Input - Models and Techniques

From CS2610 Fall 2016
Jump to: navigation, search

Slides

Slides

Readings

Optional

Reading Critiques

Haoran Zhang 15:39:40 9/10/2016

A Morphological Analysis of the Design Space of Input Devices The author of this paper proposed a graphical and analytical measurement for design space of input devices. They provided a way to generate points in the design space. They use 2 dimensions to represents how designs can be critiqued, they are expressiveness and effectiveness. Also, we can use footprint and bandwidth to illustrate the method to do the systematically analyzed. By using these systematically analyzation methods, we can easily to compare the different input devices by using the results of the measurements. Multi-Touch Systems that I Have known and Loved In this websites, author explored some interesting things related to multi-touch systems. Such as Touch-tablets vs Touch screens, Discrete vs continuous, location specificity, degrees of freedom, size of target, orientation, single finger or multi finger, multi point or multi touch, point vs gesture, stylus or finger, hands and fingers vs objects, different vs the same, and the inter devices attributes, such as degree of touch, pressure sensitivity, angle of approach, and force vectors. In addition, authors also list some interesting multi-touch systems project. All projects are trying to explore a better way to deal with multi-touch system. In my opinion, actually the projects are the one who introduce the problem that the author talked about before the list. Nowadays, more and more big screen devices become popular, and it makes multi-touch task become more possible on these big screens. Because it is hard to perform multi-touch movement on a small screen. This is the reason why how to design a better multi-touch control method is important. But like what author said, there is no free lunch. It is hard to say which design is better, especially in the same generation. All solution is state of art solution, and they have their reason to make a design as they look like. And due to different target users, they may have to focus on different aspect of design principle. I think we cannot make a final conclusion of what is the best design in a short time, but so far, end users will make their call.

Steven Faurie 19:24:28 9/11/2016

Steve Faurie A Morphological Analysis of the Design Space of Input Devices: This paper describes the design space of devices as a collection of primitive movements that can be made with a device that can be put together as a sentence. The sentence is a collection of primitive operations of the device that represent some action the user wishes to take. The authors go on to formally define the primitive movement vocabulary of devices. The paper goes on to define three different types of composition operators as well. Merge composition, layout composition and connect composition. They go on to describe how a mouse is a collection of compositions. For example the xy movement is a merge composition. The collection of buttons and movement sensors that form the body of the mouse is the layout composition. Connecting the output of the mouse to the input of a computer is the connect composition. The taxonomy of input devices is also described. It is basically a plot of position, change in position, force and change in force applied to x, y, and z axes. The author plots several devices on the taxonomy. Another important aspect of devices described in the paper is the practicality of the device. The properties discussed include size, pointing speed, precision, errors, time to learn, time to grab device and use it, preference and cost. An interesting portion of the paper describes how different muscle groups are much better at different tasks. For instance, your finger is a much better pointing device than your head. Perhaps the is one reason computer mice remain so popular. They use muscles capable of very precise movements. Multi-Touch Systems that I Have Known and Loved: The author describes many different types of interactions a touch device can expect. He also describes different types of devices. For example old style touch screens that allowed a user to just push an area of the screen to the more modern smartphone type screen that lets someone drag, pinch, etc. An interesting point about a downside of touch screens, at least as they’re implemented now is that there is no tactile feedback from the device. Consequently doing something like typing on a touch screen keyboard can be frustrating. The history of touch devices described in the article is interesting. Seeing that the first multi touch table that used a capacitive screen was from 1985 was surprising.

