 In this session, we'll look at some basic terms. Then we'll try and describe what information systems are. In the context of that information systems, we'll quickly examine what do we mean by information technology. Then we'll take an example of an information system from the industry. We'll conclude this first session by describing the organization of the course. So here are the key terms. Data, everybody is familiar with that term. Data is usually a quantified value of some attribute or characteristic of an entity. So like 14 is a number, 25 is a number, 25 alone does not convey any meaning of what data it is. So you typically have to associate something called metadata about that. Metadata is nothing but data about data. So if we say 14, it makes sense only if we say something like this is the price of sugar in rupees per kg. If we say 25, it may make sense only if it is associated with a similar description like price or, for example, age of a person. Similarly, a string like this, Deepak B. Fartak. Now, since all of you know Fartak is a human being, but it was abarakadabarakad, you would not figure out what that string represents. So there has to be a metadata saying this string is nothing but the name of a particular person. Consequently, when we talk about data, we know that it is quantified value, but quantified value of what has also to be specified when we talk about data. Information is a term which describes the next level of data. It is typically generated out of analyzing, consolidating, correlating a large amount of data together for some meaningful purpose. For example, if we evaluate sugar prices in three different shops by actually asking, then we have three different pieces of data. If we compare them and say minimum amount is 14, then suddenly the same number 14, which was only one data element earlier, now becomes a consolidated and analyzed information which says the best sugar price available from the nearby shops is 14 rupees. So I think you can see very clearly that information is something superior to just data. Information conveys something more meaningful to us and it comes out of some analysis, some correlation, some consolidation of data. Knowledge is the next higher term. Knowledge is not information, but knowledge is something that serves a purpose. That's the objective. So for example, what information is needed itself is knowledge for achieving some purpose. So for appearing for a competitive examination like gate, you need to know this information. You need to know what is the last data application. You need to know what is the syllabus. You need to know which books to study. You need to see old papers to practice, et cetera. This is the information that you need. How to get it, that's also important knowledge. And finally, how to use it effectively. An example of knowledge is correct diagnosis and treatment by a doctor for any illness that we might have. So what doctor does in terms of diagnosing our illness is not nearly data or information. The doctor does collect a whole lot of data. Could be pathological tests, it could be measurement of temperature, blood pressure, whatever, whatever. But the elements of data, even when converted after analysis into information, still does not represent knowledge. The consolidated information may say that these tests indicate symptoms of this or this or this disease. But a doctor will use all of these together, all his knowledge together to correlate, these are the symptoms, these are the diseases, these are the medicines which apply to those diseases. And probably this and this is the correct code. So knowledge means something more than information. Last but not the least is the notion of wisdom. Wisdom actually talks about knowledge which gives you multiple options and says which option is most suitable in the given circumstances. Wisdom is what we all attempt to achieve all of our lives through the experience. For example, instead of advising, let's say a very strong medicine like antibiotics, which is apparently perpetually required. A doctor might say, look, I know you for 10 years and this is not the right thing for you. You might want to do a diet control, you might want to do some exercise so that you prevent such things occurring. Now that is clearly wisdom, that is not just knowledge. These are simple examples and I believe all of you are familiar with these terms, but it is important to understand that data, information, knowledge and wisdom are actually growing, are indicative of the growing importance of what we do with the piece of information that brought around us. We loosely use terms like data and information sometimes, but this is where we wish to go. The role of information is well understood, but let me just reiterate that the information is generated out of events that happen in our lives. Events could be anything, events could be, let's say, your decision to go and attend the friend's marriage. Now that decision in your mind is an event. All the activities that you do generate information. You make an application for railway reservation and it generates information. Railway office prepares a chart for reservation per bogey for a particular train that is information generation. You send a letter to your friend saying, I am coming on such and such day, you do a hotel reservation. So every event generates information. You take the event of your admission to IIT Bombay. Someday you got admitted to IIT Bombay. Before that you would have given K, J, whatever, whatever, you're given interviews, but the day you come in, you get a new identity, you get a roll number. So new information is being generated. You get allocation to some hostel room, new information is generated. You fill up some forms, new information is generated. You meet friends, okay? They generate new information for you. You get their names, places of stay, hobbies, whatever, whatever. Essentially, when you want to take a decision in life, you often require information from multiple events. So for example, whether to take admission for M-Tech or to take up a job. Now you will analyze information from multiple sources to take this decision. So you need to tap this information, you need to know the sources of these multiple pieces of information. You need to have information processing capability within to move the data to the higher value in the chain to make it information, further make it knowledge and wisely take a decision. So decision-making requires information from multiple sources and in turn, a decision that you take causes new events to happen and new information to be generated. And this goes on and on perpetually. For the purpose of our course, we will appreciate the fundamental fact that any decision-making, whether operational or strategic decision-making requires effective handling of information. There's no doubt in anybody's mind that this is so. This is what we have been doing as human species, all of our known history, but now we need to do it in a more organized fashion because the amount of information that we cause to be generated and which is generated around us which we are required to handle has increased humongously over the years. That's the pain point. Since decision-making is the context in which we are looking at information systems or information handling, it is important to understand the temporal dimension of decision-making. Temporal as in time. The amount of time you are able to, willing to or permitted to spend on taking a decision has an important bearing on the amount of effort that you are willing to take to collect and analyze information associated with that. If you have infinite time, you can collect infinite information. If you have to take a decision in a snap second, your analysis will be limited to very basic form of information. That is common sense here. I have tried to identify some of the temporal dimensions of decision-making. So one