Design II: Tools

From CS260Wiki
Jump to: navigation, search

Lecture Slides


Extra Materials

Discussant's Slides and Materials


Reading Responses

Airi Lampinen - 10/17/2010 20:37:58

Schneiderman's "Creativity support tools" is a brief discussion of what kinds of tools are needed to support creativity. A framework of activities for creative work is established, including eight tasks: searching, visualizing, consulting, thinking, exploring, composing, reviewing and disseminating. On a higher level these relate to collecting, relating, creating and donating.

The authors suggests that we need similar patterns of actions as cut-copy-paste and open-save-close for creative purposes, notably support for "protocols" such as annotate-consult-revise and collect-explore-visualize. The overarching claim is that getting such basic tools at place will facilitate creativity and speed up innovation processes.

While it is easy to believe that *when effective* such tools could indeed be desirable, I was delighted that Schneiderman uses a significant chunk of his five-pager for discussing the potential problems and unwanted outcomes of pushing the development of creativity support tools. While there is nothing wrong with innovation and creativity, they are perhaps not necessarily something every knowledge worker using ICTs should be concerned about in their daily activity?

The second piece, Klemmer, Newman, Farrell, Bilezkjian & Landay's "The designers' outpost: a tangible interface for collaborative web site design" presents a user study of a tangible user interface that combines the affordances of paper and large physical workspaces with the advantages of electronic media to support information design. The authors studied the system in the use of web architects and visual designers, finding that the current implementation was appreciated by the former but somewhat questioned by the latter.

The studied system is another interesting example of mashing physical and digital together to reap the benefits of both but it is also yet another example that doing so is not an easy thing to achieve. Introducing such systems is might be difficult as next to learning how to use a system (which generally is not too difficult) the users would need to change their current social and professional practices to accommodate the new tool. This is always a big challenge and maybe an especially big one when trying to combine two spheres, the digital and the physical, that we traditionally are used to think of as separate and maybe even contradictory. On the other hand, the promise of benefit is clear: better archiving, better portability, enhanced possibilities to modify and update drafts as the work proceeds. It will be interesting to see how long it takes until one of these systems makes it way to the mainstream.

Charlie Hsu - 10/19/2010 13:03:43

Creativity Support Tools

This paper aims to describe the creative process and explore better ways of creating support tools to aid creativity. Shneiderman condenses the creative process down to four activities: collection, relation, creation, and donation. Innovators collect knowledge first, consult with other innovators, create their own work, and then disseminate results. He then cites integration of these activities as one avenue of improving creativity. He also cites eight tasks commonly used in innovation, and suggests that software designers attempt to address these tasks in supporting creative work.

It was interesting to see the creative process broken up into activities and checklists. On the surface, it doesn't seem to be that groundbreaking of an analysis, but seeing each concrete task may inspire more ideas by narrowing the scope of "supporting creative work", a grand and vague idea. By focusing solely on "visualizing", or "searching", or "simulating", creative support tool designers can tackle a much more well-defined and solvable problem.

Integration is another key component of enhancing the creative process. Dating back all the way to the idea of information trails in the memex and realized in hypertext, creating assocations allows for a huge increase in creative flow by creating an environment where logical thought trails and exploration in the mind can be followed (Shneiderman calls this the "smooth coordination across windows"). Compatible data formats are certainly a great idea on paper, but may be difficult to implement. How do you encode the musical note data of a complex, layered metal song? How many bands actually score down all their songs? Compatible data formats may force new methods of creating data itself, going against the grain of the existing creative process.


The Designers Outpost

This paper described The Designer's Outpost, a tangible user interface for collaborative web site information design. The Designer's Outpost combines the physical functionality of a whiteboard and Post-It notes, two common tools for website information architects, and the electronic functionality of a touch screen, computer vision, and digitized data. By adding Post-It notes and freeform ink drawings, the system can save each entity digitally, and allow the user to create links between entities, erase/remove them, and capture the entire state. The authors then conducted an usability study with 15 professional designers, learning that the environment needs to be calm (no more automatic grouping, context menus), but that there is utility in enhancing the whiteboard/Post-It model with digital capabilities, especially in capturing board state.

I found that the idea of enhancing the existing physical world with digital capabilities, like the DigitalDesk reading before it, is a powerful one and holds many opportunities. There are so many applications simply in text and writing tools where digital enhancement could prove useful; e-books, annotation of any type, office and design environments, etc. However, as seen by the usability study results in this paper, designers need to be careful not to interfere with existing thick practice. In this paper, it was the distractions caused by automatic grouping and menus that interfered with the users, interrupting the normal design flow. Even seemingly small design decisions, such as the color of the board background, show the importance of avoiding introduction of negative side effects due to digital enhancement.

One high-level observation I discovered is the importance of versioning and state capture in so many physical applications. The user feedback in the paper seemed to suggest that state capture was possibly the most useful digital enhancement offered by The Designer's Outpost. Allocating a scribe simply to document changes as they occur seems wasteful, but is often done anyways, a testament to how important documentation of the creative process is. Secretaries and minute-takers for meetings in the corporate world run a similar task. Since computers offer such powerful capabilities for storing, duplicating, and accessing data, looking into new ways to harvest and mine for such data seems always to be a rewardable effort.

Luke Segars - 10/19/2010 13:53:25

Creativity Support Tools

This paper discusses the types of tools that could be used to assist knowledge workers in creative work. Schneiderman outlines four activities that creative knowledge workers undergo to do their work, and also lists eight tasks that make up the activities. These tasks and activities are designed to be generally applicable across disciplines, not relevant to any particular topic. At the end of this paper he confronts skepticism that states that technology cannot aid in the creative process.

I like a lot of the author's reasoning. His four activities are both general enough to apply to many applications and specific enough to be able to design for. His four stage model seems accurate, but most of the components already have a number of effective tools in place. Collection, relating, and donating information and knowledge are somewhat well-supported by new innovations like online communities, email, and the world wide web. There is certainly room to improve in these three areas, but they do not seem to be the most challenging part of the process. The "create" component, on the other hand, seems to be the one that is somewhat resistant to technological innovation. Most examples of creativity-enhancing technologies that exist today fall outside of the domain of the "create" task, which is still very often done using very basic tools like whiteboards, simple prototypes, and so on.

