Seminal Ideas

From CS2610 Fall 2016
Jump to: navigation, search

Slides

slides

Readings

Additional Resources

An application of GOMS in a modern context (eye tracking and text input)

Reading Critiques

Keren Ye 20:14:38 8/31/2016

Direct Manipulation Interfaces Direct manipulation is a good form of interface design, thus the authors try to discuss both the advantages and the disadvantages of direct manipulation interfaces. In particular, they identify two factors: distance and direct engagement. These two give rise to the feeling of directness. In the first chapter, the authors explains the concept of direct manipulation interfaces by giving examples. In sum, the sensation of directness is always relative. It is often due to the interaction of a number of factors. However, we can generally sense the directness: interfaces that are classified as direct are the ones that are easier to use. For the directness, two aspects are discussed in the second chapter: distance and engagement. A short distance means the translation is simple and straightforward. Thus the better the interface to a system helps bridge the gulfs, the less cognitive effort needed and the more direct the resulting feeling of interaction. For the other aspect, the authors state that systems that best exemplify direct manipulation all give the qualitative feeling that one is directly engaged with control of the objects. For my understanding, engaging with the objects that concern us helps to make us feel more direct to the interface. In the next chapter, two forms of distance are explained. The problem of semantic distance concerns if the interface language is closed to the human language. To reduce the semantic distance, we could either optimize the system side or the user side. While semantic distance has to do with the relationship between user’s intentions and meanings of expressions, articulatory distance has to do with the relationship between the meaning of expressions and their physical form. One way to shorten the articulatory distance is to make the physical form of the vocabulary items structurally similar to their meanings. In the fourth chapter, direct engagement is discussed. Direct engagement is a kind of a feeling of involvement directly with a world of objects rather than of communication with an intermediary. In order to have a feeling of direct engagement, the interface must provide the user with a world in which to interact. In the following two chapters, the authors firstly give a graphical demonstration of all the concepts discussed in the paper (A space of interfaces) and discuss the problems with direct manipulation. User Technology: From Pointing to Pondering The improvements in humans and computers interaction accumulate into user technology, which includes hardware and software techniques for building effective user interfaces. Also, it includes the technical understanding of the human-computer relation. The main purpose of this paper is to trace some of the history of understanding of users and their interaction with workstations. In the first chapter, the authors give a brief explanation about the strands that would be discussed in the later chapters. Then, in the following chapters, each of the interfaces is discussed individually. From pointing to pondering, there are: physical interface, cognitive interface, conceptual interface, the task interface. From chapter two to chapter five, the authors explain the interfaces alternatively. They design several experiments and reach the results. The conclusion of the paper states that: 1) at physical interface level user performance with pointing devices is constrained by the information-processing capacity of the user. 2) at cognitive level, human-computer interaction does not involve problem solving but rather cognitive skills based on the execution of known methods. 3) at conceptual level, users often have mental models of the systems they use, thus the conceptual models provide the basis for users to acquire mental models, hence are important for system design. However, the mental models are cognitively intensive. At each of the three levels, it has been possible to base an applied science on a theory of the cognitive mechanisms underlying user behavior, while the most interesting problems lies in task level: understanding the nature of complex intellectual tasks and finding ways to build idea-structuring tools.