is instant decision. Instant decision is taken in seconds or minutes. Let me give you two examples, completely different domains. Say, you find somebody drowning in a lake. Now the urge to save that life often will compel you to jump in. Provided, of course, you know swimming. If you don't know swimming, the urge will be countered by the knowledge of the body that not only that fellow will die, but idiot, you will also die if you jump in. So you will take a decision or not to. But you will agree that this decision is not taken after a deep analysis or something like that. This is a gut field decision that you will take. Another decision which is required to be taken quickly, but which may imply a whole lot of background preparation of information handling is appearing for a test where you have to answer large number of objective questions in a short time. Let's say you have to answer a question in five seconds on an average. Now do you have too much time to do detailed computations, this, that, that, nothing? You must have actually handled that type of question earlier to be able to have developed a gut field by which you can answer that. Anyway, such decisions fall under the category of instant decisions. The second category which I mentioned, answering what you call objective type questions is something that is relevant in professional life as well. You are required to take quick decisions, but they can't be gut field decision of jumping into lake or not. They have to be practiced decisions because you would have handled similar situations many times in the past, or you have all the background information analyzed and stored in your mind and you are wise enough to take a quick decision. So instant decision is, in such a case, will fall under that category. Thought-out analyzed decisions will take hours, days, months. So typically if you are working on an assignment, say assignment submission due date is one week, so you have one week to work on. Or you are working on a mini project, okay? Which will have let's say two months or three months time period to work on. So this is the next level, thought-out or analyzed decision. Research decisions are an extended form of thought-out decision. So you take all your final year MTech projects or a PhD projects, clearly these fall under the research decisions. And then finally, and more let's say notionally, philosophical decisions which take centuries milling. For example, when a human being asks, what is the purpose of my existence? My has not in me alone, but of humanity. Now you all, does God exist? I mean, these are the questions which we struggle for centuries and millennia to figure out an answer. And this is usually, although it is articulated by a few people whom we call philosophers, what they are articulating probably is the composite and combined wisdom of the entire humanity as that has been filtered through the process of interaction with people, knowledge access to other knowledge, et cetera. What we are looking at in information systems is clearly not philosophical decisions. Perhaps not even research decisions, but we are looking at the first two type of decisions which are required to be taken using information that is generated and handled by us. Here are some categories of instant decisions. Some examples I gave you. A reflex action is an instant decision. Like hand is withdrawn from a flame. Okay, there is a lot of information processing that happens by the way. If you have studied some basic biochemistry in the human body, you would know that the moment the nerves and the fingertips are affected by the flame, there is a chain of messaging that happens. It's an electrochemical messaging. So at synaptic junctions, chemicals are formed which actually transfer some electronic voltage to the next cell and so on. And the whole chain actually conveys some message to the brain. And the brain then in the reverse manner sends a message for actuating muscles to withdraw the hat. Since all this happens in a split second, we don't realize it. So this you may call to be an example of an embedded system. If you consider human body to be hardware, the software of all these action, including the networking of messaging is all embedded. So we get it for free. God when God creates us, we have all of this thing ready. Rest of it of course we better. So this is one kind of reflexive. Impulsive action like jumping to save a drowning person. Quick decisions, answering objective questions. Railway reservation, cash withdrawal from bank. They are very quickly done. Gut-filled decisions, which is experience-based instinct cannot be explained why. Decisions that need longer time. I told you analysis-based, research-based, philosophy, eternal, there are various, I think this slide merely extends what we briefly discussed so far. In this context now let's understand the need for efficient organization of information. That we have to handle information effectively and preferably we have to handle it quickly is very clear to all of us I believe. How quickly can we handle it? In the process, how do we organize the information for quick handling of information? These are some of the issues that we would like to answer in this course. So first and foremost, one of the things that human society does is that we attempt codification, systematization, standardization, et cetera to reduce decision-making time. You're familiar with grammar rules? Those of you who understand Sanskrit will know that Panini reduced the grammar to a very few simple rules of grammar so that people will be able to apply them and use correct grammar. Multiplication tables, very simple. I don't know, nowadays of course, I don't know how you get milk in your houses but when I was young, a milkman would come and deliver the milk and at the end of the month he would know exactly how much milk he has given which could be let's say 58 and half liters and he would multiply that in those days. It would be 12 anas or 1 rupee, 2 anas or something like that and he did not require a paper or pencil to calculate that because he wouldn't remember multiplication tables even including fractions. The purpose is not that person was doing something great. It was a need to be able to quickly calculate and come back so multiplication table is nothing but a codification of some standard information handled. In exactly the same fashion, forms, sorted records, files, they all represent organization of information so as to reduce decision-making time. To do lists for activities. This is not very common in India but in almost all developed societies where information is handled far more carefully than we do in a written articulated form we don't articulate our thoughts or actions in writing generally, our society has verbal traditions. However, to do lists is nothing but if you plan an activity what all things you need to do to perform that activity is to be listed there. So any event that you want to organize and you have seen mood Indigo organized you would have participated in mood Indigo you will participate in tech fest. Many of you would be participating in some of those organizations. At least some of those activities would have been documented. Step this, step this, after this, do this, if this happens do this, et cetera, et cetera. So to do lists are of that type. Job card is a component which is used by the industry to describe the actions that are to be performed for a certain activity. So job card will list one, two, three, four, five, six, do this, et cetera, et cetera. It is worthwhile at this stage to give an example of what appears to be a very mundane description but why it is considered important in an industry. The example of banking industry. Many, many years ago a friend of mine, colleague, Professor Gayathonde had gone to I think US or Europe I don't remember and he had to visit a bank, a particular branch twice in succession, once today