I really want to believe that there is a class of tools that can assist with the "create" component of creative thinking, but I'm not convinced yet. At least now, this seems to be something that requires soft knowledge and experience moreso than advanced tools. Transferring soft knowledge is something we haven't figured out how to support with technology particularly well, and that limitation may impede our progress towards coming up with "creation" aids that could finish the loop of accelerating creativity to a 21st century pace.

Kurtis Heimerl - 10/19/2010 14:47:39

Creativity support tools. I’m mostly guessing but I think this paper described how the author thinks design happens, and then described technologies to aid each part of that process.

I’m really confused by this paper. I attempted to read most paragraphs twice, but I never really grokked the thing. The skeptics corner is a particularly good example. Do we really need to argue with people who think technology can(and should) play no role in creativity? Even in 2002, it played a large, tangible role. The negative uses of technology in a 5 piece “how to build for creativity” paper?

I have no issue with anything said. Most of the technologies described are of meaningful value. I just fail to see the point of these disparate thoughts. My conclusion, after reading all of this, is “what?”.

The designers' outpost: a tangible interface for collaborative web site design This paper describes outpost, a tangible interface for supporting web development.

The problem with older papers is that you can evaluate them on their impact in a much more feasible way. For instance, when reading this, my question is, “are systems like this used?”. I think the answer is no. It probably hasn’t taken off for business reasons, and that’s fine. Still, it argues against the “build it, put 5 design teams in front of it for 2 hours each, write a paper” model of research. Maybe I’m just a mean-spirited systems guy. Who knows?

The system itself is totally reasonable, web developers do a lot of whiteboard based, post-it style development and discussion. This automates some of that and allows for them to keep a consistent, stable state. That seems like a win, though what I wanted to know was “how big a win?”. That was not described, because it’s really hard to measure objectively. Oh well.

Thejo Kote - 10/19/2010 15:15:37

The Designer's Outpost:

In this paper, Klemmer and co-authors describe Outpost, a tangible user interface which allows a designer to take advantage of the physical affordances provided by paper, while at the same time enhancing the experience with electronic features. It was targeted at the activity of website design and the authors share their experience in building and evaluating the concept.

Outpost focuses on enhancing the advantages of the whiteboard based design process, like the large amount of available space, ability to paste notes in groups etc. But, a purely physical process makes it difficult to transfer ideas into the final destination of the design - a computer, with all the benefits of the information available on the board. It is also not possible to collaborate with remote individuals in a purely physical setup. Outpost augments this design approach with a set of interaction techniques that allow the designers working together to electronically add notes, create links and move notes while maintaining the electronic links among the physical notes. It also allows the transfer of these designs to the DENIM system on a computer for further work. The experiments conducted by the authors showed that it works well for information architects who work at a more abstract level when compared to visual designers.

Tangible user interfaces which bridge physical and electronic worlds are always interesting, but I suspect the reason none of them have really taken off is because of the very narrow focus of each application. Or perhaps, the value is lesser when compared to a lower cost solution of photographing the whiteboard and then performing the heavy lifting through computer vision aided by the designer on a computer.

Creativity support tools:

In this paper, Shneiderman discusses the characteristics of tools that support human creativity. He focuses on the nature of such tools and not any specific implementation.

He proposes a framework with the following activities - collect, relate, create and donate, where each one is associated with one step of the process of creating something, though not necessarily in a linear fashion. He argues that creativity and problem solving are not lonely activities as they are generally made out to be, but a very consultative and social process which his framework incorporates. Shneiderman also provides a number of examples of how a task which follows his framework might make it easier for people to be more creative. He also proposes eight specific tasks like searching, visualization, thinking and exploring that are involved in completing any creative task in his higher level framework.

This was an interesting mental exercise, but that's about it. I thought that the process that Shneiderman suggests had traces of the same arguments for how the semantic web can unleash knowledge and creativity. At least, most of his examples are related to the same concept. Some of the tools and tasks he envisions are already available, and I think with time, the tools supporting creativity will only improve.

Dan Lynch - 10/19/2010 15:44:01

Creativity Support Tools

This article was great---I really like the idea of writing software that enables creativity (I would like to think that’s exactly what I do). The first thing the author points out is that the target user is way too broad and that the tasks are ambiguous and hence a more horizontal, or generalized method should be taken into account. The framework laid out was based on Collect, Relate, Create, Donate, terms that basically encompass the collaborative and creative aspects of creation.

The main challenge the author claims is the integration of word processors, graphics, email, music, etc. It seems as if the author is dreaming of an all-encompassing monolithic operating system-program. I think the main problem is that people will always write software for distinct entities, although some may collaborate. This will lead to discretized windows on a users machine and hence disparate actions between programs.

However, perhaps this could be created if you use the signals/slots analogy from Qt and apply an open protocol between all programs. This could be one implementation, while allowing separated development.

The Designers Outpost

This research studied the benefit of using a tangible user interface for designing web pages. The study was done on fifteen web designers and architects. The results were that the designers could have done with out it, but the architects loved it.

The implementation itself is interesting because they can use paper and pens to interact with pseudo-digital content, similar to the digital desk. In fact, much of their design was similar to the digital desk in that there is a camera system set-up to do object detection/recognition.

I definitely think this topic is extremely relevant, although their implementation may not be. What I am most intrigued by is the idea of developing digital content through tangible means. I think this is the next wave of the future beyond the mouse and keyboard.

Luke Segars - 10/19/2010 17:23:17

The Designer's Outpost

This paper discusses a new tool for helping web site information architects collaborate. The tool, called "The Designers' Outpost," is tested with a group of professional web developers to determine if it improves their workflow without getting in the way, and the developers tend to generally appreciate the tool after a couple of overly-distracting features are removed.

The field of tangible workspaces seems to be relatively impotent in terms of tools accepted by a wider audience, and very few products have succeeded in the market that blends the benefits of the virtual and physical worlds. The Designers' Output is interesting in that it is actually supports a relatively minimalistic set of features, but focuses on optimizing the user experience of the ones it does support. The device is essentially a tool for keeping track of relationships between pieces of information, and allowing this information to be stored digitally and recalled. The group mentioned a couple of additional features (such as automatic grouping) that were removed because their users didn't find them to be helpful. An interesting followup to their study would have been to investigate whether any of the web developers missed their prototype once it was taken away, or whether the benefits were somewhat inconsequential anyway.