nannan wen 23:09:36 8/31/2016

User Technology: From Pointing to Pondering written by Stuart K. Card and Thomas P. Moran. Summary: In this paper, the author stated the intimate partnership between human and computer based on what they learned about users during their study. Their project is called Applied Information-processing Psychology Project(AIP). There are four strands are organized by them, which is physical interface, cognitive interface, conceptual interface and task interface. Within these interfaces, the breakthrough in physical interface are the design of the mouse by Bill English in 1967. They also did some experiment on test which method will improve the speed or error rate using direct empirical comparison between devices, but they didn’t find any, which lead to another question: why the result came out the way they did. Under cognitive Interface, they tried to figure out human information processing characteristics, they have a prototype called Model Human Processor, what it can do is compute predictions about human performance, for instance, the speed that human can read and scribble and so on. In conceptual interface, they first proposed the notion which refers to the conceptual model that the user can have of the system. For task interface, they want the user to be able to express their ideas as objects, so they proposed an representation, which can help users develop explicit mental models of idea structures. Direct Manipulation Interfaces written by Edwin L. Hutchins, James D. Hollan, and Donald A. Norman. Summary: In this paper, the author seek to make a cognitive point on account of advantages and disadvantages of direct manipulation interfaces. First the author gives some examples on what is a direct manipulation interface. There are two aspects of directness, one is distance, which means there’s a distance between the user’s thoughts and the physical requirements of the system that they use. The other one is engagement, which the feeling of the user directly manipulating the objects of interest. In relation of distance, there are two form, one is semantic distance, which measures the the possibility how well the user can say using the language. The author also provide one illustration of how to change semantic distance, which contains two basic ways. The problem for direct manipulation systems are direct manipulation interface have difficulties handling variables,

Anuradha Kulkarni 0:18:20 9/1/2016

Direct Manipulation Interface: Direct Manipulation systems aims at providing an easy to use, familiar method of interaction that helps novices to easily use these systems i.e. provide user interfaces that is in close proximity of individual’s initial expectations. With help of examples, this paper gives an insight about the remarkably powerful properties the direct manipulation interfaces possess. Directness being of prime importance, the paper exemplifies the two aspects of directness and covers the pros and cons. In my opinion, direct manipulation was clearly a large and important step beyond programming languages. However, it is not a panacea for human computer interaction as there are situations in which manipulation can be clumsy to perform. These situations need to be recognized which can further help in taking the direct manipulation to next step where the one sided programming of computers can be replaced by two sided conversations. User Technology: From Pointing to Pondering The paper gives an insight about the advances in the ways humans and computers interact, what might be called a user technology. The paper discusses about sequence of ever-broader interfaces and vision driven by the future with respect to the close proximity of the user and machine on the intellectual tasks. The paper goes in depth over the five different interfaces. This paper serves an example of a scientific approach to Human-Computer Interaction: hypotheses are proposed and then tested via user studies. Models of human behavior that are not directly corroborated are cited from earlier studies (e.g., working memory and unit-task behavior), adding significant credence to the arguments. It demonstrates the value of using simple models of human behavior: it is feasible to test them through experiments. And yet low-level theories are straightforward to apply to a wide domain of problems (e.g., predicting expert user performance when you can only observe novices). The crux of cognitive psychology is breaking down the complex process of thought into fine-grained steps than can be modeled and studied. The idea that user interface should simplify the necessary mental model of the system is also powerful, and practical advice for system designers. Today’s technology there are varied types of user interface that ranges from graphical user interface to touch user interface to gesture interface and many more.

