Reflective Physical Prototyping
From IEOR 170 Spring 2007
Required Readings
- Reflective Physical Prototyping through Integrated Design, Test, and Analysis Björn Hartmann, Scott Klemmer et al. ACM UIST 2006.
- Authoring Sensor Based Interactions Through Direct Manipulation and Pattern Matching Björn Hartmann, Leith Abdulla et al. ACM CHI 2007
Discussions
Please post your critiques/commments on the required readings below. To do that, first login by using your user name and password, then click the "edit" tab on the top part of this page (between the "discussion" page and the "history" page), New to wikis? Read the Wiki editing guide. . Hint - Please put a whole line == ~~~~ == at the beginning of your submitted critique, so the wiki system will index, sign and date your submission automatically.
[edit] Bryan Boling 10:52, 4 February 2007 (PST)
“Reflective Physical Prototyping through Integrated Design, Test, and Analysis”
The d.tools software seems like a fairly powerful design tool that has a wide range of uses in the design process. Working through my own design classes I know that trying to build a good prototype is often one of the most challenging issues to deal with during the design process. By introducing d.tools as a fast prototyping device, this gives the designer more time to spend working on functionality and testing. Also, having a prototype on hand early in the design cycle can be beneficial for most designers. I know from my own experience that a prototype can often reveal many real world challenges that calculations or research just can’t. For this reason alone, it would be highly beneficial to employ something like d.tools during the design process. In addition to the prototyping capabilities of d.tools, it also allows the designer to test and analyze the design with real users and a multitude of analysis functions. By providing this real-time feedback and analysis, d.tools can serve as a powerful tool from start to finish in the design cycle.
“Authoring Sensor-based Interactions by Demonstration”
As a mechanical engineering student I often shy away from programming, and mechatronics design has always intimidated me. However, through the use of applications like the Exemplar prototyping tools this article discussed, “programming by demonstration” is possible. Although I lack a lot of experience programming, I understand many of the concepts behind program architectures, and subsequently a graphical tool for quick prototyping and programming would be very beneficial. Additionally the real-time feedback control and iterative functionality of these tools allows for extended testing and modification of sensor-based design in a much shorter timeframe than the more traditional design process. Also by applying the design-test-analyze paradigm to this design tool it can be used throughout the design process, ultimately allowing for increased testing and refining of the prototype.
[edit] Erin Palermo 14:09, 4 February 2007 (PST)
"Reflective Physical Prototyping" This article seemed to highlight a lot of what we have been talking about already: iteration in the design process. It discussed the importance of developing a prototype, then testing it with users, then making adjustments after gathering data from the useability study. I am a little bit confused as to why this seems like an "innovative" concept, because to me, it seems like common sense! I think a design would be incomplete if it were never tested with users to determine its flaws. The only way to make something better, and ensure its success, is to test the product using a prototype, so that the user has an accurate model of what they will be using. I was encouraged that the article mentioned that a designer doesn't have to be fluent in programming, just have some experience. That is nice from an Industrial Engineering standpoint, seeing as I do not take courses in computer science. The "d.tools" device seems very useful. I like that it records users and their interactions with prototypes so that you may later review their body language, their excited utterances while using the device, and any other helpful information. I also like that d.tools correlates the video with states on the state chart, for easier navigation later in the redesign process. This new tool will greatly aid the design iteration process.
"Authoring Sensor-based Interactions" Overall, I wasn't impressed with the artice. I was a bit confused at first about Programming by Demonstration (PBD). It was difficult for me to understand what the goal of this type of programming was. I couldn't really see how programming in a different way was actually valid and useful. Further, I didn't think that the point about 'generalizations' was made clear. How would they generalize? How would they know what information to implement to other projects? The article was, in my mind, rather technical. It didn't use technical equations or anything like that, but it used lots of language without defining it, which made it confusing to someone who is not well-versed in this type of system.
[edit] Onyi Lam 23:27, 4 February 2007 (PST)
"Reflective Physical Prototyping"
This paper introduces d.tool, a toolkit/software that "embodies an iterative-design-centered approach to prototyping information appliances", which basically provides an alternate solution towards expensive prototyping by simulating the products' behaviors. I like how the author also discusses the shortcomings of d.tools noted in the testing of the device and what are his recommendations to improve those insufficiencies. The author also acknowledges the success of d.tools by recognizing that the students were able to create some innovative design through d.tool. The one thing I don't like about this article was that it is highly technical and it's hard to follow and visualize if you don't have d.tool with you.
Authoring Sensor Based Interactions Through Direct Manipulation and Pattern Matching
It is another highly technical paper and I had a hard time following it given that I don't have a lot of background in computer science nor programming. The author starts off saying that specifiying the relationship between sensor input and application logic, and the software Examplar, will allow users to focus on design thinking rather than "algorithm tinkering". The article then goes off explaining the architecture and logic behind this application and then adopted two approaches to test the system. I like the second approach - the first use study - in which the researchers ran a controlled study on the usability of this application, very much aligned with the design process that we've been studying. The other thing that catches my eye in the article was the mention of "LabView". I've used LabView before and I was very amazed with its functionality and visualness. Exempler targets to a different population and it is also very powerful in allowing designers to do rapid iterations
[edit] Urvashi 03:24, 5 February 2007 (PST)
"Reflective Physical Prototyping through Integrated Design, Test, and Analysis"
I found this article to be very technical and its basically about a toolkit called d.tools which allows designers to place physical buttons directly onto their prototypes and hence run simulations to find out how it works without having to go through the entire process of recording and observing people that was featured in Monday’s readings. I found this article very technical and didn’t understand many of the specific functions. I do, however gather that this software is a powerful tool for designers as they can test and simulate many different scenarios on the prototype and also record user’s body language and reactions for a more detailed study later. The authors also point out the shortcomings that their toolkit had but then explained the features they added on to counter those shortcomings and they do a good job of convincing me of the usability of their product.
"Authoring Sensor Based Interactions Through Direct Manipulation and Pattern Matching"
The aims of this technology are to contribute a method of programming by demonstration for sensor based interactions that emphasize design control, integration of manipulation and pattern recognition through a visual editor and rapid exploration of interaction techniques on a much shorter timescale. I found this paper much harder to read than the first one, as it is even more technical and no effort has been made to make this reading accessible to someone without experience in programming. Apart from that, I felt it to be very similar to the first article, as both are describing software that can help test prototypes and the information gleaned can help designers continuously upgrade and refine their prototypes until they are ready to present the product in the market. I think on the whole these kinds of software have a lot of potential in the market due to the increasing pressure on product designers to come out with reliable, efficient and stylish products.
[edit] Michael Chen 03:48, 5 February 2007 (PST)
"Reflective Physical Prototyping through Integrated Design, Test, and Analysis"
This homework article is a scientific paper describing the development of a program called d.tools, a GUI program which allows for rapid prototyping. To be quite honest, I found the paper to be not that revolutionary. While there are a couple of novel ideas, on the whole the concept was none too mind-blowing. With that said, if widely adopted, such a program would lead to a veritable explosion of design. Such a situation is kind of like the iPod. When you think about it, it's only a sleek-looking MP3 player with a cool interface. Apple was neither the first company to make MP3 players, nor the first to use a touch screen interface. But it was the combination of the two that created the modern music revolution. It's the same with this program. There are two really good uses of technology that I can identify: the concept of programming by demonstration, and the remarkably intuitive creation of new states via drag and drop of the signalling icon. Though it doesn't go into too much detail about it, seamless plug-and-play of input and output devices, while pretty much a given designwise, will be crucial for the success of the product.
"Authoring Sensor Based Interactions Through Direct Manipulation and Pattern Matching"
This is a later paper, by the same group, with the same first author, that further refines the signal processing component that will be inputted into the d.tools GUI. While a little more technical, the signal processing component is still a crucial step in the development of a prototype, and is just as important as the macro view provided by d.tools. The author once again explains modifying signal input through the concept of programming by demonstration, and goes on to explain the GUI from which to implement threshold determination, signal scaling, and pattern matching. In both papers, I found the last portion of the paper, describing the end-user testing, to be especially indicative of the efffectiveness of the system, and was also entertained by the description and practical results of the experiment.