and let's say once two days later. And on both times he found out that two different clerks whom he dealt with, they all did whatever was required but at the end they thanked him profusely for banking with them. So he thought it was a very nice gesture and he went to meet the manager to thank him that this was very nice of your bank, I mean they behaved very nicely. The bank manager laughed and he showed him a job card. He says every clerk who is handling a particular activity or portfolio has this card. For example he showed the card where the check that you give for in cashing is handled. So the instructions on the card read like this. Take the instrument from the customer that is called, the check is called an instrument. Any paper which has financial implications is called an instrument. Take the instrument from the customer and keep it to your right. Instruction number two says keep a paper weight on it. Instruction number three says on your computer enter these details, et cetera, et cetera. Okay, the last but one instruction says hand over the cash after counting. Last instruction says thank the customer for banking with you. Now this is a job card, it is considered a essential component of the, not just the training but of the infrastructure that exists for every clerk. You would not have seen that in our banks. Although such system exists in Air Force for example, maintenance technicians in Air Force always have job cards for everything they did. So what could be the purpose? After all if I'm working as a clerk after doing this 20 times, 21st time I don't have to read it. So why do you think would be the need for such a card? The other additional thing is that suppose I'm the regular clerk and I'm on leave someday. And during that time because the customers have to be handled the bank manager simply asks somebody else and says hey go handle this function. That person is not trained in this. But to the customer the behavior of the bank will be exactly identical because that fellow will very carefully follow the steps. I spent some time describing this to illustrate the importance of written documentation of the workflow. And please remember what I said. Indians are naturally very poor in this because our strength is in verbal communication. But going forward as we move towards becoming a developed society, every developed society does a very good job of written articulation and discipline of following written instruction. And I think we'll have to adopt it as we move forward. So is this clear? This is what the two do list or whatever it is. Hierarchical organization for decision making is common sense. We don't let everybody in a large organization to handle everything. So for example in our own institute we have a hostel coordination unit. We have an academic office. We have an accounts office. We have a purchase office, store's office. In an industry there will be a separate group handling inventory. There is a separate group handling HR. Another group handling accounts, et cetera. We do that for simplification of our activities essential. Having seen this now we pay some attention to the notion of information system. An information system can be formally described to be a system which effectively and efficiently handles all information related tasks. What are the information handling related tasks? This is something which all of you would know but perhaps would not have emphatically understood. So let me mention these very clear. Capture, verify and validate. This is the first task that any information system is supposed to do. Take information system within the institute. When you come in as students the first job of the information system is to capture information about you. How it is done? Either you enter that online on a computer today or if you don't have a computerized system you will fill up some forms. In all these schools or colleges that you would have attended at the point of entry you would have filled up such form. What does that form do? That form represents a mechanism for the back end information system to collect information. Usually such information when collected in the form of a physical paper will be transcribed or entered into some register file or in a computerized system. This process of transcription has two problems. One is the person who is transcribing that is a clerk in academic office for example might make mistakes. The second, the basic information that you have written in the form itself is inaccurate. It may happen. I mean we have seen so many people who apply for jobs and when we say date of application some person writes date of birth. When we say date of birth somebody writes today's date. Now if I receive the form tomorrow I certainly don't want to employ somebody who was born yesterday. So there is apparent mistake in that document. The process of verification and validation ensures that only correct information gets into my records. Whether it is file, whether it is register or whether it is a computerized database. And this is vital. If this does not happen and if funny information gets in then everything that I do using that information is liable to lead to problems. You agree with that? There is a simple principle initiated in the information system. It's called garbage in is garbage out. Garbage represents incorrect, inaccurate or incomplete information coming into the system. And the second phrase means that if you have garbage in with then the decisions that you make the work that you do with that information is more likely to be garbage. This is just to stress the extraordinary importance of ensuring that the correct information is captured and transcribed to the back-end records that you have that there is no mistake in the truthfulness of the information and there is no mistake in the transcription. Validation and verification are the two terms. One relates to ensuring truthfulness of the information. The other relates to ensuring that the correct transcription has happened from what the in-person has given. After having captured, you need to store this information and you need to store it such that you can recall or retrieve that information quickly. Obviously, this is a fundamental importance particularly when the amount of information is large because if it is not stored in an organized fashion then the retrieval may be very slow. Simple example, if the dictionary is not organized alphabetically and let's say somebody asks you what is the meaning of the word geopolitize? How will you search for geopolitize? Meaning of Zebra, whatever, some word. Today you can do that very quickly or take the result declaration, old-time newspapers used to publish results. They would publish roll numbers. Imagine if the results are published not in the order of roll number but in the order of marks which are obtained by those roll numbers. Now obviously everybody, if I am appearing with an exam I will expect to score 100 so I will start reading from the top. Probably I will find my name in the bottom. So I will have to read through all 100 or 1000 or 1 million names. But you all know we use binary search even in such circumstances. In dictionary we use a binary search to quickly go to the page which will not be feasible if information is not organized properly. What I gave you are examples of visible information to you but internal information must also be equally valid. So storage is important. Organization of storage is very important and its importance is purely from the perspective of quick retrieval of information. Any kind of storage which does not permit quick retrieval will be considered useless. The largest example will be our library. You would have visited our library many times. It has large number of books. How do you reach a book? So you have indices at the beginning. There are three card indices. Nowadays you have a computerized index. So you can give the title of the book, you