The tool that this group designed was specifically targeted at information architects, but I wonder if it could be applied to a broader class of problems (those that are centered around graph-based problems at the core). I could see this being useful for doing mindmapping w/ Post-it notes as well. Post-its serve as effective physical nodes in a graph, and I wouldn't be surprised if this tool was equally useful outside of the specific realm that it's being tested in. It would be interesting to see if other user types found additional features, such as the automatic grouping feature, would be useful. If so, the "graph building" tool could be developed as a customizable (and extensible) framework that could apply to a number of different classes of problems. It's interesting to think about, anyway.

Krishna - 10/19/2010 17:33:18

Creativity Support Tools

The paper is about designing software that aids creativity. The "collect, relate, create and donate" framework seems to be a little too simplistic for me. An important, obvious factor this framework misses is "inspiration" - the author briefly mentions this in his conclusion. Inspiration may be taken for granted while designing software for artists, but it becomes the most important factor to focus on while developing such software for businesses with the aim of promoting creativity and innovation among employees.

The eight tasks suggested by the author are intuitive, relevant and offer a good checklist for developers of such software. However, there are infinitely many ways of accomplishing them. For example, there is no one generic way of visualizing information; search results, and thus the creative output, can vary based on query strategies; consulting peers is obviously a good idea but the problem is which peer to consult when there are many options. Much of the other criteria such as relating, thinking, reviewing are heavily dependent on the expertise level and how much the user has assimilated and understood the collected data - this seems to depend on how good and effective the user was in the visualizing, searching and consulting phase. It would have been interesting if the author had discussed, more directly, possible strategies to address these issues in software.

In other words, it seems to me that the suggestions provided by the author will be effective when users are well versed in their domains and have a reasonably good understanding of what they would like to achieve. An interesting question is whether these are necessary and valid assumptions to have when design such software.

The Designers’ Outpost

An interesting project that combines the features of physical spaces like walls and papers with electronic media. The initial sections of the paper outline the features offered by mediums like walls and papers: they are tangible themselves and can support many tangible objects, they support collaboration among multiple users, they permit representation of large information structures without loss of context and associations, etc. They also discuss the features they lack : support for remote users, lack of persistence and versioning, related information should be manually shifted while shifting the objects, etc.

Their architecture facilitates direct manipulation using the physical medium - post it notes and pen are used to create information artifacts, the stylus is used to make links between post it notes, tapping the notes allows users to replace them with images or delete them, both the post it notes and the electronic media represented by them can be moved in a direct way. All this was accomplished by a suite of hardware and software: a touch sensitive board, a system of cameras, projector and computer vision software. The rear camera detects a new note and the location, the front camera detects the ink. An interesting, notable idea was the visual feedback system - a glow around detected notes, such cues signal the users to stick the note again in case it was not detected thus handling probable misclassifications by the machine learning system.

Their central theme seems to be towards developing an interface and a system that documents collaboration rather than transform it, their argument is that such interfaces are suited for tasks such as early stage design. This makes sense and their interface can be extremely useful for archiving brainstorming sessions that thrive on mixing physical and digital artifacts. However, they have not addressed some of the other limitations they initially stated : remote collaboration and versioning. It would be interesting to consider a DigitalDesk like approach here, the post it notes can be removed once the system has detected them and the projector can then display its image, this would facilitate both versioning and remote collaboration - however, editing contents of the now digitally represented post it note is a problem and can get tricky.

Matthew Chan - 10/19/2010 17:37:01

===Creativity Support Tools===

In this paper, Ben Shneiderman proposes a 4 step framework to spur innovation, more specifically in an 8 step process. Just like sewing machines facilitated fashion and telescopes accelerated the progress of astronomy, Shneiderman believes that new tools for creativity will broaden and accelerate innovation in several domains.

This paper is important because it emphasizes the importance of creativity, especially tools for those who work need to be creative for software. Society can't remain the same with current practices or tools, we need new ones to help us think and explore things we normally wouldn't. Unlike most papers we've read so far, this paper doesn't have any new results, but it does propose a new technique or methodology mentioned in the 4 step framework or the 8 step process from Collect, Relate, Create, and Donate, plus the 8 sub-steps within.

This paper relates to every aspect of today's technologies because developers and designers are actively designing tools and technologies for different user groups with different needs. In order to do this to the best of their abilities, they need good tools to help them think and explore. A great and often mentioned example is the iPod. Just one button and a wheel for an mp3 player and it beat out all the competition in the market based on its superior design.

This paper definitely relates to my work. In one instance, we had to design a speech interface for children who don't speak English very well. Moreover, it was a game we had to design as well.

The Designers' Outpost: A Tangible Interface for Collaborative Web Site Design

In this paper by the Group for User Interface Research, the authors explore the effectiveness of Outpost where users collaboratively author and build web site information architectures on a white board and physical material like images. Moreover, this paper emphasizes "informal user interfaces" such as regular human input (ie. speech and writing) as opposed to recognition and transformation.

The techniques/results/methodologies used were 3 fold: first they evaluated the concepts with paper prototypes, then build mock-ups of what Outpost would do, and finally created a wall-scale prototype for participatory design sessions with professionals. The authors then examined existing practices and observations and then compared it with new interactions, where all the designers were working together on the white board--oneof the most high valued items to work on/with. In one instance, one designer would be at the front guiding the team's thought process. WIth the mock up of Outpost, all design team members were interacting with artifacts on the board.

This paper is fairly important. It's important because it has found a more effective way for groups to collaborate tangibly, but the paper was written in 2001 and teams today still use post-it notes w/o anything close to OutPost. In fact, i just finished doodling and brainstorming with a team member on a white board and we finished by taking a picture of the contents on the board and left. This has no relevance to my line of work since i'm not working on anything involving group collaboration or tangible media. However, it's certainly a good domain worth exploring and we could possibly borrow ideas from this paper down the line.