[edit] Siu Pan Lam 16:21, 5 February 2007 (PST)
“Reflective Physical Prototyping through Integrated Design, Test, and Analysis”
Most of this paper is focused on presenting the diverse functions of d.tools, a toolkit that embodies an iterative-design-centered approach to prototyping information appliances. One of the functions of d.tools is its ability to log all user interaction events and employs this log to automatically structure the test videos. I find this a very powerful function because the product designers can gain significant usability insights on their products. Often times, the major disadvantage of using test videos is that the analysis of these videos can be time-consuming. The toolkit, d.tools, is able to provide fast access to video data by enabling access to the interaction code (the statechart) from a record of its execution (the video), hence eliminating this major disadvantage. I think it was advantageous for the designers of d.tools to establish a controlled laboratory to assess the usability of their toolkit because it allowed them to discover some of the shortcomings associated with their product design. For example, participants of the study found it disruptive that d.tools lacked software simulation of an interaction model. This prompted the addition of the toolkit simulation mode. This shows us the importance of conducting laboratory studies in the assessment of the usability of a product design.
“Authoring Sensor-based Interactions by Demonstration with Direct Manipulation and Pattern Recognition”
The focus of this paper is on sensor-based interactions. Specifically, it introduces a rapid prototyping tool, Exemplar, which enables users to focus on “design thinking (how the interaction should work) rather than algorithm tinkering (how the sensor signal processing works).” I was impressed with the designers’ evaluation of Exemplar, for it covered a broach range of criteria, some of which included visibility and juxtaposability, viscosity, diffusiveness, and error-proneness. This comprehensive evaluation was very helpful to the designers because it enabled them to discover the lack of visual feedback for resizing regions. Furthermore, a controlled study of Exemplar was conducted for the assessment of usability and felicity of the tool for design prototyping. I feel that this type of study is very effective as it enables the identifications of areas in the design that are in need of improvement. As a result of these findings, designers will be able to produce finer products as part of their future work.
[edit] Christine Shih 19:38, 5 February 2007 (PST)
"Reflective Physical Prototyping through Integrated Design, Test, and Analysis" This article talks about prototyping and how there is a demand for more realistic prototypes which is why they use this toolkit called d.tools to facilitate the process rather than have to code and build from ground up. d.tools combines visual authoring with programming, offers architecture for physical interfaces, and integrates design, test, and analysis of information appliances. One thing i thought was interesting was how d.tools is integrated into the Eclipse platform and how java is usedto make changes- i wasn't expecting programming to be necessary skill in designing prototypes.
"Authoring Sensor-based Interactions by Demonstration with Direct Manipulation and Pattern Recognition" This article introduces a prototyping tool called Exemplar which allows a wider audience of people to be able to turn raw sensor data into useful events for interaction design through programming by demonstration. The main idea of Exemplar is to have users focus on design thinking rather than algorithm tinkering.The idea behind programming by demonstration is that you don't have to know how to program textually, you can do it through this often visual interface by observing examples of logic. The rest of the article talks about Exemplar's functions and a study done on 12 people to test how easy the toolwas to use. One interesting result was that there wasn't much difference in performance among beginners and experts.
[edit] Monica Tanza 21:14, 5 February 2007 (PST)
Reflective Physical Prototyping
This article seemed like a how to manual for a prototyping tool, d.tools. It seems to help in the design and testing of electronic prototypes. The test mode also seems like it would be beneficial in recording where users had problems and improving upon the design. However, this article was very technical and I did not understand some of the components being mentioned and still don’t understand the real world application of this product.
Authoring Sensor Based Interactions
This article was a how to for Exemplar, a sensor date analysis program. Like the previous article, there is no good background and the authors expect readers to be experts and familiar with terms. I am not, so it was hard to follow (yet it was a bit easier to follow than the previous article because I have seen Lab View, which is somewhat similar). I guess the point of these articles is to show an example of using the testing methods that we read about on Monday, but I find these articles dry and overly technical, making them very hard to follow.
[edit] Andrew prasetyo 11:05, 6 February 2007 (PST)
“Reflective Physical Prototyping through Integrated Design, Test, and Analysis”
This article seems to be interesting. It explains the actualization of design, prototyping, and evaluation process. However, the article mainly explains the use (how to use) d.tools as a prototyping tool for any product design. Though the use of d.tools is complicated and requires some background in electrical engineering and computer programming, the tool seems to be very useful in building a prototype. From the pictures shown in the article, d.tools has been used in many design projects, such as: toys, digital camera, console controller, and GPS console. The author also explains the importance of having evaluation shortcoming discovery in the design process. The evaluation will tell the designers what is successful and what fails in the design, and this step can also be used as a building block for the designer to improve the product. At the end of the article, the author also provides some other prototyping devices as comparison to the d.tools and elaborates the plus and minus sides of the prototyping devices. This is very useful because readers can choose which prototyping device will suit their needs. Despite all that, what I dislike the most about this article is the narration which contains a lot of technical and technological terms. I found this article is very difficult to read for a person with limited knowledge of the terminology.
“Authoring Sensor-based Interactions by Demonstration with Direct Manipulation and Pattern Recognition”
Similar to the other article, this article explains about a prototyping device called Exemplar. This article gives us with what the device does and how to use the device as comparison to other prototyping devices. Since this device is still under development, the author also provides readers with an elaboration of first-use-study. The study shows that the device is easy to use and encourage the user to think more about the design rather than the algorithm behind a product design. In addition, the device gives designers more freedom because the device consists of numerous sensors which can be applied to many different product designs. In the first few pages of the article, the author explains about how the device works. As a result, I feel that the narration is difficult to understand because it is full of technical terms (similar to the other article) and electrical engineering/computer science jargons. However, after reading about the first-use-study, I began to understand about the mechanism of the device and the purpose of the device. It is also interesting to notice that Exemplar is developed based on d.tools, which is elaborated on the previous article.
[edit] Esha Datta 19:43, 6 February 2007 (PST)
“Reflective Physical Prototyping through Integrated Design, Test, and Analysis”
It seems like d.tools would be very useful in creating useful prototypes more efficiently. I thought it was nice that designers can use a variety of techniques such as Wizard of Oz techniques, using simulations, and actually manipulating hardware. The combination of software simulations with physical hardware is especially interesting to me. I also thought it was great that the tools support user testing by organizing video capture for easy analysis. Video analysis can be challenging, but organizing the data by state seems like an efficient way to see the key points. I would really like to see a demo of these tools, as it was a little bit hard to get a picture of their full capabilities and how they work.
“Authoring Sensor-based Interactions by Demonstration with Direct Manipulation and Pattern Recognition”
Exemplar and sensor-based interactions also seem very useful for speeding up the prototyping process. Again, I think that the combination of using physical hardware and sensors with software simulations is very powerful. In this way, the designer is able to concentrate more on the design process than the actual tiny details of creating a functional prototype. From the user studies that were conducted, it seems like these tools are very successful so far. Again, I would enjoy seeing a demo of how these tools work.
[edit] Timothy Manglicmot 21:38, 6 February 2007 (PST)
[edit] Reflective Physical Prototyping through Integrated Design, Test, and Analysis
From what I can tell the d.tool is a very useful program in coming up with a design. Its ability to rapidly create prototyping models gives the designer a lot more time to analyze the steps in the design process. I know that when I design a model I tend to abstain from prototyping because it takes up a lot of time; by having a fast, easy way to prototype (with the d.tool) I can focus on other aspects of designing including testing and debugging. Additionally, having the d.tool at a very early stage in the design process is invaluable because the user can identify problems a lot faster in the design process eliminating problems in the future. In my own personal experience, prototyping my design almost always reveals problems so by being able to identify problems at an early stage I, consequently, can save enormous amounts of time. The d.tool is a very helpful tool in the design process and can save the user a lot of time.
[edit] Authoring Sensor-based Interactions by Demonstration
The article discussed Exemplar, a rapid-prototyping tool, and how it can be used for “programming by demonstration.” This article was very technical and began by discussing how Exemplar allows the user to focus on the design process rather than “algorithm tinkering.” Because I have little programming experience I had a hard time following the article however I can definitely see its relevance to the design process. The Exemplar software is aimed at helping test prototypes and can be a very useful aide when coming up with the design, similar to the article described earlier.