Tazin Afrin 0:30:31 9/12/2016

Critique of “A Morphological Analysis of the Design Space of Input Devices”: In this paper Card, Mackinlay and Robertson discussed the morphological design space for different input devices. They have also summarized the generation of design space and testing points of input devices such as footprint and bandwidth. They have also compared between the mouse and headmouse and their calculation shows that mouse is more effective than the headmouse. Some previous works provides three development lines on human-machine input devices, they are toolkits, taxonomies and performance studies. Toolkits sketch only a limited picture of design space, taxonomies do not define the design space completely and performance studies do not agree as a whole. Hence the authors propose a morphological analysis of design space both for generating and testing. They propose that the interaction between human and computer can be modeled among at least three agents, a human, a user dialogue machine and an application. They describe primitive movement vocabulary and composition operators to express the sentences in an artificial language and to combine vocabulary into a combinatorically richer set. The two basic criteria to evaluate the input devices are expressiveness and effectiveness. They also show by calculation that small footprint is better than a bigger footprint. The systematic analysis of the semantics of communication in this paper suggests a promising direction for developing a better device than mouse which I found very interesting. Maybe in future more systematic analysis would open the door for more direct communication between a machine and the human at a high level language. ------------------------------------------------------------------------------------- Critique of “Multi-Touch Systems that I Have Known and Loved”: Bill Buxton gave a detailed history of multi-touch systems in this article and how this technology has been evolving for last few decades. The author discussed all the research advances towards the new technologies. Different attributes and features and also short coming of touch screens have been discussed too. Now-a-days almost all the smartphones, tablets and pads are incorporating touch screen. It has become one of the essential feature in todays’ technology. It is so widespread that, maybe in future a computer screen without touch technology will become obsolete. The different considerations and attributes discussed in Buxton’s article are important when developing a multi-touch screen. However, till today there is no single decision design that maybe suitable for every task on a multi-touch screen. According to the author, we are more comfortable while drawing on a screen using a stylus instead of finger tips. Also touch screen requires more attention, so it might not be very suitable while driving. I found this article and the way of the author’s description very interesting. The most important information that I received from this article is how important it is to have a universal design for a multi-touch screen and how much scope is there to improve.

Zhenjiang Fan 18:47:17 9/12/2016

Morphological Analysis of the Design Space of Input Devices::: Based on previous studies and engineering disciplines, the field of human-machine interface has developed enough theories both abstract and subjective to have a systematic summary or conclusion on how we design human-computer interfaces through so called design space. Before this work, they had developed a set of abstractions that could provide potential or basic theory works to systematize the human-computer interface design, such as toolkits, taxonomy and performance studies. The last step, or the last step as the author argues, a morphological design space analysis, can be used to integrate all the previous work to finalize the whole definition of the design space. The paper uses a very valid and interesting example or analogy to vividly demonstrate how the paper is going to define its design space subjectively. The paper also uses some abstract concepts too to build an image of its design space, by describing the design space has kind of semantic meaning or structure. By mapping the design space with semantic sentences, the paper could break its subject down into several segments, which in turn bring all the relevant terms or composition operators of the design space. Then the paper represent the input device as a six-tuple, which includes (M, In, S, R, Out, W) six factors in the design space. The example of a simple radio device provides a perfect cover of all the paper’s proposed ideas and terms, how things are mapped between abstract concepts to physical controls, how the actions produced by those physical controls translate into application level commands. The paper spends lots of time on the test of its conceptual propose, which I think it is not necessary. But It did successfully connect its propose to the popular theories in the human-computer interface field. Multi-Touch Systems that I Have Known and Loved::: According to the author, multi-touch technologies have been evolving for decades since the term was mentioned, given the fact that most people think they are relative new technologies. The author gives a similar example of development of the mouse to demonstrate that “new” technologies like, multi-touch, do not grow out of a vacuum. Then the author tries to explain why it took so many years for such technologies to finally be put on the market. Then author gives some distinctions between relevant terms that seem to be similar in the technology field, but they differ greatly on lots of levels. I think every term that the author mentions in his work is worthy to be noticed, discussed and studied by researchers in the field. It is important because we should make such distinctions and discussions for further researches on the topic. I agree with the author on the thought that multi-touch technologies greatly simplify the ways we interact with computers. In comparison with the old-fashion interaction with computers, multi-touch helps us get rid of physical devices that we have to carry and learn in everyday life. And it also expands our knowledge about the human-computer interface study and opens an unexplored area for us to focus on. And I also agree with the author that touch-screen devices with on physical controls have a lot of strengths, given the fact that they too demand many unnecessary attentions just like the author mentions in the paper, they bring more flexibilities on how we configure the display dynamically, which is a great thing, because in that way, we can improve a better user interface experience.