Aaron Hong - 10/19/2010 17:50:06

In "Creativity Support Tools" by Schneiderman he talks about tools that support creativity. They should lower the barrier that people can even work cross domain. He comes up with some terms to describe the creative process (it's not linear or iterative, he just left them as disjunct bubbles). And also talks about 8 tasks that support creativity

What he says here, may be only a clearer reformulation of what has been already said. All the things he noted are to some degree in existence. Granted this was published in 2002 so the full realization of this vision was not there yet. I'm curious to know whether he actually seeded things we've seen now like interoperability among tools (like CCS3 and other web tools that even let you drag files into the browser based email client to be attached). Also what he describes are to a big degree "protocols," which obviously we don't all abide by. But in the networked world, people are more and more willing to submit to someone else's design in order to be more efficient. It's good to keep what he said in the back of the mind, but much of it seems to be common sense.

In "The Designers' Outpost" by Klemmer et al. they talk about "a tangible user interface that combines the affordances of paper and large physical workspaces with the advantages of electronic media to support information design." Personally, I'm favorably disposed to these kind of beyond reality implementations and it is nice to see researchers constantly bringing in tactile and tacit knowledge back into the picture for these designs. However, I think most people agree these characteristics are valuable. But does this then become not a theoretical problem, but one design and engineering?

Brandon Liu - 10/19/2010 17:52:49

Creativity Support Tools'

I think the paper could have spent more time coming to an account of what 'creativity' means in the context of HCI. When I began reading the paper, I initially thought that 'creativity' would relate to work such as sketching, painting or music composition. It only became clear to me that 'creativity' meant any creation of an original product, including scientific research and law papers.

It was interesting to think about how the applications in the paper relate to some of the previous topics in the class.

For example, we can try to understand Creativity support tools from the perspective of CSCW. The article discusses how people can review and disseminate creative work. When reading about CSCW, we discussed the need for anonymity sometimes in collaboration. How should creative work, which needs to be subjectively evaluated, be spread through collaborative systems?

Also, how does research in input devices apply to creativity support tools? We investigated earlier the idea of direct manipulation and the space of input devices. How does the state space of an input device affect how creative people are? For example, some people are very adept (expert users) at creative tasks such as Photoshop, but most of their initial sketches are done on paper. Is there any way to rectify this difference?

I wasn't too hot on the discussion of implementation details, like how to move items across windows. However mundane those points may have been, it is still interesting since we don't have any unified way of getting content from one application to another, other than copy-paste. I am also curious about what the author was thinking about for higher-level operations like copy-paste/open-save-close for creative tools.

'The Designer's Outpost'

This article was interesting since it touched upon a lot of the things I have done while working. For example, I've often had to sketch out sitemaps and database schemas on whiteboards for other people to critique. Entity-relationship diagrams used by 'information architects' are often sketched out on a whiteboard before the design is committed to a schema. The utility of this is that the representation is much more fluid than entering it into something like Microsoft Visio.

The fundamental issue the paper was addressing is how to rectify the difference between physical media and computer systems. Physical systems can have fluid associations and are a shared experience among lots of people. Computer systems give you remote access and version history.

The most interesting point in the paper was how designers 'unanimously felt that automatic grouping was not useful, as they already knew the layout of the notes.' When we discussed ubiquitous computing, we discussed the problems of inferential systems, and that was when we were only considering 'tame' problems. Since creative problems are often wicked problems, it is an open question what use inference (if any) has in creative tasks. The automatic grouping of the notes may actually have hindered the abilities of the users of the interface.

I feel like a followup to this paper would be to map out the space of all interaction techniques for the domain (designing a web site). the spectrum would go from the most physical to the most structured (whiteboard and post its to Dreamweaver) and from the most early stage to the final product. The fundamental question of the paper is whether or not structured environments like DENIM and Outpost are effective in the very early stages.

Drew Fisher - 10/19/2010 17:53:17

Creativity Support Tools:

"Smooth coordination across windows" is a nigh-impossible task. While we have MIME and drag-n-drop, the expectation that one could "...get an English definition, a French translation, or a medical dictionary report, all in a predictable screen location" suggests that either 1) the user himself chose where to put this location, or 2) another designer reserved that space for that functionality. The problem is that the number of directions that a single context can go in far outstrips the amount of screen real estate available. There is simply no way to present EVERY possibility of what a user might want to do with a piece of data. The solution is context-aware software - on the upside, it adapts options to fit the situation. On the downside, it is likely that those options will not be commitable to memory, since they may vary for each situation.

Thus, an approach that classifies and limits the options available to those particularly useful for the user, or those in support of the user's tasks, is a valuable contribution. This is precisely what the paper offers. However, I can't help but find it to be a somewhat vague description of design itself.

Offtopic: wow, this paper is firmly entrenched in the WIMP paradigm.

The Designers' Outpost: A Tangible Interface for Collaborative Web Site Design

This paper details a project that improves upon current physical and digital whiteboards by creating a hybrid system called "The Designers Outpost" that plays to the strengths of each system. This hybrid system offers the benefits of physical interfaces, like the ability to easily add notes and draw free associations on a shared board, as well as the permanence and easy archival of digital artifacts. The system also provides for the potential of realtime and asynchronous remote collaboration via the internet.

It seems that information architects benefit from this system, whereas graphic designers saw less benefit from such a system. This suggests that success of collaborative systems are largely dependent on the problem the system attempts to solve - what may work well for one group dynamic may not function well for nor benefit another group.

Matthew Can - 10/19/2010 18:00:54

Creativity Support Tools

In this paper, Ben Shneiderman lays out a set of criteria for developing successful software tools that support creativity. His framework consists of four high level activities that creativity support tools must facilitate: collect, relate, create, and donate. Furthermore, he proposes eight tasks that should help people be more creative and that also accomplish the four activities.

I agree with Shneiderman that it would benefit us to increase the rate of progress on creativity support tools. As he states, such tools can open up new opportunities and accelerate innovation. The eight design tasks can help guide the development of new tools because they are the fundamental tasks undertaken by designers. Visualizing and relating are particularly important tasks early on in the creative process that are not as well supported as other tasks in today’s applications. While I do think the eight design tasks can be helpful, I wish the author had provided some rationale for why he chose these eight tasks. The paper presents no argument for why these tasks are the most relevant.

One thing I liked about the reading was that it addressed the idea of integrating the creative activities. In particular, the concept of higher level action patterns, like annotate-consult-revise, makes us consider what kinds of often performed actions can be packaged together and shipped as features of creativity support tools. It’s worth noting that Shneiderman highlighted the importance of integrating novel tools with existing ones. This is in contrast to Olsen’s view (from a previous reading) that supporting legacy systems is unimportant because it only serves to slow down the progress of UI systems.