[edit] Richard Chen 22:13, 6 February 2007 (PST)
“Reflective Physical Prototyping through Integrated Design, Test, and Analysis”
I found that the discussion on the d.tools toolkit to be very interesting because it once again reinforced the main principle of design—iteration. Having to brainstorm and then test that idea—while repeating the process over and over—is a testament to the principle that good designs are almost never completed on one sitting. This tool would definitely ease the process of constant revision and constant improvement, mainly due to the system’s ability to keep a very accurate and thorough record of all trial runs by the user. The d. tools system would truly expedite the process of allowing the designer to witness the test-runs by the user and is a very effective means for allowing the designer to detect which areas to improve upon in their design. After reading this paper, I was reminded of a previous homework reading about the importance of using interviews and studies to determine usability for a design. The previous readings mentioned techniques for interviewing that would bring out the most important information, but with the d. tools system, a lot of that time and energy would thereby be saved, allowing the designer to be able to spend more time improving their products..
“Authoring Sensor-based Interactions by Demonstration with Direct Manipulation and Pattern Recognition”
Here was another prototyping tool, similar to the one from the previous reading, called Exemplar. This tool would thus provide another means for the designer to see and understand what areas of the product would need improvement. What impressed me most about this system was that all participants successfully completed the first two tasks and build at least one game controller. The success of the system is further validated by the fact that many participants spent the majority of their time sketching and brainstorming design solutions—while testing and refining their design. The beauty of all this was that implementation time only took up less than a third of the total design time—a remarkable feat, I thought. I see this system as being an integral part of the iterative cycle of design because it truly expedites the process, allowing the participants to spend their time on the more useful act of thinking about the design, as opposed to tinkering around with the implementation.
[edit] Brian Loo 22:20, 6 February 2007 (PST)
“Reflective Physical Prototyping through Integrated Design, Test, and Analysis” Overall, this article was quite dry and boring. I had to have several sittings to get through the article as it highlighted how useful d.tools was for designers. Additionally, it kind of came off as a sales pitch or marketing brochure of the product at times. This article though does highlight lots of similarities to what we have been discussing in class about storyboarding, prototyping, testing, and feedback. D.tools seems like a very helpful tool for designers as it provides a more visual storyboard, simulation of a product, and very helpful user testing feature to analyze user interaction. Furthermore, the much technical details of this article of the d.tool features seem irrelevant since we cannot physically play with the program. If these technical descriptions were shortened and summarized, it would have made this read a lot more interesting and enjoyable. As a side note, it seems so strange that we keep discussing the idea of testing and prototyping and how it is crucial to the design stage. This seems like it should naturally been done from the start and not something this new for this designing industry. Doesn’t it just make sense to do it this way? Why would designers make gadgets and not have it designed to the user?
“Authoring Sensor Based Interactions Through Direct Manipulation and Pattern Matching” Side Question: why are we reading all this information from Stanford? Does UC Berkeley not take part in such studies of design? This article was also hard to follow due to the technical aspects and it failed to keep my attention and interest. Sensor data can be used in the designing process as it can be converted into interaction data. Exemplar, a sensor reading program, can be very helpful in designing products as it gives sensor data from interaction that can be analyzed and reviewed. For most of the article, I was very confused due to the high technical language and that the article never explains clearly these technical details for someone unfamiliar with computer science and technology.
[edit] Johannesleholm 23:03, 6 February 2007 (PST)
Reflective Physical Prototyping through Integrated Design, Test, and Analysis
d.tools looks like a very excellent tool for prototyping. I was impressed by all of the features that included – lots of things that I wouldn’t even expect from a program of this nature. This might be because I have never prototyped, but still this looks like an amazing product. I really like how easy it is to build your model with the tool. Whether the designer is testing a physical I/O device, or graphical interface, they can build it relatively easy with this program. I also like how it is built for the iterative process described at the beginning of the article: design, test, analyze, repeat.
Authoring Sensor Based Interactions Through Direct Manipulation and Pattern Matching
The tool described here, Exemplar, offers exciting new advances in sensor monitoring and computer logic design. Hartmann et al. explain that developing sensor based interaction requires three key steps: designing the hardware, creating application logic, then linking the two up. Exemplar helps the designer linking sensor data and application logic. I really thought this was interesting because I have some experience building robots and configuring the sensor data to application logic was somewhat difficult and very time consuming – like Hartmann said, it was a trail and error activity. I really like how he explored different type of signal data, binary, continuous, and discreet and how they are treated differently.
[edit] Christopher Ling 00:48, 7 February 2007 (PST)
"Reflective Physical Prototyping through Integrated Design, Test, and Analysis"
This technical design article discussed a design software, D.tools, which allows product designers to conveniently and spatially map out their prototypes and run simulations to facilitate the real-life use of the product. I feel that this software is extremely useful, but almost expected in today’s design environment, where successful product designs, such as the iPod, must have been continuously tested and altered before it’s final product.
“Authoring Sensor-based Interactions by Demonstration with Direct Manipulation and Pattern Recognition”
I felt this article was too technical for the average reader to understand, given by the terms used throughout the article, it seemed like you had to have used some similar program in the past or have programming experience. On another note, the article introduced a program called Exemplar, which points the user towards the actual design rather than the “algorithm tinkering”. I felt this is an effective method that provides a different perspective of seeing things. I never knew the complexity of design software used today.
[edit] Joshua Funamura 01:47, 7 February 2007 (PST)
"Reflective Physical Prototyping through Integrated Design, Test, and Analysis"
This paper explained the motivation for, and walked through the features of d.tools, a product design toolkit created to streamline the design for combination physical and digital products. It seems to be a pretty comprehensive solution, based on the author's description of all the hardware, programming interface, hardware modules, and especially the testing and video features. Everything seemed to fit logically into a system where iterative design is able to flow seamlessly through the different steps. I liked the fact that they made the program open source so others could build on and improve it.
"Authoring Sensor-based Interactions by Demonstration with Direct Manipulation and Pattern Recognition"
This paper explained the author's particular solution for a "programming by demonstration" (PBD) design system. Based on first-use survey results, it seems to be a pretty good system compared to the current competitors. PBD in general sounds like a neat concept, and brings to mind probably what the programmers for Nintendo Wii games go through to design the game controls, as the Wiimote has 3-axis accelerometers in both handsets, as well as an infrared sensor that also detects the orientation of the controller. Overall, Exemplar seems like a good development system where the designer, similar to in d.tools, doesn't need to be overly concerned with programming during the design phase. On a side note, it's interesting to see how well-designed systems can be when they're designed by design professionals for their peers.
[edit] Katherine Tong 07:08, 7 February 2007 (PST)
"Reflective Physical Prototyping through Integrated Design, Test, and Analysis"
Using d.tools as a easier way to understand one's prototype, seems simple and convenient enough. While the article does a good job explaining how d.tools is used, it seemed that one would have to be familiar with programming to use it. I wasn't exactly too sure on how clear the user interface would be. However, the idea behind the system and the main functions appear to be very user-friendly to those familiar with other types of programming.
"Authoring Sensor-based Interactions by Demonstration with Direct Manipulation and Pattern Recognition"
What I liked the best about this article was their overt descriptions of what maid their sensor-based interactions successful or not. In addition to being a step-by-step explanation on how to use the program by demonstration design, the article gives a very good reasoning for its choice of programming and various other abilities.
[edit] Ricky Surachman 11:26, 7 February 2007 (PST)
"Reflective Physical Prototyping through Integrated Design, Test, and Analysis" This article is mainly discussing about d.tools, a toolkit that simplify the product's protoype, using simple interface, in order to help users to test the protoytpe. This article also shows that d.tools will really become handy in modern era. But, although this article provides many interesting information about d.tools and how to implement it, I really don't like how it is written. There are too many technical terms which make this article difficult to be understood.
"Authoring Sensor-based Interactions by Demonstration with Direct Manipulation and Pattern Recognition" This article discusses about the use of sensor as a tool in interactive design. One of the problem with using sensor is that there is no definitive way to collect sensor data and connect it to the application logic, except the trial and error way. So, this article mostly discussing about a tool that help designer to sensor input data to application logic, which called Exemplar. I think that this article is have more technical terms than the previous article, which make this article is even more difficult to follow.
[edit] Rey Doctora 13:18, 7 February 2007 (PST)
"Reflective Physical Prototyping" This paper introduces us to d.tool which provides an alternative method to prototyping over the classical method of actually producing an expensive prototype to analyze a new product. I think the reading would have been more understandable if we were given the actual d.tool software to work with; without it, the reading was very technical and difficult to follow. Overall, however, I think the tool is nifty and clever. I particularly like how d.tool is hardware extensible and allows users to extend hardware architecture in three different ways: the hardware-to-PC interface, the hardware communication level, and at the electronic level. I also liked how d.tool enables analysis of a single user or multiple users who have used a product. This flexibility in analyzing results, I think, is one of d.tools strong points.