can give the author of the book but that search mechanism will tell you which place that book is. Well for our library we will say which place that book is supposed to be. So you may not still find it there but this is the way it is organized. Storing and retrieving information is an important aspect of any information system. Equally important is the next aspect of analyzing, classifying, summarizing, consolidating. Here you can see what you have so far is data. You have captured data, you have verified it, you have stored it and you can efficiently retrieve it but as we all know and we have seen briefly the data alone is not sufficient. We need to convert it into information which is useful and that can happen only through these jobs. So make information useful in a given context. That is an aspect of effective and efficient information handling. Having done that of course formatting that information in the form of a report. As I said report of results declaration should be formatted in increasing order of roll number for people to easily locate there. A chart on the railway bogey could be organized either by birth numbers or by names. So that people can find it. You don't want for a full 16 bogey train to have a single chart at some place organized by whatever. It will not be very easy for people to locate. Anyway there are different ways of doing things. Then you need to ensure that information is transmitted and disseminated to the desired people. Dissemination of information is vital. For example I write a book but there is a mechanism of getting that book edited, printed, published and disseminated through channels. So there are distributors, there are book sellers through whom the book will reach you. Without that paraphernalia the information is useless. If the best of the book return which is available only with one person is useless. This is true for any information. In general even for operational information which is of strategic and tactical importance to a commercial organization the information must reach a few key people very quickly. Let's say five people are involved in making a decision using that information. Some vice president, some senior advisor, some executive director, something like that. Simultaneously you want to ensure that other fellows in the organization or outside do not know about that information. This brings in an aspect of security in the dissemination of information. So information must be made available and accessible as quickly as possible to anybody and anywhere wherever the information is required simultaneously protecting it from general dissemination unless the information is intended for general. This is also a task that information system has to hand. The last but not the least archiving information is important. You are all students here, your records are available. Now in the computerized database but in the old times or in several of the institutions which use paper records active student in an institute the records will be available in a file in an academic office. So if you are studying there for four years or five years or whatever for your undergraduate program or MSc program the people will keep that file. The moment you pass out that file will be put into a archiver storage. You have cleared the courses, you have been given a certificate but five years later you may come back and say I want a copy of my performance in the third semester. The organization should be able to give you that. Why do they keep that in the long term storage? Simply for space. There is no space to keep it in the operational store because other people who are active I need to keep their data. So archiver storage is something where the information which is important but not of day to day use is kept. In computerized system archival is done by copying onto magnetic tapes or CDs, DVDs and putting them in an index fashion in some packet. But that's an important thing. Last thing mentioned here is deletion. Deletion of information is an absolutely, absolutely important item for an information system. Why? Because information once deleted generally cannot be recreated. It is lost forever. A whole lot of history of humanity is lost because information which was created in the early days may be on the form of Bhujapatra or in early libraries in cities wherever these things were destroyed information has been lost. Consequently and why would you delete information? You primarily delete because you have no place to keep it or maintaining that information is hazardous. It costs too much money. It's not worth it. Or you think that information is truly unessential. Many of smaller details of information may actually be unessential. For example, in the old houses at least my again I refer to my younger days there is a tradition in every family that every money spent is accounted for. So every day my mother would write a diary, daily expenses, whatever we purchase she purchases whatever everything will be written. One would say what is the importance of knowing how much my mother spent in Raipur or let's say 15th March 1959 for purchase of 1 kilo of sugar. The only importance could be that if somebody wants to compare the average price of sugar in the 21st century with the average price of sugar 50 years ago then somebody might want to know what was the average price. That person is not interested in the price of sugar on 10th March 1915. That person is interested in consolidated and averaged information perhaps across the country on the average price of sugar on those days. That's the reason why even when you delete you would delete detailed records or entries but not the consolidated or averaged information which is perhaps a pointer in future. And of course my mother who is now 80 years old might want to go back to that diary to tell me you are spending too much money on sugar, I used to spend only so much but that is for her personal play. In general deletion of information should be avoided if possible and if it has to be deleted then you should be very much aware of the reasons why you are deleting it and you should do that only after taking all attempts of archiving that information in a cost effective manner. That's the lesson we learn from this analysis. In general when we talk of a formal information system for an organization we need to worry about these three things. One, the information structure and procedures. In any system what would be the workflow and steps in an activity. Let's say academic office information system in IIT Bombay or in any academic institution. There is a process of registration for courses. So what is that process? The process is all courses are listed first. Now on a website but earlier on a notice board or in a bulletin then people will fill up a form saying I register for these courses then there will be a process of validation. There is a process of endorsement by faculty advisor and then the registration process would complete. At the end of the registration process course list will have to be prepared and distributed to all teachers telling me that these students are attending your course students are attending somebody else's course etc. This is the workflow. This must be elaborately defined for every activity that the information system is supposed to handle. Then we must identify the information to be processed. In this case the information to be processed is names, roll numbers of people, course course that they register for sorts of courses etc. We must know processing logic. How to convert information from individual registration forms into course list for individual courses is a non-trivial logic. We will look at that as an exercise later in fact. And doing so what is the role of people? Who does it? Does a clerk in academic office does it? Anybody who does not have a computerized information system today also is required to produce roll list from registration information. How