I would also like to add my criticism that Shneiderman was too extreme in describing the potential deleterious effects of creativity support tools. Was it seriously necessary to hypothesize that the tools might be used by “dictators, terrorists, or criminals who seek to dominate, destroy, or plunder”? That is potentially true of any tool, and I don’t see why it’s particularly relevant to creativity support tools.

The Designers’ Outpost

This paper describes and evaluates the Designers’ Outpost, a tangible user interface for web design tasks. Similar in spirit to the DigitalDesk, the Designers’ Outpost combines the affordances of physical objects (especially paper and large surfaces) with the benefits of computing technology.

The key contribution of this paper is a proof of concept; the authors developed a system for web designers, demonstrating through a study that it supports existing work practices and that it provides additional value to designers because it leverages the advantages of the physical and digital worlds.

What I liked most about the paper was its emphasis on how much the authors learned about web design from the professional design study. They originally thought that a potential benefit of their system is that it can provide immediate visual feedback. It turns out that users find the automatic visual feedback distracting. This suggests that only explicit actions should cause the system to provide feedback. The work processes of the information architects also informed the authors that their system should support linking annotations to notes. These examples illustrate the challenge of developing creativity tools for designers. My experience is that software engineers often approach these problems with preconceived notions of the "right" solution, as if they know the designer's job better than he does. I think it benefits both parties if the developer takes an ethnographic approach to understanding the designer's task.

Arpad Kovacs - 10/19/2010 18:06:35

The Klemmer reading describes a the "Designer's Outpost" concept, which combines the affordances of paper and wall-scale tangible user interfaces with the benefits of electronic manipulation in order to create a more efficient website design tool. The benefits of this approach is that it facilitates explanation, development, and communication of ideas during the early phases of design through a tangible, easily manipulable and organizable interface. However, it also allows the user to maintain their ideas in persistent, electronic storage, with associated benefits such as versioning and remote access. The authors classify the system as an informal user interface that aims to document, rather than transform the input, and thus avoid obstructing the users' workflow. The paper then proceeds to describe the user interactions permitted in the system, and the results of usability tests with actual users.

The contribution of this paper is effectively an existence proof for tangible wall-scale tangible user interfaces with electronic integration. The authors show that this interface can be used for collaborative design, integrated with other electronic systems (eg DENIM and computers), and brings clear benefits over existing solutions (such as plain whiteboard and Visio); however to me it seems that the scope of interactions could be greatly expanded. Right now, the system is excellent for recording the state of the system and sharing it with remote users; however I think that the addition of electronic manipulation can also help assist information architects extract additional data from the layout/positioning/interaction of multiple nodes, as well as recognize editing patterns that would otherwise be lost. For example, I think that a useful addition to the system would be the ability to cross-reference notes, eg by selecting a particular note, the system could highlight related notes, or historical positioning/links of that note. Another option would be identifying the most or least edited clusters of nodes, in order to inform the users of which parts of the system are controversial or may need additional polish.

It was interesting how much the designers and information architects valued calmness in the interface, to the point of one group turning off the smartboard functionality. It seems that one of the greatest challenges in creating an electronic tangible interface is ensuring that the system captures and interprets input correctly, without disrupting users' creative processes due to excessive and obtrusive feedback notifications. I think that the limited approach taken by this system of simply identifying particular notes, without attempting to recognize their content is a good compromise that should reduce the error (and hence disruption) rate, without reducing functionality (since the target users are more concerned with capturing the layout and relations between various notes, rather than say searching/modifying the content of the notes themselves).

Schneiderman's article provides a framework for designing Creativity Support Tools. The first question a designer should ask is who are the users, and what activities do they engage in. Schneiderman identifies the 4 main creative activites as collect (learning from earlier works), relate (sharing with peers and mentors), create (explore, compose, and evaluate candidate solutions), and donate (dissemination of results). The real challenge for the designer is integrating these activities with existing and new tools via data sharing, compatible and consistent action patterns, and finally smooth coordination between windows.

The main contribution of the paper is identifying 8 tasks that designers should engage in, in order to better familiarize themselves with the 4 creative user activities. 4 of these tasks, namely searching, relation, composition, and dissemination are direct manifestations of the collection, relation, creation, and dissemination activities respectively, and I feel that these can be found quite readily in most software packages that I have used. In my opinion the remaining 4 tasks are less obvious, and therefore more interesting. Visualization is an excellent approach that transcends all of these boundaries, and in the process makes it easier to spot patterns and interesting quirks in data; unfortunately it has become widespread only recently. The task of assisting thinking is quite difficult to develop tools for, since it must be extremely flexible, and at the same time help maintain concentration; perhaps this is why people often try to brainstorm in isolation, using low-tech tools like paper that avoid the distractions of cellphones, instant messaging, email, etc. I have not seen much work in exploration tools that try to evaluate decisions other than simulations like the aforementioned simCity; perhaps this is an area that is ripe for research. Finally, a few programs (such as Google docs) have started to support tracking revisions, although this is often transient and only available for the current session (eg in Photoshop); it would be intriguing to have a running history of all actions you have taken, and then run this dataset on an analytics engine to identify inefficiencies and stimuli that result in creative high points.

Pablo Paredes - 10/19/2010 18:21:26

Summary for Shneiderman, B. - Creativity Support Tools

This article reveals a very important aspect of design, which I consider key in innovation, which is the notion of sharing. The author describes an approach with four phases: collect, relate, create and donate which define a high level process of design that enable a series of tools for creativity.

The author further dissects this view to a series of tasks that allow people to be more creative more of the time: searching, visualizing, consulting, thinking, exploring, composing, reviewing, disseminating. In turn, this additional layer of tasks becomes a structured and pretty complete list of the types of tools that are necessary in the process of creation. These tools can of course be as simple as a pen or as complex as mind-mapping software. However knowing that all these tools are needed and provisioning to have them available helps to make sure to attain the overall cycle of creation, which in turns makes creativity more effective.