“Authoring Sensor Based Interactions Through Direct Manipulation and Pattern Matching “
Reading this paper was difficult to follow because it was dense in technical details and I lacked a physical Exemplar product to follow along with. Overall, however, my impression of the product is that Exemplar has a lot of promise, especially when it gives the user the power to focus on design rather than focus on algorithms. This feature again illustrates how the design of a product should always be concerned with the user and not on the product itself. One part of the article that caught my attention was the mention of the user’s gulf of execution. Exemplar is helping to lower this gulf of execution by actually carrying out what the user intends to do. This added benefit will, in the long run, empower Exemplar as a user-friendly data sensoring product.
[edit] YeeWan Cheung 13:22, 7 February 2007 (PST)
Reflective Physical Prototyping through Integrated Design, Test, and Analysis
The d.tools is a toolkit that helps us to apply the thinking-by-doing principal to our design project by integrating “design, test, and analysis for physical prototyping”. I was amazed by the flexibility that d.tools provides. It can be applying to both the early-stage prototyping and the higher-fidelity prototypes. The author mentions many advantages that d.tools provides, however, after I read the authors description of the way how d.tools works, I found it is rather complicated. I might be wrong on that since I haven’t used d.tools before. After all, I still think d.tools can make the design process becomes more efficient once the designers got used to the way d.tools works. Also, I really like the authors share the successes and shortcomings they found from their three studies.
Authoring Sensor Based Interactions Through Direct Manipulation and Pattern Matching
The techniques talks in this articles author sensor-based interactions by demonstration. Exemplar is a rapid prototyping tool, and its main feature is that it enables users to focus on design thinking rather than algorithm tinkering. The authors then suggest by using Exemplar, it can smaller the gulf of execution, which is the “gap between the users’ goals and the actions needed to attain those goals with the system”. I am excited to see the term gulf of execution because that’s what we talked about in class. Then the articles lead into much technical stuff. I like this article better than the first one because its analysis is more boarded. Besides, the 14 CDN dimensions that mentioned are very clearly written. However, in order to display multple sensor visualizations simultaneously, there is still a lot to do to improve the interface of Exemplar.
[edit] Asmita Karandikar 13:24, 7 February 2007 (PST)
“Reflective Physical Prototyping through Integrated Design, Test, and Analysis”
This article discusses the importance of prototyping and the toolkit d.tools. D.tools is a toolkit that can be used for a “design-centered approach to prototyping.” The article describes why d.tools is an effective tool because it allows much more interaction between the designer and the design. Also, d.tools is a low-fidelity alternative to traditional prototyping, which is often costly and is only created once. While I found the first half of the article to be interesting because it related to techniques that we have learned about in class, I did not enjoy the second half of the article. I feel that the second half is too technical and difficult for me to understand, as I have never heard of or have used d.tools before. However, I feel that is can be useful to read articles such as these because it gives us exposure to the different tools that are available to us during design.
“Authoring Sensor Based Interactions Through Direct Manipulation and Pattern Recognition”
This article focused on the design tool Exemplar, a rapid prototyping tool. The goal of this tool is to allow for programming by design, and should allow the user to concentrate on thinking about the design, rather than manipulating algorithms. Similarly to the previous article, I find that this article is too technical. I also find it to be less interesting than the previous one. However, it is nice to read that Exemplar is made so that it does not lead to a “gulf of execution” for the designer, since this was a topic that we previously discussed in class. While I agree that tools such as Exemplar are important in product design, it would be nice if they are presented in more interesting, engaging ways.
[edit] HAK-SEUNG KIM 13:28, 7 February 2007 (PST)
Reflective Physical Prototyping through Integrated Design, Test, and Analysis
The first article demonstrates about the prototyping tools. Especially, the author introduces one of the toolkit, d.tools. Toolkit support for designer thinking, and d.tools can integrate design, test, and analysis for physical prototyping. It seems interesting to me because I realized there are certain types of toolkit we can use for the specific physical prototyping throughout reading this article.
Authoring Sensor Based Interactions Through Direct Manipulation and Pattern Matching Next article seems little bit hard to understand because they use many technical words. I know the sensor, and the author tries to introduce how important sensors are. However, all terminologies that he uses are only for the people who are in the same area. Therefore, I do not really like this article. But I can recognize we use sensor everywhere in our lives such as sensor for the bus rear-door, sensor for the elevator, sensor for the saving lighting energy, etc. I hope there is much easier article to understand about the sensor.
[edit] Megan Whittey 13:29, 7 February 2007 (PST)
"Reflective Physical Prototyping through Integrated Design, Test, and Analysis"
This article was a bit hard to read but I understood most of it. The d.tools toolkit seemed like it would be any designers dream in developing a prototype, which we touched on in class and how important a prototype is to a design. I like the features d.tools included in their product. My favorite was that you could watch the videos that recorded the users using the product. You could either watch one user continuously in each stage of the using the product, or you could combine watching all the users using the product in a certain stage, which I thought was pretty cool. After reading about the product and seeing pictures of how it worked it really made me want to use it when I design something because I feel it would make the first few stages so much easier. I also liked how the d.tools used the java language to help program the product/prototype because I already have taken a java language programming class, so I feel I would be able to understand the d.tools product and write code even easier.
"Authoring Sensor Based Interactions Through Direct Manipulation and Pattern Recognition"
I found this article also hard to read and a bit harder to follow than the first article. With the first article, I feel like the pictures explained more than this article. Whenever I was a little bit confused in the first article I was able to understand alot better by looking at the pictures they had displayed. In this article the pictures didn't tell me as much. I wasn't use to all the technical language used in this paper either. I liked how this article was linked to the first article a little by saying the user could use the d.tools and java language with the Exemplar. I also was able to understand the article a little bit more once it described the experiment with the 12 people. I realized how the sensors were used with demonstration and the input went into the computer. The bike helmet, with the blinkers, helped me understand best out of the three different experiments because users would turn or lean back and forth to get data on the computer screen. Overall, I wouldn't want to read another article like this. I would want something a bit more easier to understand for people who aren't as involved in computer programming and designing.
[edit] Rabia Siddiqi 13:42, 7 February 2007 (PST)
Reflective Physical Prototyping through Integrated Design, Test, and Analysis
The article discusses why prototypes are essential in design and describes a toolkit called d.tools which is useful for effective prototyping. I liked the introduction section that described the importance of prototyping because it was interesting and easy to follow. I completely agree with Michael Barry's comment that "the companies that want to see the most models in the least time are the most design-sensitive; the companies that want that one perfect model are the least design sensitive." It is almost impossible to create a perfect model on the first try. The best way to determine the problems with any given design are to test the prototypes and improve on them. An iterative-design-centered approach allows the designer to continuously analyze and modify the design. Once the introduction ended and the author began describing d.tools, I felt that the article was harder to follow and the language was too technical. From what I could understand, I believe d.tools is definitely an asset for any designer. The integration of video and event logs is something a was most impressed with. The videos would allow designers to observe every detail of the users interaction with the design and make note of aspects they had trouble with. This way the shortcomings of the design can be determined early on. I believe d.tools is definitely something that I could benefit from if I want to test a design, but I would need more training than this article to fully understand how it works.
Authoring Sensor-based Interactions by Demonstration with Direct Manipulation and Pattern Recognition
This article is about the problem of specifying the relationship between sensory input and application logic and how the Exemplar program can be used as a solution. Exemplar uses sensors to help the designer determine the exact interaction between the user and the product. It helps the desinger focus on the design rather than the "algorithm tinkering," and improves prototype testing. A study was done to assess the effectiveness of Exemplar and from the results, it seems that it is a very powerful tool for collecting and interpreting sensory inputs, but it has some shortcomings that need to be addressed. The video game examples reminded me of the new Nintendo Wii, which translates the sensory inputs from hand motions into actions on the screen. Like the first article, this one was also loaded with technical details that made it difficult to fully understand the functions of Exemplar, but I do think it is innovative and can be valuable for designers.