does that clerk do it? So what is the processing logic and which person plays which role all these has to be defined. Next technology and products. Books, pencil, typewriters, telephone etc. are all technologies which handle information. Computer system, word process, modems, switches etc. are all modern technology. Please understand that as far as information is concerned a telephone or a book or a pencil is as important and relevant as a computer and a modem as long as they are used for handling information of some kind. There is no distinction between them. Finally people. In our case of course in this course we will be emphasizing on technology and products such as technologies such as database technologies, web technologies, internet technologies. But don't forget that our generation is going through a flux where a whole lot of information exists on paper and we are trying to migrate to a situation where we want the whole lot of information to be handled digitally and we are caught in between. Large stack of information is on pieces of papers. As you can see most of you are notebooks. Some day 30 years hence all of you would be sitting with laptops. In which case there will be no paper information to be converted to digital information. But during the flux years information handled in both cases is important and more important is the handling of transition when information moves from paper to machine and machine to paper. All of these are aspects of technology and products that we need to understand with the associated process. Finally people. Operational staff, supervisors, managers and external users or customers of an information set. So a clerk sitting in a Bombay VT office of railway reservation system will have a different process to follow. A customer standing in a queue to make a reservation at the window will have a different procedure to follow. But both these processes must be understood by the respective people who handle these things. Information system in general would then require these things. How do you achieve these? Here is by the way an example of a familiar academic system. Each activity in the real world has some information system intrinsically all defined explicitly which is associated with that activity. There is no activity which does not require the use of an information system. Many times there is no obvious information system associated with that activity. In many of our informal engagements with each other we just handle things without defining an information system. Many times of course it is explicitly defined. We are of course worried in this course about explicitly defined information systems. Enrollment, course registration, result processing system, hostile mess mill computation system, organization of one study material, management of events such as sports and cultural events. These are all examples of activities which will have information systems associated with them. Without the appropriate information systems we could make a mess of these activities. Every information system defined from an engineering system perspective has one or more sources where information or data comes in. The system has some control and operations logic through which the system is operated and outgoing information will have to be disseminated to the identified destinations. So this is a standard systems view of any system. This is how an information system look like. What is important here is sources, destinations, control and operations. These constitute the information system. Please understand that in terms of the handling of information the activities that we identified are associated with the respective portions of activity as indicated here. For example, with the sources is associated the act of capture validation verification of information. With the control and operations are associated storage, retrieval, processing, archival of information. And with the destinations is associated dissemination, decision making, origin of next action, etc. You agree with this? So while the information system would have to have ability to handle all of these these are the portions for which that ability is required to be implemented. What are the main considerations that one must apply while evolving or defining an information system? This is the old adage. So some of the audience will forgive me for saying this. It's called five wives and one husband. Five W's and one H. That's the easier way to remember it. Please remember that this can apply to many other aspects of life. I am discussing or describing these in the context of information systems. For every information system that you want to conceptualize, you want to build and operate, you must ask these questions right at the outset. What or which information is required? This will tell you what are the sources, how the storage and processing will happen. Who needs that information? So which are the users, staff members, customers, people who will need that information? This will describe the output. What is the target output? When is it needed? Priority for that information? Is it urgent? It is routine. What is the nature of that urgency? Where is information needed? Location. Is the location of the person static? An office where the clock sets? Or is it mobile? What do we mean by mobile? Inventory supervisor is moving from place to place. But in order to take a decision, that person need to be told about an information very quickly. What do we use today? A mobile phone. Next level, SMS. Next level, an automated SMS generated by a back end information system which will send that information to the person when the person can take quick decision. But it is important to identify where is that information. Why is that information required? Most critical. Purpose. What are the actions which depend upon this information? What is the kind of decision making? Is the information meant for decision making? Is the information classified as FYI? Anybody knows what FYI is? For your information. So that means I don't expect you to take any decision or take any action. Somebody else is taking the action. This is for your information. CC in your email is something like this. Say FYI. Archiving. So why is that information needed? All these things. And finally, if you know what information, who needs it, when it is needed, where is the information needed, and why is the information needed? Only after you know all this, you can then think and design a system based on how you decide to provide. In the conventional sense, how is the information currently being provided that you need to understand in order to build a better information system or first time a formal information. So what are the resources? What is the plan? How will you execute that plan? You will agree that these are important questions. I would also suggest that in many of our activities, we actually don't ask these fundamental questions at the beginning. Take your research project for example. How many of you have asked a question? Why am I doing this project? How many of you have asked the question, who will benefit from the results of this project? We do that very tentatively in our mind, but we don't articulate, we don't write it down, we don't debate it, we don't argue with it. Perhaps if we did, we would have a much greater clarity on many of the things that we do. So I humbly suggest that while we are discussing this in the context of building information systems, I would think that doing something of this sort at the beginning of any activity would be extremely useful. There are some additional considerations. The additional considerations are ease and convenience of use. How easy it is to use the information. How convenient it is. Usually in terms of modern information system, this boils down to what we call a user interface. Just to give you an example of a bad user interface, Government of India many, many years