Input - Models and Techniques

From CS2610 Fall 2014
Jump to: navigation, search

Contents

Slides

Slides

Readings

Optional

Reading Critiques

Wei Guo 14:29:49 9/17/2014

Reading Critique for A Morphological Analysis of the Design Space of Input Devices Wei Guo This paper offers the fourth line to systematizing input devices: morphological design study, besides toolkits, taxonomies, and performance studies. The method they use is generating points in design space. There are two basic criteria to evaluate the mappings implied by input device, which are expressiveness, and effectiveness. Expressiveness criteria focuses on exactly express of the intended meaning, while effectiveness criteria focuses on the measuring of performance. Footprint, bandwidth and so on are all the figures of effectiveness. Xbox Kinect is one of the popular input device nowadays. It is a motion sensing input device, which allows users to control and interact with their console without the need for a game controller. User can simply use gestures and natural language command. In this case, the footprint of Kinect can be very large. The human part of bandwidth is also restricted by human limitations. The application part of bandwidth is not that harsh. Since it is most using for games, and the precise is not a very critical problem. However, the user can be very tired after playing with Kinect. Since I haven’t tried it yet, I really want to know can the Kinect sense the 3D motion. Can it not only detect up-down, left-right motion, but also near-far motion?   Reading Critique for Multi-Touch Systems that I Have Known and Loved Wei Guo The author introduces a list of touch and multi-touch system features, and then goes over the history of multi-touch system advances. The author divides the touch-screen technique in different ways, and clearly points out the advantages and disadvantages for each category in each way. The author then lists some related work to multi-touch by chronology. From the listed works, we can see that the technique is switching from the mainly touch pad, to mainly touch screen. There is an interesting movie call The Iron Man. In this movie, there is a technique that the actor can create electronic model by moving, waiving hands, and touching on air… There is no screen, everything is 3D holographic imaging. To me, it seems hard to approach to 3D with a 2D screen. Each step needs endless rotation… I am interested in such technique like the one in the movie. In this case, each point of the space can be the touching point. This is definitely a multi-touch system. Is there any such techniques so far? These are not showing in this paper.

Yanbing Xue 14:35:27 9/17/2014

The first paper is mainly about classificatios of input devices by their shape, form, and functionality. Some systematic processes already exist, like toolkits, taxonomies, and performance studies. This paper first introduces a way to characterize the physical properties of input devices by their position, force and linear and rotary position. The authors address their ideas from the aspects of the basic concept of input device, which allows a portion of communicative sentences to be realized and communicated from human to machine. The design space for input devices is basically the set of possible combinations of the composition operators with the primitive vocabulary. While they can be plotted on the taxonomy, the authors provided a method to help generate points in the design space. Several devices including mouse and head mouse are evaluated and compared with the neck, arm, wrist and fingers in terms of footprint, bandwidth and device bandwidth. Applying these criteria to compare the regular mouse and head mouse, it explains why the head mouse, despite of having impressive transducer, cannot succeed in text editing. I see the most contribution of the paper is providing a graphical and systematical method to test and compare different input device on the same dimensions like pointing speed, pointing precision, errors. All properties of input devices as well as task are encoded into coordinates of design space which are 2D representation of higher dimension. The primitive dimensions of design space are defined by manipulation operations, and each device and/or task are placed on appropriate area given it support operations. The paper provided interesting example about presenting pointing precision on design space. While the graphical representation has certain advantages, I see the proposed model is still simple as it use relative positions and areas of points in design space to compare between devices and tasks. All analysis and computation should be conduct behind to obtain parameters to draw points. Of course, the design-space isn’t a static thing, it changes as our understanding of human computer interaction expands. 10:03, 18 September 2014 (EDT)10:03, 18 September 2014 (EDT)10:03, 18 September 2014 (EDT)10:03, 18 September 2014 (EDT)10:03, 18 September 2014 (EDT)10:03, 18 September 2014 (EDT)10:03, 18 September 2014 (EDT)10:03, 18 September 2014 (EDT)10:03, 18 September 2014 (EDT)10:03, 18 September 2014 (EDT)10:03, 18 September 2014 (EDT)10:03, 18 September 2014 (EDT)10:03, 18 September 2014 (EDT)10:03, 18 September 2014 (EDT)10:03, 18 September 2014 (EDT)10:03, 18 September 2014 (EDT)10:03, 18 September 2014 (EDT)10:03, 18 September 2014 (EDT) The second paper is mainly about the compilation of a number of properties of multitouch systems as well as different methods of interfacing with these systems. This compilation allows designers to consider how different elements of the device's design can affect its function. First, the author states the origin of multi-touch and also some dogma. Then, some meaningful and relevant distinctions are drew which include touch-tablets vs touch screens, location specificity and degrees of freedom, etc. He is trying to educate the public on the terms, types, and technologies of multi touch input devices. One of the first things I noticed about this reading was the several spelling and grammatical mistakes. Many times they can be overlooked, but they are glaring and make it harder for me to accept Buxton as an expert on the topic, even though he is. But he does make very good points about the lack of knowledge on multi touch. For instance, even the word multi touch is often used incorrectly. The most important points in my opinion were, one, that input devices, and multi-touch in general, are a great example of the "long-nose of innovation." That is, multi-touch had its roots in research many years ago. Although somewhat long, the presented history illustrates the progress made in touch-input over the course of four decades. Useful features can be followed for several decades whereas some features are created and then disappear quickly.

Nick Katsipoulakis 18:10:04 9/17/2014

A Morphological Analysis of the Design Space of Input Devices: In this paper a morphological classification of the design space is presented. Initially, the authors abstract each interface as a language for translating device interactions. The important features for this language are: the primitive movement vocabulary and the set of composition operators. The former is the translation of physical movements to logical parameters in an application; the latter denotes the flexibility to combine movements to a richer set of interactions (i.e. combine movement with clicking). The design space that the authors present is defined from the possible combinations of movement vocabulary and composition operators. Next, the authors discuss about methods for testing input devices, which are mapped in the design space. They believe it is important in order to understand the degree of difficulty of a device. Two metrics are presented: expressiveness and effectiveness. The former deals with the way input is expressed as output; the latter has to do with the utilization level of a user’s input. Effectiveness consists of different factors such as footprint, device bandwidth, pointing precision, errors, time to learn etc. Special attention is given to the footprint of a device in terms of a control-display ratio. In addition, bandwidth of a device comprises of muscle bandwidth and device bandwidth. These two aspects are related and they are used to define how easily a task is accomplished with an input device. Finally, precision of a device is defined as the completion time for the “easiest hard task” of the mouse and the authors use it to place a point of reference in effectiveness of a design. ---------------------------END OF FIRST REVIEW------------------------------------------------------------- Multi-Touch Systems that I Have Known and Loved: This article discusses the notion of touch systems and the misunderstandings that have been created in the past years. The author makes a point by stating that there is no leading touch technology, but different technologies specialize in different scenarios. A thorough distinction of touch technologies is provided by enumerating different dimensions of a touch system. Variations emanate from whether a system is perceived as a screen or a tablet, the time the system needs to capture a gesture and accuracy. Furthermore, the richness of interaction plays a significant role, along with the size of the touch-sensitive area, orientation and touch-sensing abilities (angle, force and pressure). Other dimensions of touch-systems described in this article are the number of fingers recognized and the level of human member recognized by a system. One important distinction is the ability of a system to provide feedback, make the interchange of devices apparent through the GUI to the user, and provide the same visual stimuli regardless of the device. The author points out that current touch systems still rely on vision rather than touch. For instance, if somebody cannot use her eyes, modern touch devices fail to satisfy basic usage. In general, this article clarifies the definition of a touch system, points out different dimensions for classifying such systems, reveals modern systems’ shortcomings and states that there is no device suitable for every task.