[edit] Patrick Liu 14:25, 7 February 2007 (PST)
“Reflective Physical Prototyping through Integrated Design, Test, and Analysis”
This article interests me because of the way it explains design, prototyping, and evaluation. The main points of the article are how d.tools can be used for prototyping product design. The d.tool is complex in its entirety and requires a lot of technical background, but it is a very helpful tool for prototyping. D.tools have been used for many projects such as digital cameras, GPS systems, toys, etc. The author also explains how evaluation shortcomings allow the designer to discern what methods are successful or not. This allows the designer to design and build a better product. There are also tools other than d.tools that can be very helpful for product designs. Some techniques such as the Wizard of Oz techniques and simulations are also very helpful. It would be intriguing for me, if I could actually use these tools myself on a project of my own.
“Authoring Sensor-based Interactions by Demonstration with Direct Manipulation and Pattern Recognition”
This article details another prototyping tool called Exemplar. The pros and cons of Exemplar are discussed in this article and are compared to other prototyping devices. Exemplar is still under development, but it is essentially the utilization of hardware and sensors combined with software simulations. The tool allows designers more independence because of its versatility. Also the author mentions that the tool focuses more on the “design process” rather than the “algorithm thinking.” Overall I find exemplar to be a very effective tool, but I still kind of do not really understand how the tool can be used in all cases because the article was a bit difficult to read and fully understand.
[edit] Michael Lovejoy 14:37, 7 February 2007 (PST)
“Reflective Physical Prototyping through Integrated Design, Test, and Analysis”
This article talks about d.tools in a way that’s hard to understand. It talks about iteration, design, test, and analysis as important aspects of information appliances. D.tools records users’ interactions with the device, so it is a useful tool for designing products. Beyond this, the article is too abstract and confusing. While I’m sure it has good ideas, it is written in a difficult manner; the fact that the subject matter is pretty complicated (though the article makes it more complicated than it has to be) does not help in understanding the subject. The many examples would have been more useful if they did not demonstrate very technical ideas, but they helped give the subject some concrete applications. Overall, the article wasn’t so great.
“Authoring Sensor Based Interactions Through Direct Manipulation and Pattern Recognition”
This article is about sensors and how to use sensor-based interaction in designs. While the concept is pretty interesting, the article is still too technical to be of much use to me. Using sensors can be really cool, as seen from the video on the first day of class, but the ideas of how to implement them are the important parts I care about. Basically, it involves creating the hardware, creating the software, and creating a relationship between those two. Although it used to simply be trial-and-error to link sensor input data to application logic, there now better ways to accomplish that. Exemplar is a program that helps designers focus on design and not get bogged down in annoying details. The article was pretty complicated, and sort of killed any enthusiasm I may have had for sensors before. I didn’t like the article, and I believe that it was too technical for the concepts we’re learning. It didn’t help me understand design any better.
[edit] ChuiShan Wong 14:42, 7 February 2007 (PST)
“Reflective Physical Prototyping through Integrated Design, Test, and Analysis”
After reading this article, I have a deep respect to old desinger. D.tool is a very useful software that we can use in the early design process. With the help of D.tool, designers will know that if their designs are practical or not. This is very important because it allows deisgners to save more time to work on other part of their designs. I can’t imagine how difficult it is for those old designers to finish their work without the help of those useful hardware and software that we have today.
“Authoring Sensor-based Interactions by Demonstration”
This is another article introducing a design tool. Exemplar is software allowing designer to focus on design thinking. This article is very difficult for me becaue it contains a lot of specific computer languages which are new to me since I am not familiar with computer programming.
[edit] Nicolas Suryono 14:46, 7 February 2007 (PST)
“Reflective Physical Prototyping through Integrated Design, Test, and Analysis”
The articles mainly talks about d.tools using a complicated and complex technical language. The d.tools records user's interaction with the device which will then allow the user to monitor what has been done and gain several useful information from it. Even though it has some good examples, I think this articles has too many complicated technological terms that does not really explain to the readers what design is all about.
“Authoring Sensor Based Interactions Through Direct Manipulation and Pattern Recognition”
I think this article is pretty similar to the previous articles. The author uses too many complicated technical language which makes the articles challenging to read. It has some pretty interesting concept about sensors. I took a robotics class and learned about sensors but I have never thought that sensors can also be used in the design. The Exemplar program example is a pretty useful aid for designers to focus on their ideas.
[edit] Karey Park 14:52, 7 February 2007 (PST)
“Reflective Physical Prototyping through Integrated Design, Test, and Analysis”
This article introduces d.tools to a user who may not necessarily know everything that there is to know about prototype development, apart from the fact that it is a pivotal step. Prototyping is the primary method by which designers innovate, test their designs, and collaborate with others on them. That said, the primary purpose of d.tools is to make this task a little bit less daunting. One thing that appealed to me about d.tools is the design of d.tools itself, ironically. It is designers' software, in the purest sense of the word. Everything supports design-oriented thinking, treating the designers as users themselves. Even designers are users themselves, users of other products by which many better ones are developed.
“Authoring Sensor Based Interactions Through Direct Manipulation and Pattern Recognition”
This paper was a pretty formal report, introducing techniques for sensor-based interaction. However, it seemed to be overly ambitious. At the very least, it proves that sensor technology is not easy stuff to be taken lightly; it is a technology that is growing because of its diversity and economic appeal. Perhaps the author would have generated more interest in would-be sensor technology engineers if had gone down this alley of economic use and diversity rather than a demonstration of sensor-based interactions, a demonstration not without its myriad of esoteric techniques.
[edit] Hong How Quek 14:53, 7 February 2007 (PST)
“Reflective Physical Prototyping through Integrated Design, Test, and Analysis”
The article talks about the motivations behind their research efforts, before delving deep into the key techniques for building, testing and analyzing prototypes that d.tools can offer. Finally, the evaluation strategies were described and the findings (in particular, d.tools' shortcomings) were discussed. Before reading the article, I always thought of prototyping as a costly approach that is not replicated very often, but with the high level of technical support offered by d.tools, designers now have a low-fidelity alternative for prototyping! I liked the fact the d.tools could monitor users toying with the prototype via video (or even watch different users at the same activity simulatneously), as it gives an immediate feedback to the designer on how effective his/her prototype is. I found most of the article very technical and difficult to follow, especially if one does not have the program to refer to while reading the article's highly technical and complex descriptions. I am looking forward to Hartmann's guest lecture, which will hopefully shed some light to what seems like a really intriguing prototyping tool that is difficult to comprehend fully by the amateur prototyper (me).
“Authoring Sensor Based Interactions Through Direct Manipulation and Pattern Recognition”
Once again, a very technical article (although less so compared to the first one) that introduces techniques for authoring sensor-based interactions by demonstrations. This time around, the prototyping tool Exemplar is introduced. My major takeaway from this article is a newfound understanding of the difficulty making sense of continuous signals (often numbers) often recorded my sensors as the relationship between performed action and reported values is not visually apparent. Hence there is a gulf of execution here that needs to be bridged, and Exemplar tries to do just that. It will be interesting to hear from the guest speaker on what other technological advances have been made in the field of linking sensor input data to application logic.
[edit] Royapakzad 14:57, 7 February 2007 (PST)
“Reflective Physical Prototyping through Integrated Design, Test, and Analysis”
This article discussed prototyping through integrated design and it also introduced d.tools which is a design tool that embodies an iterative-design-centered approach to prototyping and it integrates support for prototype testing and analysis into the workflow. I was intrigued by the smart components of d.tools that cover a wide range of input and output technologies, especially the test mode, in which d.tools records live video and audio of user interaction to capture the user’s quotes and opinion about the usability of the prototype. I think utilizing d.tools in any design process will enable designers to reach their goal in a more time-efficient manner and will help them to end up with more successful results.
"Authoring Sensor-based Interactions by Demonstration with Direct Manipulation and Pattern Recognition"
This article contributes techniques for enabling a wider audience of designers and application programmers to turn raw sensor data into useful events for interaction design through programming by demonstration. The termanology of the article was complicated, especially "programming by demonstration". There is no concrete explanation or definion of that term. However, the introduction of Exemplar,(a tool that frames the desing of sensor-based interactions as the activity of performing the actions that the sensor should recognize), was interesting. Overal, it was a highly technical paper and difficult for students with no computer programming backround to follow.