ago designed a system to collect village statistics on various things that happen in the village. So it would have the number of cows you have in the village, the number of drinking wells you have, or whatever, whatever. Unfortunately, the fellow who designed the questionnaire started from North India. So in Rajasthan, he said how many camels, how many this, et cetera, et cetera. And by the time he reached Tamil Nadu, there were 158, 159, 16 questions, et cetera. The questions which were relevant to Tamil Nadu could be reached only after that clerk who was feeding information could keep saying nil, nil, nil, nil. After all, there are no camels in Tamil. So how many camels he'll have to return? By the time he pays 140 returns, he was not interested in typing correct information for the remaining answer. So as a result, all information coming from southern states was considered to be bad quality. Nobody figured out that the reason is the way interface is designed, it is not easy and convenient for that person to use it. How many times you have felt the need on modern computerized interface. You go from here to there, that page, this page. You say it's not easy. On the other hand, imagine a very neatly organized interface which immediately focuses based on your activity to specific points or questions that you need to answer. You agree with this? So ease and convenience is a fundamental requirement. In fact, in the information system that we'll design for practice in this course, this will be an important task. Cost and time required to design, implement, operate and modify the information system. You design a very elaborate system, but to build it, to design it, build it and operationalize it, it will take five years. Or it will cost, let's say, 10 crore rupees. And you are designing it for an organization or a company whose annual turnover is 10 lakh rupees. That's why I say very nice, thank you very much. Have a cup of tea and vanish. Because I can't afford this. So affordability, timeliness are important. Additional consideration of information security. Remember I mentioned that not all information should reach everybody. So there has to be some security consideration, security classification. And last but not the least, monitoring and control. So what is the efficiency of the organization by using that information system in doing day to day tasks or in doing whatever tasks which are assigned. How do you monitor that? How do you control it? So there are activity logs, there are performance measures. So it is not sufficient to say I have 10 clerks who are doing data entry. You want to know how many hours that particular person was sitting on the machine, how many characters or records or bytes that person has entered correctly so that you can reward or punish that person based on the performance. So activity logs, performance measures, etc. have to be built into the information system. They can't be put in as a later afterthought. If you have to develop such an information system, then there is a method or there is a process by which such information system is to be developed. Over the years, the development of information system has been standardized, the process has been standardized to comprise of these steps. First is conceptualization and planning. The next step is analysis and design of the information system. The third is development and testing. In the context of computerized information system, development will amount to writing programs. So when you write programs, to read data from some file, read data from terminers, processing, that is development. And of course testing means testing those programs. But please understand that nobody starts writing the program unless one has done a detailed design of those programs. Nobody starts designing programs unless one has done a detailed analysis and represented after answering all those questions. How, when, where, etc., etc. and come up with a model which has been validated. Nobody does that. Imagine the computer programs that you write. How many of you, all of you would have written some programs, right? How many of you actually have written a five-page document in plain English describing the answers to all the questions that I asked you? Then doing a detailed analysis of the situation, the context in which you are trying to write the program. Then writing a design where you design the data structures, write them down, I'll use arrays, I'll use w linkless or I'll use strings or I'll use these files. Writing down the algorithm which is called the design of the algorithm and then writing program. We normally don't do that. We look at a problem, oh, you have to find out roots of a coordinate. You start writing a program, you know, read star A comma B comma C, whatever. This is the temptation, right? Now, this temptation is good for small problems that we handle. But if we get habituated to do that, we'll be simply unable to build larger complex and robust systems to undergo this process. It's like any engineering design. That's what you have to do. And of course, as a part of the process is review, modifications, updates, etc. In the context of modern information system, the act of developing information system is governed by methodologies defined in a field of computer science called software engineering. You will notice that software engineering is one of the fields that has been mentioned here and we'll have adequate exposure to the engineering methodology of building those systems or such systems in our course. It is in this context that we wonder what is information technology. Here I have a simple definition. Information technology is any technology that assists in effectively and efficiently carrying out one or more of the following tasks in an information system. What do you see below? You see all the tasks which you already see. Capturing information, verifying, validating, storing, retrieving, analyzing, classifying, any kind of processing, formatting, transmitting, disseminating, archiving. Any technology which permits us to do any one or more of these actions is information technology. Please remember that. If I pick up a phone and dial your number and talk to you conveying some information on which you act, then what I am doing is I am using information technology. If while you listen to a lecture, make notes by pencil on a piece of paper, that pencil and that paper represents information technology. Never, therefore, ever think that information technology means computers or switches or communication. Any technology which helps us do that is information technology. Conventional technology is what we mean when we say paper, register, file, pen, telephone, fax, camera, audio, video, recorders. Modern technology is what we mean when we say computers, data communication. But the products and tools using different technologies will exist. For example, a handwritten note, plus a printed book, plus a computer database, plus network. You all use that, right? For some purpose, you use paper and pencil. For some purpose, you use web. For some purpose, you use a web-based information system at the back end in say academic office, etc. All of these will coexist for some more time. The only difference today is we don't write on Burjapatras like old style. But the paper technology is ubiquitous because it is inexpensive, it is easily available and it has longevity. It will not die unless the new digital technology replaces the paper technology in all these aspects. Ubiquitous, that means availability across, cheap, that is affordable and longevity, that is information can be guaranteed to be read even after 50 years. Does anybody know of ability of modern computers to read data which was stored on magnetic tapes 50 years ago? It's not possible. Can anybody read data which was stored on floppies 30 years ago? You may not know that, but the answer is no because those days there used to