Tazin Afrin 0:33:59 9/1/2016

Critique for “Direct Manipulation Interfaces” : The authors of this paper Hutchins, Hollan and Norman explored both advantage and disadvantages of the interfaces that have direct manipulation property in the context of cognitive experience of a user. To determine how well an interface bridges the goal and the output, the authors used two factors – distance and engagement to measure the directness of the interface. This paper is very important because it is an early investigation of the complex concept of direct manipulation interfaces. This study investigates the sensation of directness in terms of distance and engagement. The distance is described as the gulf of execution or the user’s knowledge and goal and the gulf of evaluation or the system output. The directness of an interface depends on how the system bridges the gulfs. Semantic distance, which is concerned with the user’s intention vs the meaning of an expression on the interface, reflects the gulf of execution by how much of the structure is provided by the user vs the system. The gulf of evaluation in terms of semantic distance is the amount of interpretation to be done by the user to reach the goal. On the other hand, articulatory distance has to do with the physical output vs the meaning of expression. This paper describes the key properties of direct manipulation interfaces which is the basic of some recent technology. In today’s computer we use a file system that is closely related to this concept. We have file icons and we can run different operations on these icons and feel like we are directly manipulating the files. However, the feeling of directness is relative. Directness comes at a cost of generality and flexibility. On re-doing the work, I believe I would have done it the same way but emphasize and explain more on articulatory distance in the gulfs of execution and evaluation. Critique of “User Technology : From Pointing to Pondering”: The authors of this paper, Card and Moran investigates the early human computer interaction efforts in regards to four levels of interfaces, physical interface, cognitive interface, conceptual interface and task interface, organized from pointing to pondering. The interaction of users with workstation have been traced in this study. One of their study shows that mouse as a pointing physical interface is optimal and it is hard to designing a device that is faster than the mouse. This is important because, now in 2016 new technologies have emerged and almost everything is becoming a touch-screen and mouse is becoming obsolete. So it gives scope of a new question if the new technology is optimal. Their study of model human processor shows the use of cognitive skills instead of problem solving in human computer interaction problems, which was contrary to the authors’ expectation. Hence they use a more general GOMS model instead of a sophisticated model. To understand the psychology of human computer interaction the authors give a theory of behavioral continuum between problem solving and cognitive skills. In considering task interface, the authors proposed of a developing system called NoteCards as a medium for capturing ideas. Although it is said to be open interface, it is highly specific user oriented. It is interesting that, to understand the nature of human computer interaction the authors focused on empirical user studies. Their proposed studies are fundamental to todays’ HCI understanding.

Xiaozhong Zhang 1:00:47 9/1/2016

Direct Manipulation Interfaces: The paper described an user interface design method called "direct manipulation interface". It is intended to reduce the information processing distance between the user's intentions and the facilities by letting the system providing representations of objects that behave similar as the objects themselves. The information distance can be necessary system steps to achieve user intentions as well as mental requirement needed to realize the user intentions. The distance can take two forms. One is semantic distance, which has to do with the relationship between user's intentions and meanings of expressions. The other is called articulatory distance, which has to do with the relationship between the meanings of expressions and their physical form like sound or typed characters. The article, although written two decades ago, still has merits in user interface design today. For example, it tells the UI designer to not only focus on interface conciseness but also user mental burden in realizing his/her attentions. Actually, the latest UI gadgets like VR and AR devices are both trying to reduce this distance, ideally to zero, so that user can manipulate any virtual/concrete object directly like they would do in their real life. User Technology: From Pointing to Pondering: The paper is a history review of users and their interaction with workstations. It covers mainly four strands which were developed in sequence of time. The physical interface that contains physical input and output. The cognitive interface like the limited working memory of the user. The conceptual interface like the user mental model of the computer system. The task interface which is abstract representation of user tasks of different sizes. Among the four interface, the first and fourth are more system related while the rest two are more related to user mental states. I'm not familiar with psychology or cognitive science, so I'll mainly discuss about the physical interface and the task interface. Since the paper is a history review, I just try to append some latest advances in the areas. For physical interface, as at the paper's time, keyboard/mouse and display are the latest physical interfaces. Although, surprisingly, they are still among the most widely used interface today, other interfaces like touch screen, and wearable display devices have emerged. For task interface, today's software already can not only handle large tasks like mechanical design but also connect different tasks to become a pipeline like connecting mechanical design and mechanical analysis.