[edit] James ChunJye Tong 15:00, 7 February 2007 (PST)
"Reflective Physical Prototyping through Integrated Design, Test, and Analysis"
This article discusses d.tools, a tool that would create a prototyping environment during the designing process. Even though I don’t think the idea is a innovative one, and I believe there has already been people using similar tool, however, I like some of the features that presented in this article. For example, I believe that the single user analysis and group user analysis will help the designer get a better understanding about what’s good and bad about their design. Again, I think this is an important concept that has already existed about designing new product. This article is very technical and is pretty hard to read for readers who have no background about designing the products that the author mentioned in the article.
"Authoring Sensor Based Interactions Through Direct Manipulation and Pattern Matching"
This article is also very technical. This article introduces new method of sensor-based interaction designs. The article has discussed about many different tools that can help the designer design such products in a more efficient way. Even though this article is also very difficult to understand for people who don’t have such background, I found the user-study pretty interesting. It is interesting to read about how such tool would make impact to the users in the study.
[edit] Jason Hu 15:01, 7 February 2007 (PST)
"Reflective Physical Prototyping through Integrated Design, Test, and Analysis"
This was basically an advertisement for d.tools, a design software whose User Interface (UI) prototyping features are comprehensively explained in this report. In an unexpectedly Meta development, the text eventually turned to a detailed, technical explanation of the process by which the prototyping software’s own usability was tested and refined under various conditions. Their verdict was, of course, positive. The d.tools software seems like a useful addition to any UI designer’s kit. It certainly appears to beat sheets of paper, transparencies, or Photoshop layers in terms of accurately portraying the intended user experience. However, those rougher, analog tools are probably still useful in a early prototyping stage, and d.tools is probably most appropriate for a more intermediate prototyping stage.
"Authoring Sensor Based Interactions Through Direct Manipulation and Pattern Matching"
This was the most densely technical article we have read thus far. Like the preceeding one, it was also a thorough analysis of the software, Exemplar. This software was intended to build upon the easy prototyping features of d.tools by including sensor-based signal processing, as well as to integrate the prototype testing/refinement process better. The ability to include real physical knobs, dials, switches, etc into the prototype interface seems extremely useful. Also, the ability to synchronize camera footage of your user testing with the exact flow of their UI experience would provide telling evidence could aid in the design iteration process, provided that you had someone willing and able to take the right lessons from such evidence.
[edit] Sumeet Patel 15:08, 7 February 2007 (PST)
“Reflective Physical Prototyping through Integrated Design, Test, and Analysis”
This article talked about how to use d.tools in order to build a prototype. I found this to be an interesting approach to testing the prototype because you are using a software in d.tools to test a physical device. D.tools seems a bit complicated to use because you need to know some java and a little bit of circuitry to make it most useful. Overall the article was a bit hard to follow because it was set up like a users manual. I felt as though they were trying to sell this product more than describe a process.
“Authoring Sensor Based Interactions Through Direct Manipulation and Pattern Matching”
Like the previous article, this article talks about another specific technology called Exemplar. This software connects the sensors to the actual prototype and is a useful tool to see how the prototype actually works. These different ways to evaluate the prototype are important in refining and making the final product. Exemplar makes the refinement process easier because the designers can focus on the design and are not bogged down with programming languages and algorithms.
[edit] Anar Joshi 15:09, 7 February 2007 (PST)
“Reflective Physical Prototyping through Integrated Design”: This article described a toolkit, d.tools, that can be used by designers to effectively prototype their appliances. D.tools solves one of the most important factors of the design process, prototyping, which enables designers to learn about the problem that they are trying to solve. The article not only outlined the various features of the software, but also talked about the iterative process that they went through while refining the product. For example, after an initial test with a user group that had general design knowledge, they learned that d.tools was especially useful in enabling usability testing, shortening the time required to build a prototype, and helping to understand user experience. At the same time they also learned that the software lacked a simulation where the designer could interact with the model, and added this feature in its next iteration.
“Authoring Sensor-based Interactions by Demonstration with Direct Manipulation and Pattern Recognition”: The main problem for designers is specifying the relationship between sensor input and application logic. The article introduced a tool, Exemplar, to resolve this problem for designers. Exemplar helps designers to focus on the sensor-based interaction by graphically demonstrating the action and then allowing the designer to edit the model to better fit the user’s needs. When testing Exemplar, the group used user groups as well as a framework that is used to evaluate visual programming languages called Cognitive Dimension of Notation (CDN). I have never heard of CDN before, but thought that it was a valuable performance metric since it provided a quantitative analysis.
[edit] Minhaaj Khan 15:13, 7 February 2007 (PST)
Reflective Physical Prototyping through Integrated Design, Test, and Analysis
This article spent almost all its entirety talking about d.tools, a powerful design program that allows designers access to a variety of design tools that are necessary in the design process. It stressed the iterative process and d.tools allows for rapid iteration in the design process that wouldn’t be possible otherwise. The article walked us through how d.tools functions during the design process and how it gives designer access to tools that they otherwise wouldn’t possess without a computer programmer. I also liked how the program itself went through iterative design during creation and how the authors discussed this in the paper (e.g. using users to find faults like the fact that they added the software simulation mode into d.tools because test users felt the evaluated version didn’t provide for stepping through an interaction without attached hardware).
Authoring Sensor-based Interactions by Demonstration
I didn’t like this article as much as the previous one, although similar to the last one, this article stresses rapid iterative design through computer applications (e.g. Exemplar). It discussed sensor-based introduction and the inherent difficulty of specifying the relationship between sensor values and application logic. The authors claimed this was due to the fact that 1) current tools use textual programming to author sensor-based behaviors without visual representations 2) the gap between their goals (support of interaction design) and the actions needed to attain those goals 3) lack of tools for the process. This was somewhat dense reading and I was certainly bored.
[edit] Yang Hung 15:20, 7 February 2007 (PST)
"Reflective Physical Prototyping"
This paper talks about d.tools, a "toolkit that embodies an iterative-design-centered approach to prototyping information appliances," such as cell phones, digital cameras, and mp3 players. d.tools integrates the design, test, and analysis for physical prototyping and can save a designer not just materials cost but also time costs as well. Despite all the benefits mentioned in the paper, one problem that I see with d.tools is the fact that a designer may not be versed in programming, thus it would take a lot of time to train a designer with no formal technical training. It seems as though this toolkit is only intended for designers with a high tech background, such as a designer at Apple. In the future, I recommend that the author come up with a less technical version of d.tools to open the market up to all designers.
"Authoring Sensor Based Interactions Through Direct Manipulation and Pattern Matching" This article was very technical and hard to understand, I had to read the abstract at least three times to get the focus of this article. It seemed like one needed to actually use Exemplar to understand all of the dense terms. However, my impression is that having a tool that helps improve the linking of sensor input data to application logic would shorten the design cycle and thus introduce new efficiencies into the whole design process. I would recommend this reading for my EECS friends.
[edit] Luciana Cook 15:22, 7 February 2007 (PST)
“Reflective Physical Prototyping through Integrated Design, Test, and Analysis” I think that d.tools is a good way to use your time efficiently. I am sure that without it engineers would have to spend a lot more time doing prototypes on their own. I also think that although there are limits to d.tools the article does a good job explaining on how to work around these limitations in order to still be able to use this tool efficiently. I think that for any tool we must really know how it can hurt us in the long or short run in order to be able to use it to its highest capabilities.
“Authoring Sensor-based Interactions by Demonstration”
I felt that this article was confusing and hard to follow. Having no computer science experience and no programming background I felt lost a lot of the times. I think it took long to get to the point because it would go out on tangents. Aside from that I was able to capture that it was talking about a lot of the same concepts from the first reading such as prototyping except that I felt that the focus was a bit different. It was more about how the interaction should work.
[edit] Connie Ko 15:30, 7 February 2007 (PST)
“Reflective Physical Prototyping through Integrated Design, Test, and Analysis” This article just explains in detail the functionality of this one software product called d.tools and how it can help design, test, and analyze a potential product design. It is definitely an innovative product that can help make the designing process a lot more efficient. I thought this article was a bit too technical for me, and it was very hard to follow because of the breaks in between sentences due to notes in parenthesis.
“Authoring Sensor-based Interactions by Demonstration”
The prototyping tool, Exemplar, allows designers to program visually rather than through traditional programming techniques. This increases the user base because more people can understand it. This article is similar to the first in that it also introduces software that can ultimately help designers design products more efficiently and effectively. It was once again a very technical article that was a bit hard to follow.