Eric Gratta 19:11:02 9/17/2014

A Morphological Analysis of the Design Space of Input Devices (1991) Stuart K. Card, Jock D. MacKinlay, George E. Robertson This paper’s goal was to come up with a general or abstract way of both generating interfaces and comparing them. In a previous reading (I think another Stu Card paper) we witnessed the design space, a parametric representation of interfaces in relation to one another. The design space parameters that the authors developed are strictly morphological (related to structure and form), and a “language” of device input interaction was devised to discretize the morphology of input devices. The parameters on the vertical axis of the design space were the “primitive moves” in the language, and different styles of arrows were used to indicate “composition operators,” that combined the primitives of the language into “sentences.” By creating the space and plotting existing input devices, it’s possible to see visually what kinds of devices have not yet been created and what characteristics of those devices should be explored further. By the end of the paper, the authors had emerged with examples of how observations in the design space could expose where more direct transduction from human movement to input device language would dramatically reduce the completion time in comparison to the performance of current input devices of the time. Despite this probably being a significant contribution to HCI research methods, I thought the paper was extremely lengthy and saturated, so much so that the overall goal was almost always unclear while reading it. Figures and charts were never on the same page as their reference, forcing the reader to navigate around and struggle to keep a coherent sense of the authors’ arguments. Another reason for the confusion is that the authors chose to focus almost exclusively on pointing devices as examples of how the design space might be generated and how improvements could be explored with the use of the design space. By focusing on just one set of input devices, it was very unclear whether they were actually proposing a general design space-generating method or proposing a single design space to apply generally to all input devices. The proposed design space certainly does not apply generally to all input devices (although they conceded that it was not applicable to voice input). In the exploration of their point-and-click device case study, they attempted to classify the difficulty of movements based on mouse movement time. I thought this notion was ridiculous. Their minimum time for “difficult” tasks was taken from the mouse movement time to select a character, and from that baseline every task with the headmouse was considered “difficult.” What was not addressed, however, is that a slow task is not necessarily a difficult task. Again, the relaxed nature of the case study was conceded by the authors, but there were so many instances of soft/approximate math that I lost some amount of confidence in the results. ------------------------------------------------------------------------------- Multi-Touch Systems that I Have Known and Loved (2007-2014) Bill Buxton Beyond the description offered by the title, this compilation of the history of multi-touch systems also addressed issues in the discourse surrounding multi-touch systems. Some hints that the author was interested in giving the reader a deeper understanding of HCI were his uses of the buzz-phrases “long nose of innovation” and “Everything is best for something and worst for something else.” It was interesting that the author targeted those who think the mouse is replaceable as fools; it was suggested that the complements to the mouse’s weaknesses are the real areas to look for progress, rather than fail to duplicate the mouse’s strengths. As stated above, Buxton clears up some terminology that apparently gets muddled in HCI discussions, especially those terms relevant to multi-touch. The descriptions of location specificity, degrees of freedom, orientation, and angle of approach/friction detection in an HCI context illuminated some obvious things and offered some new information to me: Touch screens as they exist do not give any physical feedback and so cannot, in most cases, be used with hands alone. Many interfaces today allow few degrees of freedom, and cannot really capture the complexity of real, physical cues. The same interface can function better or worse, or at least differently, depending on its orientation, independently of the interface’s awareness of its orientation (e.g., using a vertically positioned tablet prevents accidental palm touching, which is not addressed by the fact that an iPad or other tablet knows what position it is in). Finally, recognizing approach angle and friction were simply interesting potential inputs that I’ve never seen used in a real-world application. Buxton extensively warned the reader to make careful distinctions between various interaction contexts, such as two fingers from one hand being very different from one finger from two different hands. There are many subtle differences in the way interfaces work that need to be distinguished accurately. Overall, the paper offered an interesting glance at how limiting our current interfaces are. We perform so many actions simply through the manipulation of a single point in 2D; inputs to real life actions are so much more complex! Even the multi-touch input devices today, namely phones and tablets, capture relatively little complexity compared with what is possible. That said, it is difficult to imagine exactly where the capturing of the various complex inputs will become useful. My only critique of the paper (since it isn’t academic) is that there were MANY typos. You would expect someone who works at Microsoft to be a habitual spell-checker.

Wenchen Wang 21:37:11 9/17/2014

A Morphological Analysis of the design space of input devices Summary: This paper introduces a systematizing knowledge about input devices, by defining design space and testing design points. It is a method of morphological design space analysis to comprehend different input device designs by combining the previous work, such as toolkits, taxonomies and Performance studies. Paper Review: The design space generation involves two key ideas, generating a primitive movement vocabulary and a set of composition operators. The primitive movement vocabulary is represented by a six-tuple, <M, In, S, R, Out, W>. The composition operators are merge composition, layout composition and connect composition. Take the mouse for example. Merge composition is the combination of two devices. Layout composition is the collocation of two devices on different places of a common panel. Connect composition is the mapping from output of mouse to input of the screen. For testing points in the design space, it could by evaluated by expressiveness and effectiveness. Expressiveness is the input conveys exact and intended meaning. The Effectiveness is how fast the input could convey the meaning. The effectiveness can by evaluated by desk footprint and pointing speed, which also called device bandwidth. Multi-touch Systems that I have known and loved Summary: The author introduces some interesting attributes of multi-touch systems, some dogmas about system designing and also the history of multi-touch system from 1965 to 2011. Paper Review: Multi-touch greatly expands the types of gestures that we can use in interaction, from simple pointing, press button to multi-figure operation, force vector operation and continuous gesture. By touching a screen, there is more to touch sensing, such as degree of touch, angle of approach and force vectors. For the degree of touch, when I keep pressing icons of iphone, I can easily delete any of an application. For the angle of approach, Android system has the virtual keyboard to do shorthand writing. For the force vectors, some video applications have up-down finger touch operation to adjust the voice, and left-right finger touch operation to adjust the progress of the video. I like the dogma that everything is best for something and worst for something else. Everything has tradeoff. You may improve speed or performance at the cost of expense. Same thing to multi-touch system: the more diversity of touch operations, the more distinct style of interaction on different devices. People need more time to get used to it.

Longhao Li 21:58:49 9/17/2014

Critique for Analysis of the Design Space of input Devices The article basically talked about the morphological analysis of the design space of input devices, which is a new development of the study of Human-machine interface technology. It systematically analysis how interface works and how they make human computer interaction possible. There are good contributions in this paper. The author finds out how to do modeling of input device by using tuples in sets, and also the author point out the different kinds of connections between human and computer. By using them, we can draw tables about how input device works. After that, according Fitts’ law, we can derive the time of movement about the input device and the bandwidth of the device. Finally, we can know if this input device works well, and how does we can improve it. In general, I think this paper is important for designing input interface. Even though we have a good idea to design input device and we make it real, there still be some chance that people don’t want to use. It is because that it may not faster than what people are using now. Just like using finger pointing on air as mouse. It seems to be a great idea. But in practice, it is hard to do some accurate operation since people’s fingers are bigger than mouse curser and also people’s finger on air is not stable. For example, if you want to close a window, you need to aim accurately on the little button, and go forward with out shake. It is hard. I think by doing the analysis that the author introduced, we can find that the bandwidth of the input device are lower than mouse since it need longer time for movement. If the designer uses this analysis method when developing the input device, they will find problem and improve the device so that the product can have more people like to use. Thus, I think the analysis method is great, and worth to use in practice. Critique for Multi-Touch Systems that I Have Known and Loved In general, this article talked about what is multi-touch, what it is good at, and also what it is poor at. Also, the author point out some suggestions that designers need to care about when they are designing a multi-touch interface. This article is so great for designers of multi-touch interfaces. It is not to say that this article shows a brand new idea about multi-touch interface. But this article gives us a comprehensive understanding about multi-touch interface. We cannot say that the multi-touch is always great. It has limitation. Just like the author said that “Everything is best for something and worst for something else.” We need to carefully design the multi-touch interface since we may encounter pitfalls. By using this article as reference, I think the designers will have the opportunity to make their interface more suitable for multi-touch operation. Therefore, to my understanding, I think this article is great. Designers will get benefit from it and moreover, it may lead to further development of multi-touch interface and also get good answers to the question how to use multi-touch interface as complementary for ordinary mouse keyboard interface. In the article, the author also point out that the touch interface should not be used in car console. I think this is a great idea. It definitely influences driving. When people pay attention to find where the button is, they may not be able to look at the road at that time. This is risky. This is same with texting while driving. Thus, to my understanding, I think car console interface should be made with buttons, which can give people tactile feedback and also since people get used to them, using real button will save people’s time to learn how to use it. This is good attempt when doing interface design.