Zhenjiang Fan 3:44:46 9/1/2016

Direct Manipulation Interfaces: At beginning of the paper, it gives me a feeling that it is like a survey, mainly because, it goes on for a long time to describe the tern Direct Manipulation. I am not saying that it is a strategy, on the contrary, I feel this is a great introduction in terms of the conception of the Direct Manipulation and its related topics. I think the matrix handling example is great, it well fits how to understand the Direct Manipulation, as well as other example the paper gives. The word in the term Direct Manipulation deserve more attention, and the paper makes it very clear for us to understand it. Then the paper goes on talking about the two distinct aspects of the feeling of directness: distance and engagement, which also are two very interesting and confusing(or misleading) concepts if the paper does not spend lots of time on explaining them. In the explanation, the paper again comes up with two great examples or analogies: gulf of execution and gulf of evaluation, which are two term that have been frequently used in the rest of the paper. Next, the paper goes on talking about distance, which has two forms: semantic and articulatory. And then the paper goes on discussing engagement and the concept of a space of interfaces. Furthermore, the paper also talks about the problems with direct manipulation, where I found some of problems that it mentions are trivial problems. And it fails to mention the problem of resource waste of direct manipulation in term of graphic computation, comparing to commands interfaces. And I found there are a lot of repeated topics that have been occurred in different places in the paper. That makes the paper kind of shallow, not well arranged. The two Shneiderman citations seem kind of not necessary. Overall, the paper does come up with good conceptual explanations, but it has a lot of flows, not a well-refined paper.

Zhenjiang Fan 3:44:00 9/1/2016

User Technology: From Pointing to Pondering: As a paper that was published in the middle 80’, it is both quite visionary and conductive, even though it has a few points of its views flawed from the current perspective of the field. First of all, from a broad point of view, it acknowledged the term of ‘user technology’ when it referred to the field and nature of human-computer interaction. And it had narrated a lots efforts to build the field as a real applied science(I think it is an ice-break deal for the field). Another very important point is that the paper had tried to explained every relative topics or term in a very scientific, empirical and statistical fashion, either by the data of their past experiments or by using other relative scientific knowledge(psychology, for instance) to make its hypotheses or proposed terms incredibly convincing. From the beginning of the paper, it successfully predicted the future development of the field, something that people have to deal with in the field at this very moment, decades later. And it defined the two basic factors: human and workstation(computer). The paper then talked about how the idea was initiated and how the paper is going to narrate their proposal, which the paper broke down to four interfaces: physical interface, cognitive interface, conceptual interface and task interface. Then the paper went on talking each every of these four interfaces. When it talked about the physical interface, it used the tested data to find a model that best fits the mouse by a version of Fitts’s law, which is something very instructive. But I did find a wrong assumption of it, that is, it came to a conclusion that the mouse back then was nearly optimal, which is a total wrong assumption, because we all know that the mouse has been experiencing lots of optimization since then. And then, it went on talking about an important topic: the cognitive interface. Given the cognitive interface is abstract and complex topic, it articulate the topic with several interesting terms like: cognitive skill something every human possesses, its related term Model Human Processor, the GOMS model, Keystroke_Level model, the Unit-Task. All the terms are relative, come from human behavior and can be tested by different experiments. The next topic is conceptual interface. There is one point that deserves attention, a user’s conceptual model is distinct from(but related to ) the designer’s implementation model. The paper also conceded they did not spent much time on the topic, even though they were very concerned about it and it deserved more appreciation. In talking about its empirical studies, there are a few terms that deserve more explanation, such as Homing time, Model space, Methods space and Task space, and some of its plotted figures kind of confusing. Direct Manipulation Interfaces: At beginning of the paper, it gives me a feeling that it is like a survey, mainly because, it goes on for a long time to describe the tern Direct Manipulation. I am not saying that it is a strategy, on the contrary, I feel this is a great introduction in terms of the conception of the Direct Manipulation and its related topics. I think the matrix handling example is great, it well fits how to understand the Direct Manipulation, as well as other example the paper gives. The word in the term Direct Manipulation deserve more attention, and the paper makes it very clear for us to understand it. Then the paper goes on talking about the two distinct aspects of the feeling of directness: distance and engagement, which also are two very interesting and confusing(or misleading) concepts if the paper does not spend lots of time on explaining them. In the explanation, the paper again comes up with two great examples or analogies: gulf of execution and gulf of evaluation, which are two term that have been frequently used in the rest of the paper. Next, the paper goes on talking about distance, which has two forms: semantic and articulatory. And then the paper goes on discussing engagement and the concept of a space of interfaces. Furthermore, the paper also talks about the problems with direct manipulation, where I found some of problems that it mentions are trivial problems. And it fails to mention the problem of resource waste of direct manipulation in term of graphic computation, comparing to commands interfaces. And I found there are a lot of repeated topics that have been occurred in different places in the paper. That makes the paper kind of shallow, not well arranged. The two Shneiderman citations seem kind of not necessary. Overall, the paper does come up with good conceptual explanations, but it has a lot of flows, not a well-refined paper.