[edit] Michael Toulouse 15:31, 7 February 2007 (PST)
“Reflective Physical Prototyping through Integrated Design, Test, and Analysis”
This was not so much an article as it was a short research paper, and it was also about a piece of software largely foreign to me. As such, it was almost completely opaque to me. From what I can gather, it greatly advocated the use of a program called "d.tools" that was excellent for creating user interfaces for consumer electronics. Mentions were made of different projects that d.tools had helped with, and the ease with which new functions were made that could interface with existing UI's to improve them, like Google Earth. Overall, this was not a good paper to read without some better background.
“Authoring Sensor-based Interactions by Demonstration”
This second research paper, though also somewhat hard to read, introduced the concept of programming by demonstration, which involves translating user feedback into code, from what i understand. It goes on to explain how sensor signals are gathered/interpreted, and finally gets to its main point, the use of a program called Exemplar. Exemplar seemed to me like a simplified, more user friendly version of Labview geared specifically toward processing and interpreting sensor signals. I think this is good, as sometimes all the filters and algorithms can get a little opaque. The paper evaluated the program for its pros and cons, suggesting future areas of work and where it was best and worst. This paper would be especially valuable for anyone who decided to improve on this open source program.
[edit] Aubuchon 15:42, 7 February 2007 (PST)
“Reflective Physical Prototyping through Integrated Design, Test, and Analysis”
This article introduces a method to design and prototype information appliances. It also introduces “d.tools”, which is a software package to help designers physically prototype. I was very interested that these tools include things like the ability to record the user as he/she uses a prototype. This lets designers see how people react to the design, what they say, and how to appear. This seems to me to be helpful towards the goal of effective contextual inquiry. If one can observe the subject without even physically being present during the experiment, this helps the promote natural behavior during testing, which is a very important part of inquiry.
Overall, the article seemed to just repeat concepts that we have learned such as prototyping, testing, data gathering, adjusting, and iteratively improving the design of a product.
“Authoring Sensor-based Interactions by Demonstration”
This paper introduces techniques for authoring sensor-based interactions by demonstration. This article introduces software called Exemplar, which emphasizes that programming by demonstration can help with testing and prototyping of a design. This article was very technical and seemed to introduce a piece of technology as opposed to describe important design concepts. I did not feel I learned much from this article.
[edit] Shilpi Verma 15:43, 7 February 2007 (PST)
Reflective Physical Prototyping through Integrated Design, Test, and Analysis: This article is about the importance of prototyping and its role in the media that structure innovation, collaboration, and creativity design. The article also talks about d.tools, which is a toolkit that is considered to be a “design-centered approach to prototyping.” D.tools can also be considered as another alternative to prototyping, since prototyping can end up being so expensive. I think it is really interesting that d.tools have been used in various devices such as digital cameras and GPS systems. D.tools also allows for analysis of single or multiple users who are using the product. Even though the article was a little difficult to read, the examples were really useful in understanding the subject
Authoring Sensor Based Interactions through Direct Manipulation and Pattern Recognition: This article is mainly about sensors while also focusing on Exemplar, a rapid prototyping tool. Exemplar allows for programming by design, and shows the user what areas of a product needs improvement. Some parts in this article about sensors reminded me a little of the video we watched in the first week of class. Overall this article was really technical, which made it a pretty difficult read. The concepts discussed were interesting and are useful for people performing tests on a prototype/product while in its design phase.
[edit] Shwan Kazzaz 15:44, 7 February 2007 (PST)
Reflective Physical Prototyping through Integrated Design, Test, and Analysis:
I think that d.tools can be a very useful program, however, one must keep a few things in mind. First of all, there is a distinct beauty to being able to do things with your hands, using a pencil on paper makes things flow easily. As people are very used to that type of medium. However, d.tools will need to be sufficiently easy to use, or the user must be sufficiently adept with the program, for the design to be created quickly and effortlessly. Otherwise good ideas could easily go unnoticed as the user is lost in the program, rather than the design experience. Some of the test and analysis features, however, do look very useful. I think a video, with land marks, could be invaluable to evaluating a user's experience, it leaves the user to work alone, without the watchful eye of the designer. It makes the user act more natural and not feel watched. The thing that makes d.tools useful is that it timestamps the video according to where the user is, in the process. This allows easier browsing and analysis for the designer.
Authoring Sensor-based Interactions by Demonstration with direct Manipulation and Pattern Recognition:
I didn't really understand most of this article, as there was a lot of jargon that I was not familiar with. I don't feel that they made the concept of sensors entirely clear. And I was confused by what it meant to program by demonstration. From what I understand it is some sort of non-textual based user interface that programs for you. The article later discusses the architecture and implementation of Exemplar, most of this part of the article did not seem important to me. I don't think the article was entirely intended to be read by someone with as little experience in the field of sensing technology as myself.
[edit] Rakesh Vij 16:25, 7 February 2007 (PST)
Reflective Physical Prototyping through Integrated Design, Test, and Analysis: This article focuses on the use of iteration as the core during prototyping. The article discussed the use of an program called d.tools which aids designers in the design process. Specifically this program helps by mapping out prototypes for designers. This is a very important step because it is through prototyping that designers are fully able to understand the problem that their device will solve. The protoyping phase can be very expensive and costly to designers, thus it is of utmost importance to streamline this process. The product helps setup usability studies, shortens time to build a prototype, and helps understand the user experience during testing. The software also now has simulation that also helps in the design process. All in all the iterative process involving protoypes is very helpful to designers but very costly, d.tools will help make the process better.
Authoring Sensor-based Interactions by Demonstration with Direct Manipulation and Pattern Recognition: Although hard to read, this article was actually interesting. It discusses the increasing use of sensors in interaction design. Within this type of design there are three main steps. choosing and connecting the hardware, creating application hardware, and specifying the relationship between censor values and application logic. Now the first two pieces of the puzzle have been solved, but the third piece is the complicated one. In this paper there are a few different approaches to this problem. One is to use demonstartion to author it. Another approach emphasizes the use of very rapid iterative demostration-edit-review cycles. The reading also mentions the use of Exemplar, a software which helps designers correlate action to the sensors. This software helps designers avoid algorithm tinerking by using a logical process to solve the issue of sensor authoring. Before reading this article I had no idea what sensor-based interactions even were, but i'm really glad I took the time to learn about this interesting process.
[edit] Umber Masood 16:03, 7 February 2007 (PST)
“Reflective Physical Prototyping through Integrated Design, Test, and Analysis”
d.tools seems to be a very powerful and useful tool. I did not know that d.tools could be used to make prototypes of various electronic devices, from music players to digital cameras. I agree with the authors’ statement that iteration is an important part of design. Only when constant revisions are made can errors be found and changed. It is similar to the editing process of a written report: multiple drafts are written before a final version is made. I thought the evaluation stage of the study was interesting. One thing I noticed in the reading was that all three groups that were part of the evaluation stage were designers or design students. It would have been interesting if there had been a group of people who did not have a background in design be a part of the evaluation process as well.
“Authoring Sensor Based Interactions through Direct Manipulation and Pattern Matching”
This reading was interesting. I did not know that sensors could be used in design. This reading was about a different design tool than what the previous reading was about: the authors of this paper used Exemplar, whereas the authors of the previous paper used d.tools. The aspect of Exemplar that it outputs both continuous data and discrete events seems to make it very useful. In the first study done for this report, there was a wide range in the backgrounds of people in the study group: the group included those with engineering, computer science, education, and humanities backgrounds. I think having a study group of people with different backgrounds helps in evaluation because each person has a different point of view.
[edit] Ryan Panchadsaram 16:06, 7 February 2007 (PST)
Reflective Physical Prototyping through Integrated Design, Test, and Analysis
We learned about the product design cycle last week and one of the key concepts is iteration. It is interesting to read about the programming tool d.tools because it is a design tool build around iteration. The only tools I have used for prototyping are Adobe Photoshop, Adobe Flash, and traditional pen and paper; seeing this tool makes the others look “basic”. The unique thing about d.tools is that it allows the designers to work in an authoring environment that has visual representations, hardware interface, and straight code. Allowing designers to work quicker is important because it should, hopefully, lead to better work. My only concern, which is shared by one of the participants, is that the prototypes are not polished. The nice thing about building prototypes is that they convey a message, but also hold the image of the final product.