Qihang Chen 22:25:34 9/17/2014

The paper A Morphological Analysis of the Design Space of Input Devices discusses a means to systematize the input devices through morphological design space analysis, in which different input device designs are taken as points in a parametrically described design space. The paper first has a introduction of previous work on systematizing human-machine input devices which are toolkits, taxonomies and performance studies. To make up the limitations of the current development lines, the authors propose the new technique to find abstractions both for generating the design space and for testing the designs contained therein. Then, a reflection of input devices and the model are provided. Further, to model the language of input device interaction, a primitive movement vocabulary and a set of composition operators are introduced in detail. The paper also discusses the basic criteria expressiveness and effectiveness and goes into detail on the merit desk footprint and bandwidth. A good example is given on the effect of muscle groups on input devices. In the example, a comparison is made between mouse and a headmouse. The Fitts' is law is adopted to calculate the movement time. The main contributions of the article can be summarized as: (1) provides a new method for helping to generate points in the design space suggesting a promising direction for developing new input devices; (2) how designs can be critiqued in terms of expressiveness and effectiveness shown in the paper is good illustration of how regions of the space can be systematically analyzed; (3) though focuses on input devices, the paper is beneficial for HCI studies to improve the human-machine experience. Given the fact that bewildering input devices exist and are being developed, the paper offers a systematic way to classify the multiple devices and some hints to better evaluate the performance and the direction to develop new ones. Pity that I don't find there is any relationship to my current work. ------------------------------------------------------------------------- This article by Bill Buxton is a survey of the characteristics of some of the multi-touch devices prevalent in today's systems. Drawing from his rich experience as a researcher in HCI, Buxton brings to light many of the technical aspects of today's diverse multi-touch systems. Buxton claims many times that "everything is best for something and worst for something else." This is a theme that is prevalant throughout Buxton's analysis of multi-touch devices. He claims that "understanding and weighing the relative implications on use of such properties is necessary in order to make an informed decision." In other words, there is no one-size-fits-all approach to HCI, particularly as it pertains to mobile/touch devices. This article is largely a survey of Buxton's thoughts on multi-touch devices - he characterizes devices based on a wide array of properties. The most important points in my opinion were, one, that input devices, and multi-touch in general, are a great example of the "long-nose of innovation." That is, multi-touch had its roots in research many years ago, dating back to at least 1982. While marketing departments like to promote the thought that companies are innovative, in fact it is simply a matter of tweaking a long history of research to the point that products become commercially feasible. Another interesting point that Buxton makes is that it is human capability rather than technology that should be front and center in HCI research. Computers will continue to improve - device speeds and capabilites will probably also continue to improve as well. But human performance, relatively speaking, will likely remain rather consistent. As such, it will be important to focus on design techniques that will aim to better utilize human capabilities, as these are the bottlenecks to interface performances.

Bhavin Modi 23:42:20 9/17/2014

Reading critique on A Morphological Analysis of the Design Space of Input Devices The paper describes an approach for ordering input devices in a design space based on a primitive movement vocabulary and a set of composition operators. The aim is to provide taxonomies of the design space of input devices so as to better understand their application and relative importance to each other, and to provide a common platform for analysis and generating new input devices by using this design space. The interaction between human and machine has three agents the human, a user dialogue action and an application. The input is illustrated as a six tuple for mapping. The composition tasks are broken down into merge, layout and connect composition to give a detailed view of different aspects to take into consideration. It is not only the input device but the human factor is taken into account discussing merits of user preference, cost, time to learn, errors, and pointing precision. The footprint of a device is an important measure and the apex factor while designing new devices. The different human parts as input device controllers are mapped to a graph showing that the fingers give the best performance. The assumption that the bigger the motor cortex for the muscle group the more the control is shaky at best, with the associated index of difficulty bound to change per individual. The design space though defined well, the mapping of input devices is not very well articulated. I failed to understand how to read the design space, how it would work for me and help me innovate. Lacking in explanation would summarize it. The comparison between the mouse and head mouse was not good proof and the degree of freedoms are different, they should have used fingers, speech, eyes or even gestures to better illustrate the point. The paper gives a lot of references to Baecker and Buxton, Mackinlay, and Foley, Wallace and Chan among others. It is important to learn from others work and improve on it, but the format of the paper makes it very complex. The conclusion is that with many limitations, it is a great concept to map it into a common design space as it allows us to explore the interesting regions and concentrate our efforts which show promising possibilities, maybe one day beat the mouse. -------------------------------------------------------------------------------------------------------- Reading critique on Multi-Touch Systems That I Have Known and Loved-Bill Buxton Clearly the objective of this paper is to discuss Multi-Touch Systems and Touch Interfaces, the related terminologies, the advancements and the difference between them. Explaining to a novice in the field the researches carried out and that Touch Interfaces and Multi-Touch Systems did not just appear but have been around for 25 years. This paper clearly points to a statement I remember from class, the technology of today was researched decades back, and you will definitely find a research paper on them. So, if you design something today, start from the beginning because you may already be late to the party. The classic example of the mouse, introduced by Engelbart and English on 1965 gained widespread use in the 1980’s. The 1980’s were a significant period for advancement in touch interface technologies, the one point touch input of vector information, soft machines, multi-touch screens and tablets, gesture recognition by Krueger, the sensor frame are among the researches being done at that time. The author tries to teach us the terminology of touch systems so that we can have meaningful conversations on the topic. He describes various attributes of such systems, difference between touch screen and touch pad, multi-touch and multi-person, location specificity, degree of freedom and many more. There is more to touching than just contact and position, the aspects of force, angle and degree of touch also come into play. A good example for degree of touch is the recently released iWatch which can detect a touch and a press. Size, multi-touch, use of stylus, and about the look and feel of such systems where again we can have the QWERTY virtual keypad example. One important point brought forward was that mechanical devices afford affordances and help in feed-forward feedback control, so with touch screen we don’t know exactly what we have touched. This requires us to look at the screen at all times and this cannot be converted into a habit like when using a QWERTY keypad after we get used to it. The understanding of different vs. same is also important, it shows us that just replacing the devices on the same interface does not necessarily return same efficiency results. Though we have talked about bi-input interfaces, they can hinder the user experience unless a good combination is provided. Reading this article we realize the amount of effort and research it takes for an idea, and interface to become commercially viable.

