Desicritics Category: BizTech: Programming

IIM CAT Snafu Continues - Mismanagement and Inconvenience

Aaman Lamba — Mon, 30 Nov 2009 08:19:45 EST

The first online Common Admission Test (CAT) for the Indian Institutes of Management presented a poor showing for the institutes and the company conducting the test, Prometric, as problems continued to plague some test centres, impacting a number of students across the country.

Problems started on the first day of the CAT, which runs up to December 7th, with students being unable to log in to the test, and a variety of excuses being made by the test centres and IIMs, ranging from hardware problems and virus attacks. Test examiners and invigilators did not appear to have been well trained. Often, computers hung while tests were underway, creating an unfair advantage for the student. The central server is reportedly affected by a virus or malware, and it is strange there was no backup server. Fail safes and alternatives did not seem to have been sufficiently planned, and students were mostly left in the lurch with little to no clarity about students taking the test at alternate dates. Many students travel across the country, take leaves, and generally prepare mentally for the test, thus new slots might be inconvenient in various ways, and affect the general outcome.

Over 50 out of the 360 labs were reportedly closed and problems have continued on day 3. A press conference convened by the IIM Convenor and Prometric representatives seemed to have nothing concrete to offer, apart from platitudes and promises to accommodate all the unfortunate students who had faced problems. The convener seemed to have not taken care of even basic courtesies like turning off his cell phone before the conference and provided generic answers to questions, before trying to pass the buck to the test vendor, Prometric, not best practices from a management institute. As the Wall Street Journal put it, the IIMs need to practice what they are supposed to preach,

The weekend CAT fiasco, in which 50 of the 360 centres had systems fail and students could not take the test, has reinforced the perception of poor coordination and collaboration of IIMs. Going by feedback from some of the candidates who took the test , it seems they not only failed to manage the technical aspects of the test but also did a bad job in training the staff present in various centres. If good management is about reducing uncertainty and systematic errors, then the IIMs have failed that test .The way the coordination committee of the IIMs reacted to the fiasco is also a matter of concern.

The HRD Ministry, under whom the IIMs function, has sent a letter to the IIMs and the CAT Convenor, Professor Satish Deodhar, asking for a factual report on the disruptions of the Test.

The CAT fiasco also creates a problem for broader adoption of computer-aided testing, that is expected to be a major growth industry for test-happy India.

Book Review: The Blender GameKit, 2nd Ed.

Ganadeva Bandyopadhyay — Sun, 30 Aug 2009 03:49:59 EDT

The history of computer games is traced back to the autumn of 1961 when two MIT scientists developed such an application on DEC PDP-1 computer in MIT,Cambridge, Massachusetts. This was the first computer equipped with a cathode ray tube monitor and a keyboard. Other notable years include 1980 when Pac Man was released, success of QUAKE in 1996 and launch of Sony PS2 and Microsoft Xbox and their advanced versions into current decade.

Another category which is having a wide following in current times is the massively multiplayer online game (also called MMOG or simply MMO) is a computer game where hundreds or larger numbers play simultaneously. Examples of these include Toontown Online, Pirates of the Caribbean Online, RuneScape and Second Life.

There is an extensive list of game engines written in C++ programming language for most of them. These allow the game designer to easily code a game quickly without building from scratch.

Blender is the free open source 3D content creation suite, available for all major operating systems under the GNU Public License. The book under review, serves as a guide for the beginner through advanced game designers using this fully integrated 3D creation suite. This is a four-colour book including a CD having editable and playable Blender games,all files for the tutorials and Blender V2.48 for Windows,Mac OSX, and Linux.

Among the topics covered in this book is the introduction to 3D and Game Engines, Blender Basics, discussion of game concepts/games in pumpkin run, pin ball, subracer-an underwater racing game,the flying buddha memory game, “Yo Frankie!” Open Game Project, cooking basic network setups for using Blender on a network (using a matching Python installation). The book also includes a reference for the Blender Game Engine.

One of the interesting chapters in this book discusses the use of armatures in game character animation. Assuming the modelling of the character and armature is made using the correct coordinate system the game creator has to take care when working with bones, basic layout, mesh deformation vertex groups, creating a walk cycle before finally putting game logic on the armature object itself rather than on the mesh.

There are a large number of example games which can be edited for having a better understanding of the concepts as well as to use the Blender suite for more effective and advanced games.

With the large number of editable games and inputs from professional Blender artists, the book is a useful read for game designers/creators looking to use the Blender game engine.