I found this article very concise and helpful, a true support to provide structure in the design process, especially in a collaborative process. I truly liked the view of the article to actually include de dissemination/sharing of the information with other designers/experts/peers an embedded component of the system, and not an afterwards left-over task that is left sometimes as an pseudo-altruistic task. I would like to see more analysis from this author related to his view on how can this process be replicated in other settings or environments where creativity should be a larger piece, such as science, policy making, marketing, etc. and observe if there are further lessons that can be learned from these different environments.


Summary for Klemmer, S. Newman, M., Raffell, R., Bilezikjian, M. and Landay, J. - Designer's Outpost: A Tangible Interface for Collaborative Web Site Design

The main lesson from this article is the notion of fidelity. The actual sensibility from the researchers/designers of the Outpost to go out and admire and pick up the tools currently user by web-site designers from an angle of perfection, rather than the opposite, i.e. looking at simple walls, post-its and markers as an ultimate goal, and not as a limitation for current design.

This humble approach rendered a very interesting tool that incorporated the best of both worlds, the physical and the digital. The design of the outpost remained simple and most of the learned lessons helped to further keep the interactions with the user to a minimum, a concept defined as calmness by the authors. The system design was also quite remarkable, as the authors wanted to provide high reliability, and exploring computer vision as a complement system to capacitive sensing provided a very good solution.

Another important concept was the explicit nature of the system. The authors could have added several layers of complexity to the system providing automated feedback, but a key element in the process of design is the ability to concentrate in the concepts being treated in the board, and all the actions, drawings and elements added or eliminated are actual reflections of the design process state. Therefore, adding complimentary "useful" information actually was considered noise by the participants.

As a whole the system provides a very innovative way for designers to overcome some of the drawbacks of physical systems, such as its lack of portability, replicability, and communication with remote partners, by providing a very valuable digital view of the system. However, an integral component of the design process was not captured, which is the design rationale, based on the discussions that many times are captured as side notes.

In summary the key lessons of this paper are 1) the notion that (new) tools can in fact enhance some features of work, but not without compromising others, 2) nothing replaces the real world (but it could be carefully enhanced) and 3) fidelity and a humble approach to designing new tools is very important. I believe one aspect not touched in the paper and that requires further definition is a view of the mental model that the users had from the physical tools and digital tools, to understand what is the value and compromises made in migrating from one (traditional) tool to a new tool.

Linsey Hansen - 10/19/2010 18:34:59

The Designer's Outpost: A Tangible Interface for Collaborative Web Site Design

In the first article, the authors discuss the methods they used to create an interactive whiteboard of sorts for web design. This device is able to use physical post-it notes, digital pen data for structures, and touch-screen based gestures for interacting with the data.

The main thing I found cool with this device was it's ability to combine the quick iteration methods of the real world (ie writing on post it notes) with the more mobile and easier to save digital world. In a way, this device was pretty much an extension of a digital desk or collaborative digital whiteboard (which was referenced a few times).

One thing that did kind of bother me about this set up was the fact that all it really does do is save the connections between post it notes. While I know that this is a really important part of web design, especially if there are a ton of nodes (or sites) to keep track of, I feel like some physical device could do the same thing (ie pieces of string with re-stickable tape on both ends), though this would be less elegant. I suppose that the long term plan for this is to allow the users to see different versions, of the design process, so that they might be able to go back and look at previous layouts for reference, but this just seems like a really expensive solution to me. Also, I do not see how this would work if a user wants to use this from home, or if a past version included a note that has been modified or removed- I feel like some other sort of information would need to be digitally attached to each note for this to work, but then if this is the case, it almost seems like it would be more advantageous to make then entire thing digital (ie digital post it notes on a giant touchscreen), but maybe this is more expensive or harder to work with.

Creativity Support Tools

In this article, the author discusses activities that need support in order to assist users with creativity, which included collect, relate, create, donate. From there he goes on to describe eight tasks that can be used to help people be more creative.

I felt like a lot of the things described in this article already exist to some degree. For instance, the Kindle uses an interactive dictionary for books and web pages (or it is supposed to support those eventually), plus, I am pretty sure that I have seen dictionary internet plug-ins. Then in the photo example, facebook pretty much does that, where it links information about the place, photographer, and people present in the photo along with the actual picture. I suppose that what this article is calling for is more general programs so that users can do these tasks anywhere at anytime (ie have an annotate option for photos and documents just like a cut and paste), and in most cases these seem completely doable and would just require a new compression format.

David Wong - 10/19/2010 18:37:47

1) The "Creativity Support Tools" paper discussed a framework for creative thinking, discussed 8 tasks related to creativity, and discussed how those tasks can be implemented in software for creative tools. The "The Designers' Outpost" paper discussed an electronic whiteboard system that helps in the web design mockup phase.

2) I didn't think that the "Creativity Support Tools" paper was very insightful. It discussed a framework for creativity and subtasks to support creativity. However, it wasn't anything new. The ideas for integrating software to include actions for "annotate-consult-revise" and "collect-explore-visualize" was novel, but not a groundbreaking idea or discovery. Altogether, I didn't think the framework was that helpful and those subtasks were, for the most part, quite obvious.

The "The Designers' Outpost" paper drew off of similar approaches to collaborative multitouch input devices and electronic and physical content managers in the HCI literature. It is a proof of concept of whether a tool that combines physical and electronic properties in the web design process can actually add value. As a proof of concept, I think it may help inspire others who want to do applications in this area. Otherwise, there were no other significant innovations in the paper.

3) The "Creativity Support Tools" paper, in my opinion, had a poorly formed argument. It proposed a framework for creativity and discussed how software can help aid that. There was no argument supporting the framework, and the last section for skeptics was out of scope--it gave some anecdotes and discussed how creative software could be used for bad, but it didn't say anything of what the paper had just discussed. Moreover, the argument seemed sound, but that is only because the paper proposed a general and broad framework.

The "The Designers' Outpost", as a proof of concept, conveyed that their system could actually work. The experiment they did was a over-the-shoulder informal user study. Accordingly, the results they got did not prove any validity of the system, but merely showed that it has value. They noted the flaws in their system, noted how they have improved on those flaws, and indicated how users felt about certain features. Overall, it was a sound proof of concept paper.

