Computational Thinking - Учебное пособие, страница 2

It is a major change agent in humansocieties throughout the world. Initially, manycomputer scientists were interdisciplinary scholars,studying both CIS and deep applications of this newdiscipline in other disciplines. Eventually, CIS grew inboth breadth and depth, and it became an importantdiscipline in its own right.• Sub-disciplines were developed such as analysis ofalgorithms, artificial intelligence, computability,databases, networking, and so on. Moreover,computers became more and more cost effective, andthe whole field of Information and CommunicationTechnology (ICT) blossomed.Computational thinking as a discipline is far more than acollection of isolated pieces.

Learning a discipline andlearning to use or apply a discipline at a high level are farmore than learning isolated facts, tools, and ideas. As adiscipline grows and matures, its leaders give considerable8thought to identifying unifying themes. Computationalthinking is a unifying theme in the computer field and in theuses of computers in every discipline. Using a computersystem involves telling the system what you want it to do. So,to get started you first need to think about what you are tryingto accomplish and what parts of the task the computer canhelp with.

You need to understand the capabilities andlimitations of the computer system that will be relevant toaddressing the problem that you have in mind.Computational thinking refers to people and computersworking together to solve problems and accomplish tasks.Computational thinking builds on the power and limits ofcomputing processes, whether they are executed by a humanor by a machine. Computational methods and models give usthe courage to solve problems and design systems that no oneof us would be capable of tackling alone. Computationalthinking confronts the riddle of machine intelligence. Whatcan humans do better than computers, and what cancomputers do better than humans? Most fundamentally itaddresses the question: What is computable? Today, we knowonly parts of the answer to such questions.Computer science interacts with almost every other discipline.Computationalbiology,computationalchemistry,computational design, computational finance, computationallinguistics, computational logic, computational mechanics,computational neuroscience, computational physics, andcomputational and statistical learning are just a few examplesof such interdisciplinary fields of study.

In education,undergraduate computer science curriculum and outreachprograms teach students how to think like a computerscientist. The message is that computer science is not justabout programming, but about thinking. Our long-term visionis to make computational thinking commonplace foreveryone, not just computer scientists.A computer program is a detailed set of instructions that canbe interpreted and carried out by a computer. A computer is a9machine that can quickly and accurately follow (carry out,execute) the detailed step-by-step set of instructions in acomputer program. Computer programmers design, write, andtest computer programs — so they are deeply involved indoing computational thinking.

However, all computer usersare involved in computational thinking at some level, as theyinteract with a computer and tell it what they want done. Thisis true whether you are playing a computer game, retrievinginformation from the Web, or using a word processor.The underlying idea in computational thinking is developingmodels and simulations of problems that one is trying to studyand solve. We are all familiar with the idea of developingmental models — we form mental representations of aproblem and often we "play the mental images" in our heads,doing a mental simulation. We are also aware of the value ofdeveloping a mathematical representation of a problem. Thevalue of math modeling lies in the huge accumulation ofknowledge about solving a wide range of different mathproblems.

If a problem can be represented mathematically(that is, if a math model can be developed for a problem)then this might well prove to be a powerful aid to solving theproblem.Computer models have some of the characteristics of mentalmodeling as well as some of the characteristics of mathmodeling and the types of modeling done in other disciplines.If a problem lends itself to computer modeling, then thecomputer may well be able to carry out the steps (procedures,symbol manipulations) needed to solve the problem.Thus, computational thinking, integrating human thinkingwith the capabilities of computers, provides a powerful newway to solve problems.

The computer aspects ofcomputational thinking require one to know the capabilitiesand limitations of computers and how one communicateswith (interacts with) a computer system. From an educationalpoint of view, a key aspect of studying any discipline nowincludes:10Learning some of the capabilities and limitations ofcomputers as an aid to representing and solving theproblems of the discipline.• Learning how to actually make use of these computercapabilities.• Learninghow to think about problems in thediscipline both from a traditional point of view andfrom a point of view of the possible uses of computersto help solve the problems.Computational thinking has already influenced the researchagenda of all science and engineering disciplines. Startingdecades ago with the use of computational modeling andsimulation through today’s data mining and machine learningtoanalyze massive amounts of data, computation isrecognized as the third pillar of science along with the theoryof experimentation.

