and their understanding of their own efficacy leverages instrumental value to create motivation [citation available].
In creative disciplines, this technique breaks down because of the difficulty of measuring creative knowledge. In my research we came to the conclusion that the way out is to discard instrumental value (get the grade!) in favor of intrinsic value: pride in quality of work. There can be no objective measure of the quality of a design, particularly at the high levels that Olin students and MHCI students are capable of. How can anyone judge whether you understand your users "enough," or whether more contact time would be worthwhile? The analytical approach would simply declare some minimum number of interviews, personae, scenarios, and competitor analyses and call it done, but these numbers -- beyond perhaps being nonzero -- have little correlation with understanding. It is much simpler if students have a desire to connect with their users, to create a quality product, to exhaust the space and determine for themselves the proper scope. With intrinsic value it doesn't matter what is being measured, or how, or whether it's possible, so the fact that these factors are poorly defined turns out to be unimportant. The hard part is getting people to believe that they can make their own decisions, without making those decisions resistant to feedback. Then "enough" can be described and defended by the students themselves, while still being revised over time. In a way, this speaks exactly to browley's "whatever happened to communication" point, since students' success depends primarily on their ability to convince others they know what they're talking about. Use a little rhetoric. I promise you, professors are susceptible.
But back up for a moment: giving students an intrinsic value of the material is not easy. It is so much simpler for busy students to fall back on "what do I need to do to pass". You can play games with quantity -- given two ceramics classes, the first graded solely on a final project, the second graded on the total weight of work created over the semester, the latter will generate a higher-quality piece for the end-of-semester showcase [citation available]. You can play games with nontraditional rubrics -- "You must generate at least 1 fantastic failure" -- or nontraditional tasks -- "Research a local bank and come up with a detailed plan to steal something from it" -- but then you run aground of students' belief that success is possible. Most of the designers I have spoken with stand by improv games and brainstorming exercises. These feel like hazing when you first do them, but feel like fun once the embarrassment wears off. I suspect this is a big part of why UOCD feels dumb or frustrating. You don't know what you're doing, the activities seem to have no point, you don't know what you're supposed to be getting out of them. As far as I've been able to find out, designers don't know either. They just know that they help. In some nebulous way, they help. When we do a silly sentence-completing improv game for 5 minutes to warm up, the brainstorm that was the point of the meeting is easier to get rolling, and creates a greater variety of results.
Creative people have to push the boundaries of society's comfort zone. If we knew what it felt like, someone would've been there before, and it wouldn't be outside our comfort zone. There's no way to tell someone how to get there, aside from having them try over and over again, each time a little harder, each time in a different direction. What's hard enough? You tell me. I don't know. You're the first person who will ever have been there.
It's like trying to argue with a Zen monk. Sometimes you have to just say "Yes" to everything with joy and enthusiasm and see what happens.