Authoring Sensor Based Interactions Through Direct Manipulation and Pattern Matching
Exemplar is an interesting way to calibrate and manipulate sensors because it chooses to use programming by demonstration. I think this is a novel idea because rather than starting at the code, you analyze and interact with the direct input from the device. I can see this speeding up the process and also being able to “visualize” the incoming signals better. Interacting with visuals is much easier than writing tedious code that tries to manipulate analog and digital signals.
I think both articles were interesting to read. I especially took note of the format that they were written in. They were like formal papers about topics, except they were showcasing a product that they developed. Each paper had a section dedicated to the product description, typical user scenario, architecture, first-use study, and evaluation.
[edit] Justin Hsu 16:14, 7 February 2007 (PST)
“Reflective Physical Prototyping through Integrated Design, Test, and Analysis”
The article "Reflective Physical Prototyping through Integrated Design, Test, and Analysis" introduces d.tools, a software/hardware tool that is able to help designers easily construct prototypes of their product. This article mainly discusses the benefits and shortcomings of d.tools. In doing so, the authors describe the importance of iteration in the product development process – a theme that we have previously discussed. I did not particularly like this article, not because it was not well written, but because it really did not help my understanding of design principles. While it did touch briefly on some important themes in design like iteration, these instances were mostly minor explanations used to support points about the functionality of d.tools. There were times I found it difficult to visualize some of the d.tools issues the authors were talking about, simply because I have no experience with it. In conclusion, when the authors described the reasons behind the effectiveness/shortcomings of d.tools by describing design principles, I enjoyed it. However, I believe that too much of the article was based on technical and experimental explanations of d.tools, which is largely irrelevant in the context of this class.
“Authoring Sensor Based Interactions through Direct Manipulation and Pattern Matching”
The article “Authoring Sensor Based Interactions through Direct Manipulation and Pattern Matching” describes what is involved in the linking of sensor data to application algorithms. The paper then proceeds to explain some of the experiments used to study and analyze sensor-based interactions. While this paper was well written as far as research papers go, it suffers from many of the same issues in the previous article. The article uses extremely technical and experimental language to describe how Exemplar was used to study pattern recognition techniques by analyzing the iterative demonstrate-edit-review cycles pictured in Figure 1, but not enough of the article discussed fundamental design principles. I did not particularly enjoy this article because it lacked relevance to this class in the sense that it did not significantly help me understand design principles. The instances where the article talked about design principles were only used in the context of Exemplar.
[edit] Fenelia Kosasih 16:16, 7 February 2007 (PST)
"Relative Physical Prototyping through Integrated Design, Test and Analysis"
This article is a technical paper about the development of d.tools, a program which allows fast prototyping and hence lessen the time needed for the design process. The program simulates the products' behaviors so that the designers can test out their design, evaluate and analyze their design. With d.tools, the designers might be able to bypass the need for interviewing and testing their products on user, as has been recommended in our previous readings.
"Authoring Sensor-based Interactions by Demonstration with Direct Manipulation and Pattern Recognition"
This paper focuses on techniques for authoring sensor-based interactions by demonstration, also called Programming by Demonstration (PBD). The system introduced here, Examplar, is another prototyping tool which helps the designer to evaluate their design to see which areas can be improved. The tool encourages the designer to give more thought to their design rather than focusing too much on the algorithm behind the design. Overall, the article definitely introduces interesting and innovative idea, however, I find the paper too technical to be enjoyable.
[edit] Marshall Anderson 16:17, 7 February 2007 (PST)
d.tools This paper was an interesting read that got me thinking. This d.tools program sounds like it is the wave of the future for designing use interfaces. d.tools essentially allows the designer to nail down the user interface without investing huge amounts of time and money. The whole idea of refining a system image that is consistent with the user's conceptual model becomes easier to accomplish with d.tools. This software is reminiscent of the Kodak User Experience prototype. By that I mean, d.tools seems to be a universal way to study interfaces without overwhelming concern for the physical form.
Exemplar This is another handy program that I am going to have to try out. The ability to take raw sensor signal data and condition it to create consistent event inputs without actually worrying about the technicalities of the signal is powerful. Both of these programs make me wonder about the future of design. Exempar in particular seem to somewhat automate the design process of sensory control. It will be interesting to see how far automated design will be able to go...will it ever replace human imagination?
[edit] Luna Dai 16:30, 7 February 2007 (PST)
"Reflective Physical Prototyping through Integrated Design, Test, and Analysis"
This article goes through how a designer can use the d.tools program to create quick prototypes that have both the functional and form aspects of the product design. Many product designers have only an "exposure" and not "fluency" in programming, so current prototyping has become more about implementing the functions of the product through computer programming instead of working through the design of the product. This then allows designers to get fast user feedback on their designs without spending a long time to create the prototype for testing. I found d.tools use of statecharts to show the transitions between the graphical states that the user sees in sequence interesting. I think that just as important as the design of each graphical state is the transition between one and the next. In order to create a logical user interface, the transition must be intuitive and contextual.
"Authoring Sensor-based Interactions by Demonstration with Direct Manipulation and Pattern Recognition"
This article introduces Exemplar as a program to facilitate design of software utilizing sensor-based interactions. It goes through the benefits of programming by demonstration, in which the designer can give a physical example of an action on the sensor and then edit the resulting signals. This speeds up the process of designing for sensor-based interactions and allows for, as with d.tools, faster prototyping and more user testing. I found it interesting that the paper sites a test trial in which designers used Exemplar to design game controllers. The conventional game controller uses buttons and so has not need for sensors. Recently, however, with the Nintendo Wii, designers have found the increased possibilities from a sensor-based game controller. It would be interesting to see what other consumer products will use sensors in the future.
[edit] Patricia Wang 16:34, 7 February 2007 (PST)
"Reflective Physical Prototyping through Integrated Design, Test, and Analysis"
This article details the features of d.tools, a toolkit that enables designers to iteratively develop prototypes. The author describes d.tools to make three major contributions to the design process. Its statechart-based prototyping model is an extension of the traditional storyboard-driving design method and guides the designer through the early stages of prototyping. Additionally, d.tools offers the hardware architecture for the purpose of physical interfaces. The last feature facilitates the integration of the design stage, testing, and analysis. All in all, d.tools cover all the issues that arise in the design.
“Authoring Sensor-based Interactions by Demonstration with Direct Manipulation and Pattern Recognition”
I thought that this article was really technical and sometimes hard to follow, since I was unfamiliar with a lot of the terminology. From what I got out of it, this article introduced the “programming by demonstration” design system. This sort of system uses Exemplar, a rapid prototyping tool, to incorporate and manage sensors within a design.
[edit] Raymond Kim 16:34, 7 February 2007 (PST)
"Relative Physical Prototyping through Integrated Design, Test and Analysis"
In this article d.tools, a program used to design and test prototypes was introduced. It is able to replicate anything from electronic devices to kitchenware and analyze various individual aspects of the design to uncover any flaws. Throughout the article the author continually stresses the need to iterate and reiterate product simulations in order to discover their flaws, which is something that just can't be done by normal user observation and testing. Overall, I found this to be a fairly interesting reading.
"Authoring Sensor-based Interactions by Demonstration with Direct Manipulation and Pattern Recognition"
This paper discusses sensor interactions using programming by demonstration. The author uses Exemplar as an example and goes into a deep discussion of its design and purpose. This was a fairly interesting read since it gives insight into sensors, a crucial and often overlooked part of many products. The style of writing however, made the article a bit difficult to read. A more tech-savvy person working in this field would probably appreciate this paper more.
[edit] Ryan Jue 16:37, 7 February 2007 (PST)
Reflective Physical Prototyping through Integrated Design, Test, and Analysis
This article mainly talked about ways to use d.tools, a prototyping software, to design new products. While seemed like a good way to prototype different designs relatively quickly, I felt that it was also very limiting. The design would have to based around objects already programmed into the software and more innovative designs wouldn’t be possible. Also, I felt that this wasn’t a typical prototype given that it was all digital. It only shows a demo of the product. I feel that even for hardware design, which is probably the focus of d.tools, a hands-on prototype is a lot better. Most of the design is the usability of the product and d.tools didn’t seem to provide that, just a way to demo the user interface.
Authoring Sensor-based Interactions by Demonstration
This was another article on a software implementation of prototyping like Exemplar. This was a much more technical article and dealt with the use of sensors in rapid prototyping. Even as an IEOR and EECS double major, the component parts got a little complicated. However, the gist was the same, software implementations can improve the speed and efficiency of prototyping.