Debarun Das 5:15:15 9/1/2016

“Direct Manipulation Interfaces” by Edwin L. Hutchins, James D. Hollan, and Donald A. Norman:: This is an old paper (published in 1985). It describes the concept of Direct Manipulation with the help of several examples. The paper proceeds by explaining about the two basic aspects of attaining the ultimate goal of improving the user’s feeling of directness – distance and direct engagement. Later, it goes on to describe the drawbacks of Direct Manipulation Interfaces. One of the striking aspects of the paper is that every concept is very well explained using illustrative examples in details. Also, it discusses several concepts that are still relevant today in interface design. For example, most of current user interfaces do attain the levels of balance between directness and ease of use, with proper mix of Distance and Direct Engagement. Also, some of the disadvantages that are mentioned stand valid while designing today’s user interfaces as well. One of these is that the directness of the user interface should not be improved at the cost of the ease of use by the user. This paper contributes greatly in introducing many possible frontiers of future research at that time (in 1985). ……………………………………………………………………………………………………. “User Technology: From Pointing to Pondering” by Stuart K. Card and Thomas P. Moran:: This paper discusses about building effective interfaces by studying the psychology of humans. This paper proceeds by explaining about the various types of interfaces and the approaches of understanding human behavior with respect to the design of interfaces. It starts with the example of a physical interface device (the mouse), explaining about the different experiments done before coming to a final design. Similarly, other examples are discussed that broadly throws light on the how user behavior is needed to be studied before developing a user interface. The most striking contribution of this paper is that the discussed principles are still relevant in the current era for developing user interfaces. Also, this is important because in today’s world, with the advance of technology (like smartphone apps), interface plays a pivotal role in the success and popularity of an application. Also, it emphasizes on the correct direction of research (in the task level) to pave the path for improvement in design of interfaces to “overcome user’s cognitive limitations and to aid user’s cognitive abilities with computer based tools”.

Steven Faurie 8:06:18 9/1/2016

Direct Manipulation Interfaces: The authors describe human computer interactions in a way that focuses on translating actions to be taken by the computer into simple direct actions humans should intuitively understand. A simple example would be moving a file on a modern system via drag and drop. Click the file representation on my screen, drag it to a new directory represented by a picture of a folder, etc. It’s easy and intuitive the same as moving a paper record around in a filing cabinet. I think WinForms is a good example of some of the power and drawbacks of a system that attempts to make complex actions “easy” via an interface. Microsoft has a program in Visual Studio that lets you create programs with interfaces containing text boxes, tables, graphs etc. by dragging and dropping components onto a window. A configuration file is written by the computer while you directly move and place objects into your program window. It’s a very powerful tool for letting people create modern interfaces quickly, however representing the relation between different pieces in the program requires very specific knowledge. You must know where to name the pieces in a labyrinth of menus, then you must know which label in that menu system to refer to in your program in order to access data in that piece. The portion on articulatory difference made me recall one of my first interactions with a computer as a child. I’m a little older than most of the students in the class and when I first used a computer in the late 80s they didn’t all have a mouse. You would move a cursor around using the WASD keys. Which as a five year old I found difficult to wrap my head around. I remember using a mouse for the first time and understanding it instantly. I thought it was interesting to think about how far we have come in the field of direct manipulation. Looking at your phone it seems the authors captured very well some of the advantages and disadvantages of controlling a computer so directly. Some tasks are incredibly intuitive on my phone, while others are an absolute nightmare. Pointing and Pondering: It’s surprising to me how perfect a mouse really is. This article is roughly 30 years old and they came to the same conclusion I have. That a mouse is by far the easiest most precise pointing tool available. User model of a system was interesting. The authors noted how a user’s model of the way a system works might not be accurate but is good enough for the task they’re completing. For instance a user might learn they can undo changes by closing a file without saving and reloading it. They need to have no concept of volatile memory vs hard drive. Just a basic understanding that saving something makes it permanent. It was also interesting and not surprising that users tend to default to learned routines, rather than considering an entire model of the system if they can avoid it (the calculator portion of the article). The idea of keeping the requirements for performing a task below a person’s working memory threshold is an interesting way to look at that problem. By making actions like that simple it can make using a computer almost second nature. The idea of a “Desktop” on a computer is so central to how they’re used now it is difficult to imagine one without it. Reading about using the desktop and notecard representations to augment a user’s working memory was interesting and a good point. It’s often the case I’ll have a database query displayed while I’m coding, so I can visualize the objects that need to be created for the task I’m trying to perform. Doing all of this in my head would be nearly impossible