Yubo Feng 0:25:13 9/18/2014

Both of two papers discuss about input device evaluation, one of them used a new concept created by the author, which is design space. Design space takes different devices as a point, then mark it with a special space, then try to compare different points with their degree and so on. Basically to say, what cause me feeling interesting is the application of Fitt's law. The Fitt's law used in evaluation is too far beyond my mind, and I think it is possible to measure other devices though many ways by applying Fitt's law in different direction. The other article, mostly describe the multi-point touching input mode, I think if we use the former article's point to do evaluation, it will be more fun to find that multi-point touching many be a bad idea.

Mengsi Lou 0:48:46 9/18/2014

A morphological Analysis of the Design Space of Input Devices This paper tells about the concepts of design space and also proposes some improvement about the lines of development and then use calculations to reason about the design space. ---------Traditional lines of development are toolkits, taxonomies and performance studies. And the author adds a new fourth line, that is a morphological design space analysis. 1. The Toolkits includes the construction, run-time execution, and the analysis of a user interface. 2. Taxonomies. Buxton and Baecker proposed that includes the physical properties and the number of spatial dimensions they sense. 3. Performance studies. 4. Morphological design space analysis. The goal is to find abstractions for generating the design space and for testing the designs contained in there. ----------The author also indicates the generating of design space. Take language of input device for example. There are two ideas in modeling, that is a primitive movement vocabulary and a set of composition operators. Later in this paper, the author states two points in testing the design space: expressiveness and effectiveness. Here are some details about footprint and bandwidth. ///////////////////////////// Multi-touch Systems that I have known and loved This article indicates some basic points about the design of Multi-touch Systems, and the author also gives some examples in this field. ----------The author lists the typical frame from many sides. Actually we have been exposed to these problems everyday and everywhere, but just haven’t got the chance to think about it clearly and systematically. I’d like to talk about some of the interesting points. First, the location specificity means the degree of accurate that a user has to position a touch at a particular action. Different tasks require different levels of accurate that have a big influence on the interaction. Second, there is more to touch-sensing than contact and position. 1. Pressure sensitivity. A surfaces can sense the degree of contact for each touch point. 2. Angle of approach. Some systems can sense the angle of users’ fingers referencing to the screen surface. That extends to the new field of 3D related tasks. 3. Force vectors. We can simply seen from many products that owns this ability such as iPad and some other smart phones. It also reminds me of some fun games that requires this ability such as starwalk etc.

phuongpham 0:50:09 9/18/2014

A Morphological Analysis of the Design Space of Input Devices: this paper introduce a design space for input devices. This space can help designers to analyze and give better decision for input device designs. The authors have shown that the space can map most of the input devices. Moreover, limits of these input devices can also be mapped on the same space using new dimension. This helps designers having a single view to analyze. However, I think the expressiveness of the design space still has some minor fallbacks. For example, using multiple circles to represent a single device does not give a completeness about the device. I feel if a device can be viewed as a single point in the space, it would be better. On the other hand, if merging all circles into a multi-dimension space, we may lose the composition connections between the circles. One more point is the Measure. Positioning circles inside a cell with different meanings, e.g. to the right nearly continous while to the left indicates controls with only 2 states (how about 3,4,5 states?). Lastly, as indicated by the authors, this space does not support voice input, which would be popular in the near future. I think this would be a new research question. ***Multi-touch Systems that I have Known and Loved: this is on one hand a unofficial literature survey about multi-touch technology. On the other hand, more interestingly, the author has pointed out insights about many aspects of the technology. Beside the history and insights of the multi-touch technology, I can get something else beside research from the paper. First, as our professor has stated in the class, "everything is best for something and worst for something else". Take this point seriously, we can find new interesting research topics and analyzing what are the worst scenarios of a particular technology. Some weakness would be difficult to solve with current technologies but there would be some ideas to touch the other weaknesses. Second, "human capability, not technology, that should be front and centre in our [HCI] considerations". Instead of exploring what current technologies can help people, another way is looking for that people can do but have not been supported. Skimming through the chronology of multi-touch and related work, I have a feeling that the new exciting multi-touch technology may transit from handheld device to household device where users can access to bigger screen which allow them use more input mechanisms. However, as reflected from the early paper we have read in this course, "As we may think", the ideas are the same but can be applied for new technologies. For example, many early work in multi-touch based on camera or sensors attached to the monitor, now we see mostly on touch screen. So I think even new technologies which we haven't seen yet can be developed from today ideas and support today ideas.

Xiaoyu Ge 0:58:51 9/18/2014

A Morphological Analysis of the Design Space of Input Devices This paper introduced a method to group input devices. In order to do that, the author build up a model, to perform this functionality. The paper introduced a test method as well by evaluating the model according to two principle rules: expressiveness and effectiveness. The author systematizes the knowledge about input devices, and provided a method help generating points in the design space by calculate the footprint and bandwidth. Although, people seem to find another ways to deal with is problem as well. Because other people introduced other researched seems not working so well, this new model build by the author was a better choice at that time. And the method for creating models for input device will be beneficial to my future work on interface. And web application can be build with the help of this model nowadays. Multi-Touch Systems that I Have known and loved This paper is concentrating on the discussion of multi-touch systems such as touch screen. And the paper introduced a diversity amount of elements affecting the interaction with multi-touch system. The discussion in the paper is useful in the design application, the elements it defined can be use to determine what functionalities can be populated into the application. And there are also limitations introduced in touch screens which gives us a more clear idea about what kind of application that normal computer can have better performance of even unable to populate on touch systems because of the physical reason. For example, many companies design different versions websites for mobile touch screen devices and computers, and one of the reasons for that is the design of check box or button have to be modified to fit for the fat figure. However, there is a contrary example of this situation. The newly released windows 8 is able to use a figure point to the screen and drag it to switch, but this kind of functions is not functioning if you use a mouse to do it. The elements for multi-touch system illustrated in this paper should be considered if I would like to work on Auto adjust websites, which can be fully functional and pretty on the multi-touch system device.

Yingjie Tang 1:01:52 9/18/2014

This article mainly propose a method use morphological analysis to discuss the design space of input devices. It uses the footprint and the bandwidth as its two main function to analyze it. The main advantage in it is that it can systematically analyze the prospect of some input devices on some aspects. According to the calculation, the head mouse has less advantages than hand mouse due to the bandwidth of muscle group of neck is less than half of the hand and finger muscles. I think the view proposed by this article is quite novel nowadays, the interface technologies develops quickly and we it would be helpful for us to know the prospect of some input device by calculating, this gives us a theory and will prevent us to do something useless. Moreover, the morphological analysis is also useful for developing interfaces. Interface design should take morphological analysis into account because a design will be useless if it is not easy to control. There is a question arises after reading this paper, I remember Prof. Wang said that hand mouse is more easy to use than finger since controlling a mouse only uses our finger while we have to use the muscle group of our elbow when we control the input device by finger. However, this contradicts with the statistics on this paper, it says that the bandwidth of fingers is 3 times greater than a hand mouse.—————————————————————————— The author is an expert in multi-touch system and he gives us an thorough analysis of the development of multi-touch system. Although the multi-touch system seems to be ubiquitous in recent years, but it has been invented by a few generations. Before it has been applied in the commercial products, many researchers and scientists have made great contribution on it. Moreover, the author always emphasize the axiom that “Everything is best for something and worst for other”. He mentioned that although the multi-touch system brings some convenience for people, it need people to focus on its feedback to keep your eyes on the screen. This may cause inconvenience because you can not touch the physical button to control the device blindly. I can not agree more with the author that multi-touch system is not always better than some mechanical devices. Although the virtual keyboard is convenient on some mobile devices, I enjoy the happiness using mechanical keyboard. Since I cannot feel the position of the characters, I have to stare on my screen which will cause me a lot of trouble. For example, when I used a old fashioned mobile phone, I can feel the position of the numbers and I can dial my friends blindly and it is impossible with a virtual keyboard.