Xiaozhong Zhang 20:42:57 9/12/2016

Low-cost multi-touch sensing through frustrated total internal reflection The paper introduced a HCI approach to realize the multi-touch sensing display. The technology uses the frustrated total internal reflection (FTIR), which was introduced decades before the paper and has since fallen out of usage in the world of technology. However, the author argued that the FTIR can be again of use, because the modern machine vision technology has advanced to be capable to do realtime FTIR signals processing. Then the author introduced five variants of the FTIR implementation based on their introduction chronologically. These implementations have different setup and usages. The author then talked about their display implementation, which adopts two of the previously-mentioned methods. The first one uses a glass board as both the touch board and the display board. The second adds a compliant surface onto the glass board. The compliant surface is deformable and will get in contact with the glass board when pressed. The implementation has many points to be improved. Being camera-based, it requires a significant amount of space behind the interaction surface. When adopting first method, this scheme introduces a disparity between the display and interaction surfaces. The response of the sensor is highly dependent on the optical qualities of the object being sensed, which means that the device may not detect gloved hands or arbitrary styli. When adopting the second method, it can take up to a full second for an excessively forced depression to completely dissipate. Therefore, in my opinion, the implementation is still far from industrialization. A Three-State Model of Graphical Input The author introduced a state machine with three states for graphical input devices. Before introducing the new model, the author mention two old models, which both have two states. The first one, which is basically designed for input device like mouse, has states of tracking and dragging. The other model, which is based on touch tablet, has states of out-of-range and tracking. The new model has combined the two together to build a state model with three states i.e. out-of-range, tracking and dragging. Then two tables are presented. The first summarizes the demands of a number of transaction types expressed in terms of the states and state transitions that they require from a supporting transducer. The second summarizes the capabilities of a number of input devices, expressed in terms of this same type of state information. Then, the author discussed about some pros and cons of using certain input device for certain state or transition, e.g. with trackball, maintaining continuous motion in State 2 requires holding down a space-bar like button with the thumb, while operating the trackball with the fingers of the same hand. Consequently the hand is under tension, and the acuity of motion is seriously affected, compared to State 1, where the hand is in a relaxed state. The model builds a good design guideline for graphic input devices. The state model in the paper is still suitable for the input devices today like touch screens and touch pads. Certainly, the model also has its limitations like that it cannot represent pressure information in the states.

Keren Ye 22:57:15 9/12/2016

A Morphological Analysis of the Design Space of Input Devices This paper focus on systematization of existing research results of human-machine interface technology. Apart from the previous work that states three lines of development (toolkits, taxonomies, and performance studies), the paper argues that a morphological design space analysis can be used to integrate the results of this previous work, in which it seek to comprehend different input device designs as points in a parametrically described design space. To generate the design space, the paper states that the interaction between human and computer can be modeled as the interaction in an artificial language among at least three agents: 1) a human, 2) a user dialogue machine, and 3) an application. There are two ideas in modeling such a language of input device interaction: 1) a primitive movement vocabulary and 2) a set of composition operators. Then, the authors define the design space that it is “basically the set of possible combinations of the composition operators with the primitive vocabulary”. Based on the definition of the design space, the paper later states how to test points in the design space in order to characterize regions of it. Two criteria is then proposed: 1) expressiveness and 2) effectiveness. The authors later illustrate how they can annotate the design space of input devices and discuss two of the figures of merit of effectiveness: footprint and bandwidth. In sum, the paper illustrates a way of systematizing knowledge about input devices. It has provided a method for helping to generate points in the design space. And it has shown how designs can be critiqued in terms of expressiveness and effectiveness. Multi-Touch Systems that I Have Known and Loved In the web page, the author explains his opinion towards the multi-touch systems. The key points that the author holds are: 1) Multi-touch got traction very fast; 2) Everything is best for something and worst for something else; 3) A single new technology is seldom the cause of a product's overall success; 4) Input techniques and technologies, in terms of investment and attention, have played second-fiddle relative to displays. For the purpose of explain the idea of designs of multi-touch systems, the author then makes some distinctions among some key ideas. The general opinion is that “touch screen interface”can mean so many things that, in effect, it means very little, or nothing, in terms of the subtle nuances that define the essence of the interaction, user experience, or appropriateness of the design for the task, user, or context. However, the interaction, user experience, or appropriateness for the task are quite important factors of design. It seems the multi-touch systems have many advantages yet the author states it may have some disadvantage such as 1) the system relies on both hands, 2) your finger is not transparent, 4) drawing using finger is not effective, 5) it has bad display under sunshine (or it does not work for blind people). In the last part, the author summarizes the evolution of the multi-touch systems.