Book Review: Android Application Development

Ganadeva Bandyopadhyay — Sat, 29 Aug 2009 21:46:15 EDT

Google initially announced the development of Android, a mobile operating system running on Linux Kernel. It is now the flagship software of the open handset alliance(OHA) committed to develop open standards for mobile devices. The OHA is a business alliance of more than 45 firms led by Google and established in November 2007.

The official Android developer documentation covers the basics of Android through publishing and best practices and some sample code.

Android Application Development is directed at Java programmers to develop useful and robust applications for Android. Familiarity for some of the technologies that Android uses is assumed as a pre-requisite for using this book to optimum. These include XML, SQL, GTalk(XMPP), OpenGL-ES, and HTTP.

There are two main sections in this book – Development Kit Walkthrough, to get started with the basics needed to write applications , and Programming Topics ,exploring the effective use of the libraries.

Most of the application development discussions are organized around MicroJobs which is this book's main sample application. The aim of the MicroJobs application is to help its user locate a temporary job in their proximity, so he/she can work for a few hours and make some money. The MJAndroid Application source code and project files are available from the O'Reilly website.

The Eclipse Java Editor is extensively used throughout this book with an entire chapter devoted to knowing its features. The discussion on signing and publishing the application to android market gives the details for using the keytool to generate a key pair and a signing certificate, getting a Map API key from Google, using jarsigner to sign the application to the last steps of signing up as an android developer paying a one-time fee and uploading the application.

The two distinct data access APIs that the Android framework offers is SQLiteDatabase and Content Provider. SQLiteDatabase is Android's Java interface to its relational database, SQLite and the latter is an interface used between applications.

Google Maps as one of the location based services is included as an application with Android itself. The MJAndroid discussed in this book uses the MapViews library. When using a MapView in the application, actual Google Maps data is used to draw the map. For this purpose the developer of the application using MapViews needs to obtain a Map apiKey from Google as a legal requirement.

Some of the other topics discussed in this book is about Drawing 2D and 3D Graphics and Inter-Process Communication. The important fact mentioned near the end of the last chapter is the keeping out VOIP from the core telephone functionality of the Android. This requires the designers of VOIP technologies for android to build add-ons instead of using the same APIs as mobile calls.

The book is a useful read in addition to the official documentation for Android developers.

Book Review: The Art of Concurrency

Ganadeva Bandyopadhyay — Mon, 3 Aug 2009 20:18:23 EDT

There is the interesting definition of a thread monkey in The Art of Concurrency. Approximately put, it is a term for the programmer well versed in writing code for parallel applications. There is the strong case made for practical use of parallel programming with the Moore's law starting to reach the limits of material causing the introduction of power efficient, practical designs of chips with multiple processor cores.

The author begins with the concepts like difference between parallelism and concurrency, the different approach to concurrent algorithms and multi threaded programming, comparison of shared-memory programming with distributed-memory programming before moving onto a discussion of two concurrent design models-task decomposition and data decomposition.

One of the central topics in this book is in the principles given as eight simple rules for designing multi threaded applications. These are the guidelines for the parallel programmer to keep in consideration for efficient code. Some of the tips are - To identify truly independent computations,Making use of the thread-safe libraries as per suitability and using the right threading model.

There are other topics such as discussion on MapReduce algorithmic framework ,concurrent implementations of graph algorithms and threading tools for the intermediate to advanced level programmer. The concurrent variations on the serial algorithms for common search and sort techniques are also discussed without much of a computer science jargon.

To sum up, this is a lucid text with practical guidance for writing parallel applications. The program code in the book is written in the C language allowing the easy understanding for readers down to the beginning computer engineering students interested in trying out the intricacies of parallel applications.

Book Review: The Data Access Handbook

Ganadeva Bandyopadhyay — Sun, 5 Jul 2009 20:15:31 EDT

The Data Access Handbook is an unarguably interesting read for database practitioners of various database implementations. Starting with description of trends in performance tuning to present times going right up to middleware issues and looking at data access in service oriented architectures ,the book covers most of the relevant topics of interest for working on performance issues in database applications.

The performance of the large-scale database implementations at the start of the 80s could be only understood by a few specialized experts having an understanding of the theoretical models and technicalities of respective database types. With the increase in database features and streamlining of the knowledge disseminated from the earlier database and performance tuning experts, shifted a lot of the performance factors into the middleware.

The authors describe in detail the trends in the database industry from the monolithic to networked environments, the emergence of the database middleware, the necessity of connectivity standards and various database drivers.

The shift of performance tuning efforts from the database management system itself(in earlier times),to middleware in the current times is the major focus of the book. Part of this focus includes an explanation of the database middleware before moving on to discussions on developing performance optimized applications on ODBC, JDBC and .NET.