Brandon Jennings 2:02:54 9/18/2014

A Morphological Analysis of the Design Space of Input Devices This paper is about applying the morphological analysis to design input devices. The paper shows how to analyze and critique the design space. One of the key components of the morphological design analysis is that it does not disregard the contributions from the seemingly trivial components. The two criteria used to assist that is expressiveness and effectiveness. It is not only important that an input convey its intended meaning but also that it does not convey any other meaning, intentionally or unintentionally and needs to be done so for optimal speed and accuracy. This paper delved deep into examples showing how to use analysis, like of human muscle groups, to better define input device design using the morphological method. To further prove its case, it actually presented a direction one can head to develop devices better than the mouse by using the bandwidth of fingers. Admittedly, I found this article a little difficult to follow. It is lengthy and I did not completely. I had to use the Internet to find some relevant material to help me understand terminology and background as it related to morphological analysis. For as long as the paper was, I felt there could have been more recap about morphological analysis and the previous work. In general, however, I think this paper could have been condensed. Though I do think the notion of systemizing the knowledge about devices, in this case, input devices, is important. Multi-touch Systems That I Have known and Loved This paper is about the evolution and progress of touch technology. The author wants to address common questions and clarify any misconceptions people might have about touch technologies. An important part of the paper is the distinctions of terms associated with touch technology. It breaks down the differences between things like single-finger and multi-finger, multi-person and multi-touch, and even goes as far as to define what entails touch-sensing (i.e. degree of touch, angle of approach, force vectors). This paper also goes into issues with touch technology that some people may have not thought about, like the inability to feel the controls, hand held devices require more hands than necessary, small screens can be obstructed by your fingers, and the glare from lights. Some devices are ideal for something and terrible for something else. The history of touch technology presented in this paper is rich. It gives a chronology dating back to the 60s. It is interesting to note that there was a working mechanism for touch screens in 1967 that used capacitive sensing, which is a basis for touch technology today. The reader can see the progress of the technology and how each generation builds upon the previous. I found this paper to be a great read for people who generally want to know about touch technologies.

Zhong Zhuang 2:51:58 9/18/2014

This paper is amazing. It describe the history of multi touch technology and its pros and cons. It also addresses the design concerns about this technology. One concern that I found very interesting is the ability to sense the size of the touching area. Many users have the experience of accidentally hang up the phone because the cheek touches the phone. By sensing the area of the touching, we can avoid this problem. It arises many other points that I never thought of before. It introduces all the papers and projects that related to multi touch and how it related to today's smart phone. The main point of this paper is some technologies are best for something but worst for something else. The idea of multi touch emerge for the 80's. But it took 20 years before it finally came to fruition, because, as the author stated, one technology is never the only cause of a new product it's the whole new ecosystem that is required. Before I read this paper, I never expect that there are this many concerns about multi touch.

Andrew Menzies 2:54:46 9/18/2014

Multi-Touch Systems that I Have Known and Loved Bill Buxton This paper discusses some of the capabilities of state-of-the-art touch-based input devices, and the importance of being specific when mentioning touch as an interface element. The author points out that touch screens in general, as well as specific innovations like multiple touch and force detection have different applications and drawbacks. This paper brings up many abilities of touch detection, some of which I did not know about before reading. For example, I knew that touch screens that could detect multiple fingers existed, but did not know they could differentiate between fingers belonging to the same person or different people. I also did not know that touch screens could differentiate between a single finger and multiple fingers moving together, or that they could detect force direction in addition to pressure. The paper mentions the importance of feedback when interfaces detect different aspects of touch, such as using different colored trails in response to different numbers of fingers being moved over the screen. I think this is an important point because without it, for example, I might not know the device treats me using two fingers together any differently than using only one. The paper also provides an excellent example of the “long nose of innovation”. As many sources have mentioned, inventions typically take many years (1-3 decades) to move from first implementation to mainstream awareness and use. The chart at the end of the paper does an excellent job illustrating this—I had no idea that touch detection using the electricity the human body conducts existed well before 1980! A morphological analysis of the design space of input devices Stuart K. Card Jock D. Mackinlay George G. Robertson This paper introduces the concept of a “design space” that input devices belong to. This design space lets interface researchers talk about what sorts of motion a device measures as well as how it is used with other devices. The paper’s main interesting idea is providing a visual chart to compare input devices. The chart allows designers to visualize the types of motion and force a device captures, as well as its interactions with other devices. (Such interactions can include: its output being used as input for another device, its input being combined with input for another device, or it and other devices being located on the same larger device.) The author claims that the chart helps us visualize what sorts of input devices could deserve further research. However, I believe that the chart has a fundamental problem: it only considers devices that involve either force or motion in 3 dimensions. Other forms of input, such as light, sound, biometrics, and maybe even (through monitoring brain activity) thought, are conceivable, but do not have a place on the chart. The chart is useful, though, in that it lets us compare devices. Even though two devices may be structured similarly and involve the same type of motion, thereby appearing the same in the design space chart, their effectiveness can differ. The difference in performance between users using a mouse and a “headmouse” illustrates this. This example also relates an interesting fact: Different muscles in our body can be controlled at different levels of precision. Thus, some parts of our body can produce more bandwidth when used to operate an input device. This means input devices that rely on neck movement, for example, cannot be effectively used for pointing tasks as precise as those relying on finger movement, even if both devices have the same “domain” of input (allowing movement in the same directions).

Senhua Chang 3:20:25 9/18/2014

A Morphological Analysis of the Design Space of Input Devices This paper presents a new method for organizing input devices. The purpose of this paper, as noted in it is “bringing order to knowledge about input devices”. The whole space is composed by some primitive movement vocabulary combined together by some kind of compositions. The authors gave examples of common input devices’ positions in that space. I think they didn’t take into consideration things like digital cameras and microphones. In my opinion, once a new Human Computer communication media is created, people will quickly exploit the design space enlarged by the invention, where using such kind of analysis shown in the paper would be useful. The authors also provide some examples with these metrics. However, there are still lots of other metrics, such as pointing precision, time to learn, time to grasp, etc. I think it is necessary to formalize other rules and hope this will build the entire evaluating model more reasonable and accurate. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%55 Multi-Touch Systems that I Have Known and Loved Generally speaking, this website provided a comprehensive introduction to the history and important development of multi-touch systems. It also tries to clarify several groups of concepts that are confusing especially for the users. Apparently, this article is a valuable source of information. It summarizes the history of touching technology and refers a lot of sources that we can use later on. Moreover, since the author was (and, is) in the center of the history, the overview seems to be more informative, the comments between the lines seem to be more persuasive, and, overall, the article seems to be more spirited. The pictures and video links are also helpful; these materials can help the reader grasp each technique more directly or, at least, be good eye-candies. By the way, I like the example of finger painting vs. using stylus. It's simply not what we do in real life, except for some genre of modern arts. Another interesting thing is that, though different example might look different especially when presented in such an application oriented way, the senescing and recognition do have main trend there.