Zuha Agha 9:51:03 9/1/2016

1. Direct Manipulation Interfaces This paper presents an in-depth study on the directness of interfaces – a phenomenon that makes interfaces feel direct and natural to the user by reducing the mental load incurred on the user during interaction and increasing the level of user engagement with the interface. These aspects of interface directness are described by two properties in the paper: distance and engagement. It is shown that increasing directness of interfaces requires maximizing engagement and minimizing distance between the user and the application interface. The paper describes interface distance to be spanned by two types of distances: semantic and articulatory. Semantic distance is the relationship between the intentions of user and the meaning of expressions while articulatory distance is relationship between the meaning of expressions and their physical forms. The paper studies these distances in the context of the gulf of execution (amount of input and flexibility available to the user to achieve his goal) and gulf of evaluation (amount of processing required by the user to understand the output). It describes the impact of reducing these distances and tradeoffs associated with them. Following that, the paper describes the concept of user engagement by describing two types of interface models: conversation metaphor model where the interface is the intermediary between the user and an unknown world and world metaphor model where the user interacts directly with objects in the world. It shows that world-metaphor model results in more interface direct applications as it allows the user to engage directly with objects and makes the user feel that he/she has more control. However, increasing engagement has its own trade offs as well and can often in clumsiness in application design. The paper has many strengths. It gives a clear and well-formulated description of all aspects of direct interface design. It not only discusses the properties of direct design but also provides the trade offs associated with each of them. It also gives a clear objective picture to the reader by discussing both the goods and vices of direct interface design. At the same time, the paper has some weaknesses. Different users have different technical competence and it is difficult to determine if the interface design is not direct enough or the user is ill-adept. It would have been interesting if the paper talked about some ways to disambiguate this so that designers don’t get undue blame or credit based on the background and technical ability of the users. Moreover, the use of automation in the section that talks about ways of reducing semantic distance can be a little misleading as the authors refer to automation as a trained user which is different from the standard use of automation in Computer Science. Another point to mention is that that the authors discuss at length about the ways in which semantic distance can be reduced but do not give details on how articulatory distance can be reduced. But overall, the paper makes a great contribution and presents seminal ideas in the field of Human Computer Interaction. 2. User Technology: From Pointing to Pondering This paper formulates user behaviors and characteristics that influence interface design based on cognitive theory. It categorizes it at the level of four types of interfaces: physical, cognitive, conceptual and task interfaces. It studies the performance of interactive systems with respect to these types of interfaces by conducting empirical studies with a certain interface type use case, describes a mathematical model for it and then derives insights from the results of predictions of model predictions and observations from empirical studies. Some traits of interfaces are also described based on theoretical observations, without empirical studies. Following is the summary of the paper’s findings for different types of interfaces. For physical interfaces such as pointing devices, the paper shows that performance is limited by physical constraints of the user such as eye-hand movement. For cognitive interfaces such as text-editing, the paper shows that the performance depends on cognitive skills required to execute the methods as shown by Key-Stroke Model applied to the task of replacing misspelled words and the example of breaking a task into unit tasks, due to memory capacity constraints of users. For conceptual interfaces, the paper presents the idea of users having a mental model of the system that allows them to predict what the system will do when certain commands are executed, what methods to use for novel tasks and how to deal with error situations. It also sheds light on the idea of ‘intended conceptual models’ that the system designer has in mind for the user, though it is difficult to show how well it translated to the actual user conceptual model. Empirical study shows that for simple tasks, users with and without knowledge of the model did almost equally well, however for complex tasks, users with knowledge of the model performed much better. The study also showed that users liked to avoid mental models by mapping their tasks directly to actions, as mental models can be cognitively intense. For task interfaces, the paper gives examples of some current graphical techniques used for task mapping and the problems of scalability associated with such task interfaces. The paper makes a significant contribution as it is one of the earliest works that describes users and their interaction with the interfaces as an applied science. However, I felt that the authors did not organize the paper effectively and was not very easy or coherent to read. The paper lacked an introduction or motivation and sometimes got into the details of describing the models without giving the reader some background. Moreover, the paper studies the interfaces in the context of Xerox Star interface, which may lead to skepticism about the generalizability of the findings of the paper with respect to other interfaces.