Alireza Samadian Zakaria 23:28:29 9/12/2016

At the first paper, the authors suggest a line of development for input devices. At first, they introduce 3 previous lines of developments which are toolkits, taxonomies and performance studies and then introduce their line of development. They have developed a vocabulary to represent the input devices as a six-tuple. This vocabulary is useful but I think it is not complete enough to tell the differences between a touch-pad and a mouse. After that, the article talks about expressiveness and effectiveness and define both of them. Regarding effectiveness, it provides 6 figures of merit for an input device and talks about two of them in details. One of the conclusions that the authors have made is not good in my opinion. They have concluded that we should not focus on devices which are interacting with neck muscles since these muscles are not fast enough. However, in many applications such as virtual reality using these kinds of inputs are very useful. Actually, everything is the best in some applications and the worst in some others. That is one of the points in the second article which was written by Bill Buxton. Bill Buxton’s article talks about multi touch devices and tries to demonstrate the fact that the device is able to measure multiple simultaneous touches is not enough by itself but the application and other aspects of a multi touch device is important as well. At first, the author mentions that the multi touch technology was invented before but it was not used in any common devices. Then he demonstrates, the differences between some alike concepts. The difference between multi-touch and multi-person was interesting to me since there can be many good gaming applications for that. At the end, the author talks about some of the shortcomings of touch screens which I previously felt them when I had just bought my first touch screen phone.

nannan wen 23:44:57 9/12/2016

Multi-Touch Systems that I have Known and Loved by Bill Buxton: The propose of this article is for the author to give some detailed reply to each question he’s been asked by others. In this paper, he gives some summary, though incomplete, of some of the landmark examples that he have been involved with over the years. He gives some examples on traps, like Touch-tablets vs Touch screens, Discrete vs Continuous. They also designed the interaction so that control is asserted by continuous actions or gestures. In Location Specificity, the author asked a question that how accurately the user has to position a touch at a particular location for a particular action has a significant effect on the nature of the interaction. He also talked about degrees of freedom, size matters, orientation matters - horizontal vs vertical, there is more to touch-sensing than contact and position (degree of touch / pressure sensitivity, angle of approach, force vectors), size matters 2, single-finger vs multi-finger, multi-hand vs multi-finger, multi-person vs multi-touch, points vs gesture, stylus and/or finger, hands and fingers vs objects, different vs the same (device-independent graphics, the interchange of devices is not neutral from the perspective of the user, feedback). There are some attributes, like the path offered by touch-screen driven appliances. There is no free lunch, the adaptability of touch screens in general, and multi-touch screens especially comes at a price. User interfaces are about look and feel. The he gives an incomplete roughly annotated chronology of multi-touch and related work, in this, it contains some finding in a very early step, it has some example in electroacoustic music, touch screen technology, touch screen terminal, one-point touch input of vector information, tactile array sensor for robotics, flexible machine interface, soft machines, video place/ video desk, multi-touch screen, multi-touch tablet, sensor frame, bi-manual input, apple desktop bus and the trackball scroller init, bidirectional displays, digital desk, flip keyboard, simon, wacom etc. A Morphological Analysis of the Design Space of Input Devices by stuart k. Card et.al. This paper discusses a method to systematize different devices through morphological design space analysis. In the introduction part, the author points out that the design of input devices has been developed to the point where it is appropriate to systematize existing research results and craft into a body of engineering and design knowledge. Toolkits, which can help with a wide range of problems, including the construction, run time execution, and post-run-time analysis of a user interface, can help systematize input device knowledge by providing a library of prebuilt input device modules. Taxonomies, performance studies, morphological design space analysis, mackinlay, and card and robertson can also help. Then the author talked about how generating the design space. After descrip primitive movement vocabulary, composition operators, the design space for input devices, he explained testing points in the design space, then he give us some bandwidth, footprint, and some functions on Fitts’ law, some examples on effect of muscle groups on input devices etc.