Qiao Zhang 3:21:15 9/18/2014

A Morphological Analysis of the Design Space of Input Devices This paper introduces a way to compare different types of input technologies. The authors view different input device designs as points in a parametrically described design space. This is an interesting view of classification, as different devices are abstracted as a vector in an attribute space. If the design space can be visualized, we may see different clusters of similar input devices. Expressiveness and effectiveness are the two basic criteria to evaluate input device design. Expressiveness means the input conveys exactly and only the intended meaning, while effectiveness means the input conveys the intended meaning with felicity. A design that violates the expressiveness would be: a user tries to accomplish the job of "select pixel x=105, y=32" while she has an input device that has a resolution of 1/4 inch. The footprint is another criteria for an input device, which measures the occupation of area. However, from my point of view, smaller footprint is not always preferable. For example, a larger Wacom digital tablet has larger drawing area, thus is convenient for the artists to work the details. In this case, smaller footprint saves space at the cost of losing precision. The bandwidth measures if an input device can be as fast to use as possible. The paper gives a approximation for computing the MT, which takes the qualitative Fitts' Law towards a further step. The human muscular system will also affect the bandwidth. For example, a headmouse is less precise than a regular mouse because the neck muscules are less flexible than the wrist muscules. For this example, however, I am wondering if there is any ways to coorporate both muscular groups, i.e. use neck muscles to do less precise part and use wrist muscles to do more demanding job. This approach would potentially reduce the pressure on wrist muscles, which will benefit most computer users. And another bizarre thought that I have is that if there is any possibility to utilize human's tongue muscle as an input device controller. I know that tongue is the most flexible muscle of human body, however despite the inconvinience, I also know that tougue is not designed with a controlling functionality. Using tongue tracker may have the same difficulties as using eye tracker. But this could potentially help the disabled, if they are still able to speak. ============================================= Multi-Touch Systems that I Have Known and Loved Multi-touch systems are prevalent nowadays. I myself have an Android device, an iPad, a laptop that has a built-in multi-touch touchpad. I once owned a Macbook, and the touchpad is amazing. It supports several gestures mapping frequently used functions such as maximize a window, show the dock etc. I would say I will buy a Macbook solely because of the touchpad. But as the article suggests, a problem with this is that each of them is with a distinct style of interaction. I hope they all have a universal standard of interaction, however patents do not allow the manufacturers to do so. This adds burden to human users' mind, which causes "more is less" effect. The article points out several terminologies that people often misuse or misunderstand, e.g. touch screen vs touch pad, for which the difference lies in one offering direct manipulation and the other not; discrete vs continuous, for which the difference lies in whether actions are discrete or continuous; the feedforward term describes how the right conceptual model functions; With more degrees of freedom, users can do more actions with more gestures at the cost of remembering a bunch of instructions. Buxton is also concerned with the drawbacks of multi-touch screens. But as he said, it does not mean that multi-touch technology is not good at all. It just mean that everything is best for something but worst for something else. We should take the weakness into consideration when designing multi-touch devices. Buxton also provides a list of multi-touch devices back from the 1960s. It is interesting to see a lot of today's devices prototypes are invented far back in the early days. As the instructor suggested in the class, if we want to make some products, we should look 10 years back to see what the research work was then. As I can see, projection was a widely used technique back in the 2000s. As the projectors are getting cheaper these days, I can imagine if there is a good concept, every household would have a projector and a whole new market can be opened. It is also interesting to see that some concepts are implemented in a new way, like the answer the phone by holding to cheek in Synaptics & Pilotfish example in 2006 is implemented by Samsung using front camera today. Good to know that Microsoft Surface was as big as a table when it was invented and shrank to today's portable tablet size.

yeq1 5:14:58 9/18/2014

Yechen Qiao Review for 9/18/2014 A Morphological Analysis of the Design Space of Input Devices After seeing a wide array of devices that facilitates human computer interactions, the authors sought to how to provide a morphological device space analysis on the design space of these devices so the designers can select the right ones easier. The paper has used earlier work on taxonomies and performance, and other similar analysis in engineering, and has built the design around two criteria: expressiveness and effectiveness. Unsurprisingly, they have used formal language to test the expressiveness of the devices. And they have selected multiple metrics on effectiveness, including footprint, bandwidth, etc. The paper suggests that while some devices may be better than others in one performance metric (mechanical mouse better than joystick on speed), it is entirely possible that other devices are better for some other metric (joysticks are better than mechanical mouse in footprint for larger screens). I think the authors had made many good points in the paper. The first is the need for the designers to consider the design space. The design space of input devices are very complex, and this 2D space seems just like a toy example. However, the authors have correctly pointed out the tradeoffs between two effectiveness metrics, and the potential problems with expressiveness. One anecdote I had happened when I was in middle school. I was so happy I finally got to play with a high quality joystick, and I wanted to use it on all the computer games. I thought the joystick with over 100 programmable buttons would allow me to ace multiplayer FPS and racing games as well. Needlessly to say, that did not happen. I found out that 1) having way too many buttons, static modes and switches gave me way requires me to remember way too many things for each game, and many mappings has to be made arbitrary because they make no other sense otherwise; 2) joysticks takes much more efforts to move around, even if I set the sensitivity to be the highest (there’s a mechanical knob for adjusting sensitivity), and my wrist and arm easily gets tired; 3) sudden movements to the complete opposite directions are simply impossible. After doing some experiments, I found mouse as the superior device for most of the computer games, even many of the arcade flight games have the troubles I mentioned above if a person were to use joystick. Therefore, I also think that in deciding which device to use, the designer should choose the device that not only suits the context (expressiveness), the physics of the device should also suit the physiology of the user. This may mean allowing prolonged use of the device, or allowing efficient use of the device. The main limitations of the study is that the authors had limited the design space to devices that supports pointing and selection (1D, 2D, 3D). While at first we may think there’s this “easiest hardest task” due to FItts’ law, there may be tasks that the “easiness” of it cannot be described by Fitts’ law. For example: submenu selection, multi-selection, etc. Mouse would allow the users to efficiently multi-select objects in the same rectangular area through dragging, or any convex regions through multi-clicking. But for objects that falls outside of a convex area, mouse selection may be inferior to multi-select with multiple fingers. The authors also assumed that only one person may be using the interface at a time. This may not be true either. Some of the restrictions of human physiology may be relaxed if we were to design the system for multiple users to operate at the same time. (For example, the huge Microsoft Surface “tablet”) In conclusion, while this is definitely a good first step, there’s clearly a need for a more complete design space analysis. Multi-Touch Systems that I Have Known and Loved In this reading, Dr. Buxton had gave us an overview of the difficulties in getting industry to adapt, and finding the right model from the design space. This article had clearly pointed out that the design space of an input device such as multi-touch inputs usually is very complex, and choosing the right one is not a small matter. I can imagine countless thoughts and experiments had been spent on finding out what is the right way to multi-touch. Unlike the last paper, this one not only greatly expanded physiology attributes, but it also considered psychology. They had described how designers had sought to provide natural interactions with objects presented in the screen, such as pinching, use of force vectors, resting arm and wrist on the surface while writing or pointing. Some of the problems are still unresolved, such as not being able to operate the device without looking at it. It is quite possible that no matter how much the designers sought to improve, there will be no solution that allows it to work at least as well as mechanical objects in some degree. Hence, the author’s philosophy: best for something, worst for something else. Having used touch devices for a while, and I found lots of points made in this paper makes sense in this context. For example, I found I have trouble dialing numbers with touchscreen phones (found this out operating PDA running Windows Mobile 6.0) while driving. As a result, I did not switch to smartphone until people started to compliment Apple Siri. My iPad has a low accuracy multi-touch input without stylus support (those rubber sticks don’t count!), but I’m still using it for checking emails because the high precision device such as surface can’t operate too long without recharge, spends too much time turning it on and off when compared to iPad, heavier and thus more difficult to use without a desk. Does this mean I prefer iPad to Surface? Not at all. In fact, I spend more time on Surface than my iPad: it is good for precision pen input, it is good for flipping through webpages or reading and annotating PDFS, it is good at taking notes, it boots more than 10 times faster than my laptop with 32G RAM and a core i7, it runs applications faster than iPad, it is more stable than iPad, it allows me to run Windows and Linux applications, and lets me compile software code. Which device I choose to use depends on what task I’m trying to do. I would choose iPad over Surface in reading emails, even if iPad is 2 feet away. Similarly, I would grab Surface to take notes even I had to drive to school because I forgot to take it home. Often, picking the right device becomes picking the right devices, and sometimes the “expert reviewers” on the Internet just don’t get it…