Alireza Samadian Zakaria 9:51:03 9/22/2016

The first paper was about Direct Manipulation Interfaces. The authors first start to talk about Shneiderman’s definition of the term direct manipulation and its benefits. After providing some examples of direct manipulation, they talk about two aspects of this matter which are distance and engagement. The Distance may exist in two parts of the system: input and output. The authors use the expressions “gulf of execution” and “gulf of evaluation” for these two distances. Each of these distances consist of two kinds of distance: semantic distance and articulatory distance. Semantic distance is about the relationship between user’s intentions and meaning of expressions and articulatory distance is about the physical form of the expressions. Reducing these distances can decrease the effort made by the user but it can also decrease the generality of the system; thus, the system may not be capable of having new features sometimes. Another aspect of direct manipulation is direct engagement. It is about the feeling of involvement experienced by user; in direct engagement, the user should feel like he is working with the object itself rather than communicating with an intermediary. The authors also provide requirements of producing a feeling of direct engagement proposed by Laurel (1986). At the end, it can be seen that direct manipulation systems have both advantages and disadvantages. One of the interesting points mentioned in this article which I have not thought about before was that interfaces can bound our way of thinking and it can be one of the disadvantages of direct manipulation since in these systems there is a specific way of doing operations which can affect our thinking about the problems.-------User Technology From Pointing to Pondering is the second paper which talks about users’ interactions with workstations. This paper has four parts and in each part the authors talk about one of the four kinds of interfaces: the physical interface, the cognitive interface, the conceptual interface, the task interface. The physical interface is the interface that we use to interact with workstation like mouse and keyboard. Studies showed that none of the devices (including joystick, step key, text key) has a better performance than mouse. They tried to find a mathematical model to show this fact and the mouse was best modeled by a version of Fitts’s Law. The cognative interface is a term for characteristics of the user as an information processor such as limited working memory. There is a model called the Model Human Processor that is a simplified architecture of the user; this model tells us every event that occurs within 100 msec will be perceived as a single event. It can be used for some predictions like how fast people read and it can be used to set the maximum velocity of a mouse. The conceptual interface is about how users learn the system and what information they have about it. By studying this interface we are able to find out what makes a user an expert one. The task interface is an abstract representation of tasks that users do with the workstations and the aim is to build an structure that make us able to build our ideas by that.