Siamak Faridani - 10/19/2010 18:48:43

In the first article, Ben Shneiderman (who is a very prominent HCI researcher) highlights a number of factors in any tool that supports creativity. In his paper, he himself follows his method and starts by looking at other research done in the filed. He builds upon the older work done by Couger. Couger has reviewed 22 different methods to support creativity. Based on that Shneiderman highlights for activities withing his framework. Collect, Relate, Create and Donate are the activities within this framework and to be honest it is not yet clear to me how donation contributes to creativity. The main part of the article is his 8 tasks that he highlights for a system to support creativity. Again I am not sure if disseminating can add much to supporting creativity. I might add to the paper some other items that will add to creativity. Mobility makes people more creative for example people move around the room when they think. There are other things that may spark creativity. I know people go and jug, some people relax, play sports or sit under an apple tree to find ideas. My question is how can we add these techniques under Shneiderman’s task list?

The second article is an example of how design should be done. Authors follow an interesting design procedure. They start by looking at current tools that are being used in a creative process (web design). In addition to the tools they also go into each task that designers do and break them into individual tasks that can be emulated in a computer driven system. And they finally implement and deploy the system. To me it looks like that authors were so happy about their EM algorithm, XML and OpenCV work that they forgot to comment about the magnitude of the impact of their work. I believe their work is amazing in terms of implementation but suffers from the same problems that it inherits from the paper and post it notes system. As we saw in older papers an HCI work should be significant, I feel that they have designed another VISIO but I am not sure if they increase the level of creativity by their tool.

Shaon Barman - 10/19/2010 18:49:02

Creativity Support Tools

This paper discusses a framework and set of tasks that can be used in different creative processes.

I was not sure what mindset to read this paper in. Its not technical, nor does it discuss concrete future ideas. The main takeaway seems to be advocating the use of technical tools to enhance the creative process, a process that it sometimes hard to define. The Create-Relate-Create-Donate framework seems to be one way to breakdown these creative processes so specific tools can be designed. One aspect which I found interested was the use of a common language between software. Transferring data between two programs can be difficult. I recently had to convert a mp4 to wmv, and this wasted a significant amount of time. Automating such mundane details would greatly enhance the creative process. Overall, human creativity is just another task that tools can help automate. Such tools are evident in movies, games, tv, photography, etc. There are even certain types of art that could not occur without the invention of computers.

The Designers Outpost

The authors develop a tool which enables web designers to have the useful features found in physical world, such as a whiteboard and post it notes, but enhanced using digital features, such as playback and ability to transfer data between two places.

We design is a difficult task which takes lots of creative power thinking about possible use case scenrios and transitions that the user might encounter. Many of these tools are difficult to do on a normal desktop machine. The authors propose an enhanced digital whiteboard, basically one tuned to the use of designing a layout. The whiteboard is able to find post-it notes and the lines which connect them. It seems like this general interface language could be developed, such as capturing an image, creating a shape, etc. so that more domain-specific design interfaces could be created easily. Another interesting aspect of the experiment was that the users were annoyed when the computer automatically grouped together notes. This seems to indicate that if the computer "automatically" predicts user behavior, users will have a low degree of tolerance for incorrect decisions.

Thomas Schluchter - 10/19/2010 18:49:38

Shneiderman: Tools for creativity The article develops areas of focus for designers that are interested in supporting creative tasks. These areas of focus are grounded in a framework of the creative process that includes collection of materials, discussion and exchange with others, creation and dissemination.

I disagree on several counts. The "Skeptic's Corner" section of the article tries to preempt the criticism that creativity should be considered a prerogative of humanity; and the author readily admits that technology can't replace creativity. But that is too black-and-white of a discussion and misses a different, important point. What the article still implies is that tools can enhance creativity. I would argue that creativity is a function of talent, skill and practice. Tools can aid creative processes where talent, skill and practice are already at work. But they can also create a false sense of quality. Presentation software, explicitly mentioned in the article, seems like a particularly bad example for this: Everybody uses PowerPoint, but because of the specific limitations of the tools, people are locked into a model of presenting dictated by the software - and especially the "creative" use of PowerPoint's features by people who don't understand the underlying principles of a good presentation lead to highly dysfunctional products. Just because a task is made easy it doesn't mean that otherwise bound creative resources will start to flow freely.

Another issue I have is the general idea espoused towards the end that technology can make positive change happen. This technological optimism is very common in the tech community and a little short-sighted if I may. The sewing machine didn't *cause* a fashion revolution, it merely happened to be developed during a time when a sufficient number of people saw use for it to make it a mass-marketable commodity. The same thing applies to creativity tools: their existence predicts nothing about their adoption, and their adoption predicts nothing about the specific cultural aspects of their use. The author acknowledges the possibility of such concerns, but brushes them off pretty quickly.

Lastly, I'm not sure I agree with the creativity framework. In my mind, it doesn't say much about the conditions under which creative thinking occurs, and more about the condtions under which the production of non-material goods happens. Grandted, tools definitely change these conditions. But the author (beyond systems that support "free" association) offers few examples of tools that would actually enhance the process of coming up with ideas.

What I liked about the article was the thought that tools should offer new and appropriate task sequences to support common activities. Again, I think that locking people into tools and have their thinking take place within the confines of it is not the right approach, but for well-known and well-defined tasks, this can reduce the headache of creating media, putting together papers etc. But fundamentally, this is helping with the legwork, not with the brainwork.

Designers' Outpost The authors present a touchscreen wall system that bridges the gap between the fluidity of working with paper and pen and flexibility of maintaining a digital record of the work done with paper and pen. A user study with website designers reveals that there is a need for this kind of application, and illustrates important design concepts that can make it successful.

I think this work illustrates very well the point that a system that supports well-defined tasks is superior to a system that tries to replace the task with another task that is streamlined for a particular interface. By replicating a highly familiar interface, the whiteboard, preserving all the affordances that it offers and augmenting them in the background the Designers' Outpost does two things: 1) It addresses the gap between the analog and the digital by co-evolving both representations of the same concept. The example of the designer photographing a whiteboard and filing it away is very typical of regular design processes that lose memory of former states of knowledge as they go along. 2) It takes optimal advantage of both the analog and the digital. The analog contributes space as an organizing dimension and the digital contributes time as an organizing dimension. This is not to say that the analog and the digital have exclusive access to either of the dimensions, but they have specific properties that lend themselves more to organizing information in one way or the other.