Christopher Thomas 7:30:34 9/18/2014

FIRST PAPER *** 2-3 Sentence Summary of Multi-Touch Systems that I Have Known and Loved by Bill Buxton- In this overview piece, Mr. Buxton discusses many subtleties relating to multi-touch systems in the form of contrasts, explaining many commonly heard and used, but frequently misunderstood or confused terms, such as the difference between touch tablets and touch screens. Mr. Buxton discusses other features of innovation, such as how many things widely regarded as new “innovations” were actually invented many years prior. Finally, a nice historical survey of multi-touch devices is given, listing many devices in chronological order, each discussing some novelty. When I first approached this article, I thought of multi-touch inputs usually as just being pinch gestures on iPhones and similar products. Mr. Buxton shows however, that the space of touch gestures (and multi-touch gestures) is much, much greater than many people realize. I think the most obvious example of this comes with the unveiling of the Apple iWatch, which now will be supporting some form of pressure detection in its input, which enables users not only to tap on a particular location, but to also apply different pressures. What was interesting to read is that most devices supporting “pressure” sensing actually don’t really detect pressure, but instead use the fact that the user’s fingertip expands on the surface when the user applies more pressure. I would be interested to learn if this is also how the iWatch is doing its pressure detection. I essentially believe that Mr. Buxton has given the “landscape” for many touch devices to explore. This relates in some ways to the concept of a “design space” that we discussed in class, whereby different concepts are placed on different axes, enabling different technologies to be placed into a “design space” based on which technologies have certain design features along that axis and organized. Mr. Buxton gave many such concepts which I can see easily being used to form a design space. Some touch devices are single touch, some use a stylus, some multi-touch devices are physical, some use degree of touch, angle of approach, or force vectors, etc. By knowing all these areas of past research, one can try to leverage past ideas into new technologies. The next major “innovation” in touch technologies, will probably not be something that hasn’t already been described in this article, it will just be refined in a new and insightful way which is appealing to users. From the concepts discussed, I see many potential research directions involving combinations of subsets of these concepts for different purposes. For instance, games may be able to make use of the angle of the user’s finger when firing a torpedo or in a golf game, the angle of the finger could be used to have some impact. One thing I would have liked to have read more about in the essay was how the author felt about “multi-touch” devices that didn’t actually involve touch. For instance, some hardware displays a virtual keyboard projected onto the table. The user then can just type by pressing on the buttons on the projection on his or her table, but the device is actually sensing the inputs using a camera. There was not much (though there was a little) discussion of the use of optics in touch gestures. For instance, with the Kinect interface, a user doesn’t actually have to touch the screen of his laptop, he can just wave his hand in the air to move a window off the screen. It would be interesting to see how the challenges faced in the past with touch interfaces can be rethought and reworked now that the touch interface itself is becoming unnecessary with improvements like the Kinect.   *** SECOND PAPER *** 2-3 Sentence Summary of A Morphological Analysis of the Design Space of Input Devices – The authors begin by explaining that toolkits, taxonomies, and performance studies are often used in systematically understanding input devices. The authors propose the concept of morphological design space analysis, where different concepts/features are placed on an axes, and devices are plotted onto it. From this theoretical model, the authors give taxonomies and a primitive movement vocabulary and plot numerous devices, explore relationships between them, and show how devices can be analyzed in the space using Fitts’s law in the case of the headmouse and mouse. The concept of design space has already been brought up in class last week, but I really didn’t understand the power of that concept until I read this paper. I just assumed it was a way of organizing devices, but I didn’t really see that it had much use. This paper however, shows that the technique can point designers to previously unexplored areas. Once a taxonomy has been made and the devices have been placed, designers can see areas which haven’t been explored or that aren’t really worth exploring. For instance, if we can determine (as the authors did) that a particular area is nearly at its maximum index of performance, as it essentially is with the mouse, which indicates a design area that it is probably not worthwhile to investigate or innovate. The authors’ used Fitts’s law to do a quick calculation showing that why trying to use a headmouse to compete against a mouse for text editing was a bad idea, due to the poor control of muscles. However, we can also see that the concept of a design space can point us to areas worth exploring, for instance the authors show that the fingers have larger bandwidth than the hand. By exploiting the fact that the fingers have larger bandwidth novel devices which outperform the mouse could potentially be produced. Thus, the design space provides a powerful mechanism for reasoning about existing devices and finding spaces to innovate. I think one of the largest take-away messages from this paper was that innovation doesn’t always have to “beat” something that already exists completely and in every situation. Finding the right niche for innovation seems to be key. For instance, we know that the mouse is the most efficient entry device for the computer, but I doubt many users would want to use a mouse on their iPhones, instead, they prefer a compact, touch interface in the phone and not an external device. I do think the authors could have improved the paper by describing some ways to derive the taxonomies in a first place. How can we evaluate not just the devices, but the taxonomies we have decided upon, as this will definitely impact our design space? One interesting thing I was thinking about would be the use of design space in reference to the last article we read, about the multi-touch devices. The author provided many different taxonomies there for multi-touch interfaces. In fact, it seemed somewhat overwhelming and difficult to reason about so many devices. One way we could overcome this is to apply the concept of a design space to the touch-interface domain. We could then plot the devices on our design space (given some axes decided upon in advance). Using such a technique could reveal areas for research, of user interface possibilities previously unexplored, or whose performance we feel could be improved in some way. We could even use Fitts’s law just as in this paper to analyze that performance mathematically and do some quick calculations to determine viable research directions. Thus, the two pieces of literature from today’s readings connect in an interesting way.

Xiyao Yin 7:45:28 9/18/2014

‘A Morphological Analysis of the Design Space of Input Devices ’ focus on input devices in human-machine interface technology. This paper is important because it develop of a set of abstractions that provide one method for bringing order to knowledge about input devices. Different from the existing three lines of development: toolkits, taxonomics, and performance studies, this paper create a fourth line of development, a morphological design space analysis, which can be used to integrate the results of previous work. Morphological design space analysis which comprehends different input device designs as points in a parametrically described design space aims to find abstractions both for generating the design space and for testing the designs contained therein. A primitive movement vocabulary and a set of composition operators are two key ideas in modeling the language of input device interaction. After modeling the space of input device designs, this paper focus on two points, footprint and bandwidth to illustrate how regions of the space can be systematically analyzed. A good point in this paper is that it uses calculations to reason about the design space, which makes this paper more convincing. Mathematical prove can effectively show the correctness of this paper. ‘Multi-Touch Systems that I Have Known and Loved ’ shows different aspects to build multi-touch systems and then give an incomplete roughly annotated chronology of Multi-touch and Related Work. This reading material seems different from the papers before in this class because it doesn’t have a clear logic relationship between each aspect. Another difference is that it contains a large amount of information in the development of Multi-touch. I am interested in technologies and products in Wacom, after searching information in wikipedia I find a shining point in Wacom’s products. Wacom tablets use a expired patented electromagnetic technology. Since the tablet provides power to the pen through resonant inductive coupling, no battery or cord is required for the pointing device. As a result, there are no batteries inside the pen( or the accompanying puck), which make them slimmer.