The volume and rate at which scientistsand engineers are now collecting and producing data –through instruments, experiments, and simulations – aredemanding advances in data analytics, data storage andretrieval, and data visualization.Computational thinking has also begun to influencedisciplines and professions beyond science and engineering.For example, the areas of active study include algorithmicmedicine,computationalarcheology,computationaleconomics, computational finance, computational law,computational social science, and digital humanities. Dataanalytics is used in training Army recruits, spam and creditcard detection, recommendation and reputation services.Computers have had a much larger impact in somedisciplines than in others. The search for aids to the symbolmanipulation (the computations based on algorithmicheuristic procedures) has been going on for many thousandsof years.

The abacus was a very important early success — anaid many use today. Math tables, logarithms, the slide rule,and mechanical calculators were all quite important aids toarithmetic computation. Now, we have electronic digital•11calculators and computers. We have quite inexpensive solarbattery-powered scientific calculators that are superbimprovements over earlier aids to numerical computations.We have computers that can do the types of symbolicmanipulation needed in many different algebraic, calculus,and statistics computations.Computer systems can graph functions and statistical data,but they can also do graphical displays of the types of modelsused in many other disciplines.

For example, an architect candevelop a computer model of a planned building. Thecomputer can do computations to test the structural integrity,fire and storm resistance, heating and cooling requirements,and so on. A computer system can provide users with a threedimensional structured walk-through of a planned buildingbefore physical work begins.

These are all humongouscomputational tasks, based on accumulated knowledge inmath, physics, and other disciplines.Information as in Information Age has different meanings todifferent people. The following quote is from Marc Prensky(February/March 2009): “Digital technology, I believe, canbe used to make us not just smarter but truly wiser. Becauseof technology, wisdom seekers in the future will benefit fromunprecedented, instant access to ongoing worldwidediscussions, all of recorded history, everything ever written,massive libraries of case studies and collected data, and highlyrealistic simulated experiences equivalent to years or evencenturies of actual experience...But the digitally unenhancedperson, however wise, will not be able to access the tools ofwisdom that will be available to even the least wise digitallyenhanced human.”Humans and computer systems, working and "thinking"together, are now routine approaches to problem solving.Computer hardware capabilities are continuing a rapid pace ofimprovement (See Moore’s Law).

A steadily increasing rangeof problems can be solved through computer modeling andmaking effective use of the capabilities of computers. A12steadily increasing amount of the accumulated knowledge ofhumankind is being put into a digital form so a computersystem can do part of the work of retrieving and making useof the information.Cyber-Enabled Discovery and Innovation (CDI) is NSF’sbold five-year initiative to create revolutionary science andengineering research outcomes made possible by innovationsand advances in computational thinking. Computationalthinking is defined comprehensively to encompasscomputational concepts, methods, models, algorithms, andtools.

Applied in challenging science and engineering researchand education contexts, computational thinking promises aprofound impact on the Nation’s ability to generate and applynew knowledge. Collectively, CDI research outcomes areexpected to produce paradigm shifts in our understanding of awide range of science and engineering phenomena.Computing has permeated and transformed almost all aspectsof modern life. As computing becomes more important in allsectors of society, so does the preparation of a globallycompetitive workforce able to apply core computing concepts,methods, technologies, and tools - referred to here asComputational Thinking (CT) - to a broad range of societalchallenges and opportunities.

CT capitalizes on concepts,methods, technologies, and tools fundamental to the fields ofcomputing, i.e. computer and information science andengineering. For example, computing concepts and methodsequip us to reason at multiple levels of abstractionsimultaneously, to think algorithmically and applyfoundational mathematical concepts to solve complexproblems, and to understand the dimensions andconsequences of scale.