One interesting way of developing this work further would be to visualize the design process itself as a history of events. The whole discussion of capturing design rationale has to do with the desire to convince other stakeholders of the soundness of the process. What if designers had an immersive way of representing through which phases the final concepts went, which alternatives were under consideration, how the selection process worked? This would not only be of interest to external evaluators of design processes but it would also help the design team to build a coherent narrative for itself that could guide further work.

Richard Shin - 10/19/2010 19:02:56

Creativity Support Tools

This paper chronicles the author's experiences in trying to develop a general-purpose software tool to support creative work. The author identifies four main activities involved in creative work: collecting previous creative works from various sources, relating (consulting) with peers and mentors, creating the actual work, and donating (disseminating) the results (often to sources from which works were originally collected). To support these creative activities, the authors proposes eight specific tasks that software tools should enable: searching (libraries of content), visualization, consulting (of peers and mentors), thinking (to make new ideas), exploring solutions, composing artifacts, reviewing previous activities, and disseminating the final results.

The paper's central idea is that software can help bolster human creativity, which seems like a powerful concept to me. Since much of what people already do with software involves creativity, whether experiencing others' or creating something new, having tools that recognize this important use case and are designed to enable it would greatly enhance what many people are already doing today. Armed with these frameworks for how people work creatively, people who make software tools can make sure their designs are compatible with and bolster creative processes.

However, the distinction between the four activities and eight tasks seemed a bit unclear to me. Perhaps the four activities are too vague to be things for software to implement; but then, given the eight specific tasks, do we really need a separate notion of four activities as well? Also, while the paper lays out what things software should support in order to enable creativity, it barely discusses how existing software might already be supporting these tasks or how they might be lacking in that respect, other than some simple anecdotes. It would have been helpful if the author related his ideas more to what already exists, so that can be more easily improved.

The Designers' Outpost: A Tangible Interface for Collaborative Web Site Design

This paper describes a system built by the authors to support the types of collaboration that web designers engage in: using sketches and words, on walls and tables using Post-it notes or large pieces of paper. Called The Designers' Outpost, the authors' system provide a tangible user interface which enables collaborative authoring of web site architectures on an electronic whiteboard, using physical media like Post-it notes and electronic pens. The authors evaluate their system with professional web designers, validating the usefulness of Outpost to web site architecture authoring.

Outpost reminded me a lot of DigitalDesk; they seemed very similar in how they both attempt to bring paper and its many benefits into the digital world, by allowing tangible manipulation of pieces of paper containing information while also supporting functionality like copy-and-paste that only digital systems can offer. Specifically, in building this system, the authors took advantage of how Post-it notes are easily and intuitively manipulable by users, while other tasks (like maintaining links between Post-it notes) are cumbersome and don't work as well on paper, and attempted to combine the advantages of each.

Much of the paper is dedicated to evaluation of the system, particularly to the designers that the authors brought in to test the system and provide feedback. While the paper contains a lot of qualitative discussion, there is very little quantitative data to demonstrate the benefits of this system. I thought that such data would have helped identify more precisely the benefits and drawbacks of the system without bias, and provide an objective sense of how helpful it is over the traditional one.

Bryan Trinh - 10/19/2010 19:06:27

Creativity Support Tools

In this short article Ben Shneiderman outlines eight features of software that can help people be more creative: searching, visualization, relate, thinking, exploring, composition, reviewing, and disseminating.

Classification can be good in some cases, but I just don't see the point here. Why think in abstract terms when the developer still needs to understand the context in which she is building the application for anyways. These 8 features would come naturally out of the contextual inquiry phase or understanding phase of the design process without formulating them before hand. It might also be dangerous in predisposing the designers to certain tasks that neatly fit into these categories.

The Designers' Outpost: A Tangible Interface for Collaborative Web Site Design

This paper provides an overview of a tangible user interface built specifically for web architecture planning. The impetus for such a device came from the observations of the ways web architects are currently designing their websites. Designers tended to use large public walls to plan out the structure of their websites, and they wanted to replicate this paradigm while adding more functionality by digitizing the information.

This interface successfully replicated the dominant way that web designers plan out their projects, while providing all the benefits of digitization of the content. The direction that they took begs the question, is this the right paradigm to follow when designing tools for designers. Do we just take the existing mode of work and augment it with digital technology? The Anato pen does something similar, but many have said that, it simply was not flexible enough for them. The digitized content was not of much use to them afterwards. It is clear that the interface was popular from a user testing perspective, but will it eventually be adopted by designers, or will they simply choose purely digital or analog work flows?

Aditi Muralidharan - 10/19/2010 20:20:17

In "Creativity support tools", Shneiderman, expert at all things UI, clarifies the activities that any creativity-support tool must support. He does this by breaking up the creative process into four stages and proposing eight tasks that can guide a designer through these stages by repeated application.

The eight tasks are: (1) Searching and browsing digital libraries, the Web, and other resources (2) Visualizing data and processes to understand and discover relationships (3) Consulting with peers and mentors for intellectual and emotional support (4) Thinking by free associations to make new combinations of ideas (5) Exploring solutions—What-if tools and simulation models (6) Composing artifacts and performances step-by-step (7) Reviewing and replaying session histories to support reflection (8) Disseminating results to gain recognition and add to the searchable resources

and the four stages of the creative process are collecting, relating, creating, and disseminating.

As far as my own research goes, this is the least useful Ben Shneiderman article I have ever read, mostly because it describes the creative process in (retrospectively) obvious terms. Nevertheless, this same characteristic would probably make it useful to a designer of design-support tools. While Shneiderman himself admits that his list of tasks and stages could be incomplete, I cannot think of a way to add to it. His list is hard to criticize, especially when there isn't yet a tool that supports all of it.

A concrete example of this is presented in the second paper "The designer's outpost..." published a year before the first. Instead of aiming at a general theory of creativity and implementing web-design cases of more general creative tasks, the researchers take a bottom up approach. They study web designers and find that they use pens and paper and walls to outline site structure and pages. Then, they implement a version of this using a whiteboard, a projector, and computer vision to create a tangible UI to a creative tool that allows designers to add electronic content to their designs and save their designs electronically.

While it may not have been part of the scope of the project, I think this paper would have benefited from some analysis about how the learned principles could be applied to other aspects of tangible UI design, or other creative tasks.