changsheng liu 8:34:46 9/18/2014

<Multi-Touch Systems that I Have Known and Loved > introduces the history of multi-touch systems. It explores the different variant of Multi-Touch Systems and the drawbacks of multi-touch systems. Since users of touch systems need to use their eyes to find location on a touch screen, the accuracy of touch is important on the interaction with a touch screen. In most of the situation, you also need one hand to hold the device and the other hand to touch the screen. These restrictions could be considered as drawbacks of touch system. Some properties are mentioned to demonstrate the different types of multi touch system, for example, the touch sensitive, gestures recognition, whether it’s discrete or continuous, location specificity. After reading the paper, I feel that multi-touch techniques still has potential space to explore. The techniques are not widely available to end users until the explosion of smartphone mobile. Due to the size limitation of mobile, multi touch input has more challenges. Gesture operation is a direct mapping between physical operation and virtual operation, so it will become more popular. <Morphological Analysis of the Design Space of Input Devices> seek to comprehend different input designs. The paper compares different input devices in different aspects, such as footprint, human bandwidth, pointing speed, pointing precision, time to learn the device. Footprint means the amount of space an input device requires. Bandwidth is a merit to describe the limitation of input. Usually, the speed is a joint product of three elements: human, application and device. There are two types of input systems, Discrete Entry Devices, such as keyboard, and Continuous Entry Devices, such as mouse. It shows that mouse is the most fast input device for pointing. The author also indicates it’s possible there are input systems which are faster than mouse. I like this paper because it provides an in-depth analysis and comparison of different input devices. There is no perfect input system for all the cases, everything is better for something and worse for something else.

Jose Michael Joseph 8:50:38 9/18/2014

A morphological analysis of the design space of input devices The primary purpose of this paper is to show that previous methods of classification of input devices based on methods such as toolkits, taxonomies and performance studies were not as efficient and restricted the development to a particular region. With a newer approach like morphological design one can chart the various existing input devices based on their properties and also predict future input devices that are better than the ones that are presently used. Each of the earlier efforts in systematizing human computer interaction have led to three lines of efforts that, though impressive, still have deficiencies. The toolkit approach has a problem that it presents interface designers with many design alternatives but do little help in making the design decisions. Taxonomies were ad hoc and there was no attempt at defining completeness in design space and at times it also only included continuous devices. Performance studies, although extensive, were not always greed on as the variables had more than one characteristic assigned to it. As the paper says, the ideal solution would be to find abstractions for generating the design space and for testing the designs contained in it. To generate the design space, the approach used is to compare human interaction with any system as a sequence of mappings from one state to another. Thus these mappings form components of an artificial language. Such low level components are called primitive vocabulary components. The composition operators combine these various low level components and thus generate highly meaningful artificial sentences. The advantage of this approach is that it is simple and uses a method that humans are already familiar with (turning words into sentences). The problem though is that the model makes no way of communicating design problems that could lead to conceptual model problems. In the example of the radio using three dials, although the methods used chart the various time and complexity of the dials it makes no effort to show that having that many dials could cause a conceptual model problem for the users of the intended system. Another problem that has been encountered is that the one of the composition operators, “connect composition”, is used to map the output domain of one device into the input domain of another. It always explicitly states the device does not need to be physical. If such is the constraint then where exactly would the designer stop considering a component as a “device” and as something that is part of the OS. Thus every component of the OS that interacts with the input device could be termed a device in itself and this could lead to an error in classification. Although the paper has some flaws it does rightly talk about testing the design space on parameters such as Desk footprint and Bandwidth. In the current age these parameters are crucial as people would prefer something that is convenient and small as opposed to one that is big and takes up a lot of space. Also bandwidth clearly mentions that the performance potential of an input device is meaningless if it is connected to a muscle group that cannot be finely controlled easily.

Jose Michael Joseph 8:51:07 9/18/2014

Multi Touch Systems that I have known and loved This is an excellent paper on the various touch devices and their subtle differences. The remarkable thing about this paper is that even a layman with just basic understanding of touch devices will be able to make sense of this paper. Thus it is a paper that speaks in some level to most of its readers. The paper clearly puts forward two very important points. One is that generally innovations don’t appear out of the blue but are results of gradual work over decades by many researchers. And the other point is that each system has its own strength and weaknesses and thus will be best for some situations and worst for others. This is a very important point as it helps us to realize that what we must focus on is to find the environment in which our system has most of its strengths complemented and at the same time has most of its weaknesses negated. One of the other points he makes is that touch devices generally need a lot of attention and hence should not be used while driving. This is important as while driving the sole attention of the driver has to be on the steering wheel and a touch device is one that commends the sole attention onto itself. Although the author states that touch devices will always need a user’s sole attention due to lack of the “feel”, this is increasingly proving to be false. Apple, in 2012, patented a process by which its touch screens could give haptic feedback to the user by using electric impulses that generate certain sensations in the user. Thus such a system would enable the user to “feel” the screen of a touch screen device and this in turn could help the user to perform tasks without paying all his attention to the device. The author then goes on to chronologically go through the various innovations in touch devices that have led us to where we have today. Although the effort in part of the author is appreciated it would have been more beneficial if the author showed comparisons between the various systems of a decade or explain specifically how each system developed on the one prior to it. Eg: The author has stated that FingerWorks was acquired by Apple but makes no mention about how Apple could have possibly used that technology in their devices or what innovations it could have led to. Thus although the author’s work is extensive it feels incomplete.

Vivek Punjabi 8:52:45 9/18/2014

A Morphological Analysis of the Design Space of Input Devices: This paper provides a new line of development of systematizing human-machine input devices, Morphological design space analysis. Apart from other lines of development which are Toolkits, Taxonomies and Performance studies, the morphological design space analysis tries to comprehend different input device designs as points in a parametrically described design space. Finally, the goal is to find abstractions for generating the design space and for testing the designs contained. The two key ideas shown to model the language of input device interaction are Primitive Movement Vocabulary and Set of Composition Operators. This concept seems plausible and took most of my interest. With the help of these two ideas, we can design and analyse almost all kinds of input devices such as Radio or Dashboard. The author then describes of the testing points in the design space where he mentions several figures of merit as means of measurements. Some of them are footprint, bandwidth, pointing speed and precision, errors, user precision, etc. The examples provided were effect of muscle groups on input devices and using mouse and headmouse for selection in a 3-D information environment. More examples that could have shown are psychology of humans while selection, most common human errors and interactions by multiple humans on a single device. This study can be further extended including features like learning mechanisms, data mining and intelligent systems.

Vivek Punjabi 9:57:42 9/18/2014

Multi-Touch Systems that I have Known and Loved: This interesting article provides some of the landmarks and highlights of multi-touch systems built to date. The author uses a very cheesy and informal language to make this article more interesting, which is one of the important attributes of creating interactive touch systems also. Initially, the author gives some background of concepts that were developed way back in time and are still the fundamentals of creating touch systems. The discussions provides almost all the attributes and concepts related to a multi-touch system, such as discrete and continuous, location, degrees of freedom, multi-person and multi-fingers, objects, feedback, etc. The author also considers some of the things which get worse with advancements of this technology. Some of them are feelings, relying on both hands, sunshine, etc. As he mentions, everything is best for something and worst for something else. Finally, the most interesting sections of the article arrives which gives the timeline of the development of multi-touch systems right from 1960s till today. It starts with basic keyboards and N-key rollover in 1960s to one-point touch system in 1970s to video desk and multi-touch systems in 1980s to flip keyboard and tangible interfaces in 1990s to Handgear, plastic logic and I-phone on 2000s to finally the Surface in 2011. The section provided so many ideas and their origins that motivated me to think on their lines and come up with some amazing ideas to look forward to. I would like expand the realms of this technology by considering degrees of freedom and feedback mechanisms where the user can increase its active workspace without affecting other activities(by using transparency) and at the same time, make it more intelligent thereby reducing the need for interaction.