be hard formatted 8-inch floppies. Today those floppies will not even go into the floppy tracks. In fact, within a few years people will forget floppy. CD is what you use. This changing technology is causing a problem for information systems because information system by definition must provide adequate longevity for the stored information. And one of the important aspects of modern information system therefore is every few years as the technology changes it is the responsibility of the information system itself to ensure that old information stored in some old media is transferred to new media. Even though the old media is operable now the moment danger shoots that it may not be available from sometime in future I must face that. Modern technology as I said combines computers and communication technologies to complement paper technology. What is most important here is it can handle information in any form and it can permit access anywhere at any time. As far as information system is concerned the technology provides the back end the computers doing storage, retrieval, processing but the processing logic is always implemented using piece of software called application software. So for our purposes it is not the operating system or the compiler or any other tool which is relevant. It is the application software that we write. The application software could be a 5000 line cobalt program it could be a 30 line photon program it could be an HTML script it could be a Perl script whatever. Any program which makes the computer handle information which is an application program is therefore relevant for information systems. And of course we use data communication on networking for movement and dissemination. We don't need to understand the intrinsic of networking although we will try to understand some of it but we need to know how that networking is used and exploited for moving information across. That's our fundamental purpose. We represent information in graphic picture or video form which are called vision based information representation. Incidentally the symbols and letters that we use are actually graphic representation A, B, C, D, etc. The internal representation of symbols inside computers is in terms of bets that's a different representation. Graphic representation is what we do. We also use sound in spoken form. So there is a audio representation. So there is a vision based which is video and graphic representation. There is audio representation and then we categorize some part of the visual representation as symbolic representation. So symbols, numbers, letters this is the codification of that representation which is what actually it is vision plus logical interpretation of what is that visual symbol and that is the information that we are used to handling in terms of computerized information. So this is rendered graphically but represented internally by codes such as ASCII and UPSEDIC. These are names standardized over long decades which represent the binary representation of symbols inside a computer system. The information can be structured so if you look at a form the form will have columns or boxes a name, roll number course register etc. That information is called structured information. Unstructured information is a later written essay written or a book written. That is unstructured information. Valuable information exists in both structured and unstructured form and we need to understand how to analyze. And as I said there is an internal information representation. The most compact form of information representation which has not been achieved by human beings anywhere the nature achieved it. If you look at the genes the information that is represented inside the genes is an absolutely extraordinary compact codification of the most I mean the largest powers of information that you can think of. We will not reach that level at all but now we are trying to figure out how that information is stored and represented. Here is an example. Reservation for a train journey needs following activities filling up an application form scrutiny, availability check payment of fare and the reservation charges and issue of tickets. Alternately you can do a reservation through internet. Suffice it to say that even if you are on internet you will actually be undergoing the same steps. You will identify which train you want to take by form. You will fill up an internet form. Then once you submit it it would be scrutinized at the back end not by a human being but by a machine which will say seats are available in frontier mail not available whatever. Subsequently you have to make a payment. Observe that on internet if you make a payment the payment the system will take you to an internet gateway through which you will go to a bank let's say you give a credit card number the transaction will happen there and when you come back the railway reservation system will issue you a ticket which you can print. In the old days or even today in some places where internet reservation is not possible what you do? You go to a window you fill up a form and that fellow gives you a ticket. This action of filling up a form giving him the form, paying him the money after he confirms that the seat of birth is available and collecting the ticket is called a transaction. Going to a bank giving a withdrawal slip, collecting the money is also called a transaction. That's a financial transaction. This is a service transaction where you take go to a shop, purchase 2 kgs of sugar, pay him the money, collect the sugar. There's also a transaction. This is a good start. You will agree that every transaction is a very fundamental activity. We carry out up team transactions day in and day out but for every transaction you need a good robust information system surrounding all the activities of that transaction. So this is an example of an information system which we may call a transactional information system. What is the purpose of forms in such systems? The purpose of form is first of all structuring all the elemental information. Train number, destination, class, date a form ensures systematic gathering of mandatory data. Otherwise you may forget. Somebody may forget to write something but you say this information is mandatory, you fill it up. So a form gives you a structure in which you have to fill up all the information. Quick visual validation verification is possible. I mean imagine if everybody who goes to railway reservation counter submits a detailed letter to the clerk saying because my cousin is getting married in Coimbatore I need to travel and out of multiple trains I have selected this train and by the time that fellow comes to that train he would have wasted any amount of time in his time. Unstructured information in such a situation is pretty useless. In fact it is bad. That is one of the reasons why you require forms. Ease of ordered storage in a file or folder. In plain paper terms you can collect these forms and keep them. They remain a record. And of course you need these for legality of the process. Tomorrow we should not be able to say that he gave me this ticket when I asked for some other ticket. He can show you the form that you gave. So you cannot repudiate the transaction. You cannot go to the bank and say well I asked for 5000 rupees he has given me only 500 rupees. So he should be able to show you your signature and say 500 rupees. Alternately having given 500 rupees he cannot record that 5000 rupees are given. You can go back and say that look I have my copy of the slip and he has stamped it and it says 500 rupees. So legality in a transaction particularly when it involves financial transaction is important. I am introducing a new term now which we will discuss later when we discuss databases. This term is called OLTP or online transaction processing system. And OLTP system is very peculiar in that it occurs constantly as a requirement in many transactions