Anuradha Kulkarni 6:43:48 9/13/2016

A Morphological Analysis of the Design Space of Input Devices: This paper mainly discusses and presents the morphological classification and ordering of the input devices in a design space. It represents a graphical and analytical framework for the design space of various input devices, that can be used to compare devices based upon their inherent limitations from human to device bandwidth and from device to task completion. This framework is used to display the current input device design space and take results from the performance studies. This helps in characterizing regions of the space. This paper compares the performance of devices in a standardized way and give a means to drive new innovations using the areas of design space. Comparison is done with respect to footprint, human bandwidth, point speed, pointing precision and the time taken to learn to use this device. In my point of view, the paper provides a novel way of modeling different input devices based of their effectiveness. Good comparison examples have been presented in the paper. The gist is to keep the model as simple as possible. Multi-Touch Systems that I Have Known and Loved: This article provides a history of multi-touch systems, the variety in multi-touch systems that are possible today and the drawbacks of these variety. It also gives an insight about the properties of the multi-touch systems like the touch sensitive, gestures recognition, location specificity and specify either it is discrete or continuous. The article makes an important point that no single decision is suitable for every task. It is supported by an excellent example i.e. tablet input using stylus versus using the multi-touch screen. The direct mapping between the physical operation and virtual operation has made these systems to gain popularity. The paper is interesting giving us a background about the multi- touch systems.

Debarun Das 8:55:44 9/13/2016

“A Morphological Analysis of the Design Space of Input Devices”: This paper aims for design space analysis of input devices to compare and critique them mainly in terms of effectiveness and expressiveness. Two effectiveness metrics – Footprint and Bandwidth have been used to for systematic analysis of regions of space. Footprint refers to the amount of desk space required by an input device. Bandwidth refers to limit of speed of use of a particular input device. It is generally a combination of three parameters – the human, the application and the device. Thus, it can be used to analyze and compare the input devices by calculating and generating interesting regions of design space. This is beneficial as future researchers would know which feature to be exploited further for more effective input devices. As mentioned in the paper, exploiting and improving the bandwidth of fingers can help to beat the performance of the mouse too. This idea has been further exploited in the future to develop the modern multi touch systems. This paper thus served as an important one as it highlights the effectiveness and shortcomings of all input devices and also gives direction for future research in this field. ………………...................... “Multi-Touch Systems that I Have Known and Loved” : This article discusses about the details in the design of multi touch systems, critiques them and also discusses about the chronology of these devices over the years. The most interesting part in this article for me are the discussions on the different design choices that are to be made for different devices based on their usability and situations in which they are to be used. Some of the very interesting examples include the situations where a stylus is better to be used in multi touch screens and situations where finger(s) are better to be used. Stylus would obviously be better off in situations where precision is required (like typing fast). But for playing games (like “Temple Run”, where quick reflexes are required or where the precision required is not so high, fingers would serve a better purpose. Another good discussion is about the ineffectiveness of using touch screens for car radio. This hinders the concentration of the driver as accessing a touch screen requires shifting focus of attention to the screen using eyes. Also, unlike traditional car, touch screens do not provide “tactile feedback” to develop muscle memory. Thus, using touch screens for car radio is an ineffective choice. Finally, the list of all multi touch systems till now gives a good view of the progress in technology in this field.