however we often do not understand the implications of such system. A transaction has acid properties ACID. Again ACID means atomicity, consistency, isolation and durability. Let me assure you that not all of these will make complete sense to you at this juncture. But I want you to note this point as an important aspect of typical transaction. I will explain at this juncture only one word called atomicity. Let's consider a transaction where you withdraw money from the bank. So let's consider a transaction where you purchase ticket. Let's go back to that. You pay the money and that fellow gives you a ticket is the complete transaction. Consider just these two steps. Suppose you have paid the money but you don't get the ticket. Would you say that it is complete? You demand the money back if you can't give it. Can you insist that he gives you the ticket but you don't give him the money? No. So giving money and collecting ticket are two actions which must either happen both of them or nothing should happen. Atomicity means the sequence of two actions or four or five whatever may be required either all of them happen or nothing happens. And the information system must guarantee this. In physical railway booking it is guaranteed because you are standing there. So if you give him the money there are other people watching it. He can't run away with that money. He can't say, so when you give him money he will give you the ticket. In the shop you offer an exchange and he gives you the goods you give him the money but these are physically possible to be validated. In a computerized information system what happens? Suppose internet you pay money through the internet gateway and let's say ICICI gateway has taken your credit card number withdraw money and now you come back. Meanwhile the railway reservation system computer has conned out. So you go back to that page he says who are you? You lost money but not got the ticket. Is that acceptable? No. In such cases there has to be a provision whatever happens to that railway reservation computer it must ultimately guarantee by having noted a transaction ID initially for which you are paying money it should be able to say either proactively or when you complain sorry for the goof up your credit card entries being reversed whatever one. There has to be a mechanism like why? Because these two actions form an atomic group of action either all of them must be executed or nothing must be executed. You appreciate the importance of this? Let me tell you that consistency, isolation and durability are all equally important qualities of a transaction described in the context of modern databases. When we study databases and transactions we will look at these but suffice it to say that that's the importance of this term. And online transaction processing means that the transaction is processed by a computerized system which is not just one server for one organization it could be distributed databases distributed operations but whether it is a single server for a single organization or multiple these qualities of the transaction must be ensured by that system. And online transaction processing of OLTP systems from the backbone of most operational systems in the organization take your registration you are registering for a course so you register for two courses and something comes out should that be taken as your complete and final registration? No it should say sorry the transaction was not completed roll back your deregister now re-register again so every transaction must have this capability of defining atomicity of certain steps and that is an important consideration in we will see many of these systems in future. Key expectations from any OLTP system is response time to users and throughput for the service file. What is response time? If I go to railway reservation counter in Bombay VT how long do I have to wait in a queue before I get the ticket is the response time? How many total tickets are issued by a clerk in a day is the throughput for the Indian railways? Ideally we would like a system where the response time is minimal and throughput is maximum but both are important so designing an information system is the important issue for the passenger it is not the waiting time there but the total lead time when people claim that railway reservation has been computerized and the waiting time is now reduced to only 15 minutes instead of 2 hours on an average on a crowded day I will ask this question to the NNCMC chairman that you have reduced my waiting time from 2 hours to 15 minutes but it takes 3 hours for me to travel to VT and come back so what was 3 hours plus 2 hours that is 5 hours is now 3 hours plus 15 minutes but you have not substantially saved in my time because for me the total time that I spend on that activity is important today if you see on an internet that totality time is saved significant because my travel time is reduced that is the importance of OLTP systems when computerized fully in a distributed manner so modern technology helps in greatly enhancing these expectations and there is significant reduction in cost of operations significant reduction in cost for transaction for me for service provider everywhere management information system is next level which provides consolidated or summarized information for operational convenience for example coach reservation list total revenue collection in a number of tickets sold these are not relevant for an individual traveler they will agree that for railway organization these are important piece of information a coach reservation list is important for travel all of these fall under what we call management information system and then there are decision support systems or DSS as we call them which require which are of strategic importance and which require inputs truly from different sources to be churned in our mind decision makers thought about and you say ok this is what so such systems which provide inputs for decision making are called decision support system an example of such a requirement could be planning for an academic program let's say we are planning a new program in energy system so we have to worry about what is the need of such a program in this society we have to estimate resources how many faculty members will be required we have to decide on the specialization what housing will be required for them we will have to decide on the number of students that we will admit then we will have to check whether we are hostile accommodation what will be the amenities we will have to decide on the classrooms labs, equipment we will have to work out the impact on the infrastructure how much power does the campus have what is water, what is security do we have hospital base for added population what is the capital and recurring budget you will agree that this is not something which is transactional this is the extreme other end this is a transactional system this is a decision support the information systems that we propose to look at will comprise of something in this entire spectrum from transactional system to decision support system how are such systems designed conceptualized developed etc is the job that we are going to do here workflow is a dimension which I am emphasizing again workflow defines the steps and the sequence of the steps the sequence is important because dependencies are defined by the workflow for example if you are employing staff for an office deciding on the number and capability advertising in newspapers collection of applications short listing of applications interviews of shortlisted candidates appointment of staff and then you employ joining the organization these are the steps notice the dependency I cannot interview candidates unless I have advertised and collected applications so very clear dependency all these things have to be kept in mind when you design an information system to handle any activity like this