There are some good technical reasons and tips give throughout the book for performance across database implementations Oracle, DB2, MS SQL Server, etc. There is a separate chapter given to the clearly outline the concepts of connection pooling and statement pooling and how to use them for performance advantage.

One of the effective tips given while discussing the SOA environment is implementing a data services layer that centralizes data access code as a service. Among the benefits of this is to allow the best practices for data access within a data access service to be identified and used frequently.

To summarize, the book makes a good technical read and provides a large number of suggestions and technical advise for improving the database application performance and scalability. It would be a better book if it could have covered some recent developments like the typicalities of database applications running on cloud architectures.

Book Review: SQL and Relational Theory

Ganadeva Bandyopadhyay — Tue, 30 Jun 2009 20:08:28 EDT

This book is intended to be an improvement over an earlier book on the same topic by the same author, namely 'Database in Depth: Relational Theory for Practitioners' . As the author states, the current book is a product of his realization of SQL being such a difficult language that it was far from obvious how to use it without violating relational principles even when covered in much depth in this earlier book, primarily for database practitioners.

One of the main features of the current book hence is the additional effort taken to have enough elucidation to connect the theory with practical SQL examples. While the theory discussion about the relational database model remains largely similar to the earlier book, there are exercises at the end of each chapter for better connecting the concepts with more examples. The answers to these exercises are included towards the end of the book.

There are about half a dozen chapters in the center of the book starting with “SQL and Relational Algebra I:The original operators” through to “Miscellaneous SQL Topics” which are the highlight of this book. There are some really lucid examples of implementing the relational definitions and theory in proper SQL in these chapters.

The slightly sarcastic tone of discussion of the author on the popular and theoretically doubtful usage of relational terms seems to be a constant companion for use of terms such as “duplicate elimination” instead of the more correct “duplication elimination”. This seems to be an attempt to challenging the maintenance of a continuous approach to try and match the SQL implementations to the theoretical definitions and concepts of Relational Model for Database Management.

One of the recent implementations of newer database models seem to be the Google proprietary Bigtable distributed storage systems. Since such proprietary models are out of public domain, it would be interesting comparing the principles behind them and the widely used relational database management systems.

To summarize, the book is a recommended read for all the database practitioners and other related professionals looking to match their implementations with the theoretical concepts and hence improving usability of the relational database model implemented within the enterprise.

Book Review: Using Google App Engine

Ganadeva Bandyopadhyay — Mon, 29 Jun 2009 19:54:36 EDT

Using Google App Engine starts with introducing cloud computing and the App Engine as a concept within the Google Implementation of cloud architectures. There is some discussion about the basic ways of developing web applications in HTML and CSS, Python and enough concepts to begin interacting with App Engine server using HTTP protocol commands. Python is to be used as a sort of controller to the HTML and CSS in front-end.

The concept of mostly HTML template to be rendered from Python code uses the Django project for the default syntax used in Google App Engine. The database that is at the back end of the Google App Engine uses a hierarchical object-oriented storage approach called as Bigtable technology. The interaction with this type of database being different from the widely used relational database models, there is an entire chapter dedicated to interfacing with this database.

The memory cache comes between the web application and the datastore and is one of the mechanisms used to improve speed of web applications. There is an interesting discussion about using the memory cache of Google App Engine as a sort of large Python dictionary.

In addition to the all these topics, the book gives a good amount of information to properly using the App Engine webapp Framework, setting up and using Google Application Engine accounts, using JavaScript, jquery and Ajax and finally, downloading and installing Google App Engine Software Development Kit (SDK) for Windows XP,Windows Vista,Macintosh and Linux Systems.

Google App Engine is part of the Platform-as-a-service approach that the Google cloud uses. Adding to the rapidly growing technical literature dealing with implementation of applications to clouds, this O'Reilly-Google Press book is dealing with making the usage of Google App Engine as detailed as possible starting from budding application developers to advanced professionals. This is a recommended read for the professionals looking to begin and ramp up their web application deployments quickly and as cleanly as possible to the Google App Engine.

Book Review: Gray Hat Python

Ganadeva Bandyopadhyay — Sun, 14 Jun 2009 20:05:32 EDT

With python being a language of choice for hackers and reverse engineers, Gray Hat Python gives an insight into using the python programming language and pre-built python tools within its covers ,saving time and effort looking on the web.

While the installation methods for Python and its tools are shown in the early chapter for both Linux and windows OS, the rest of the book uses more of Python on windows OS. There is a exclusive chapter devoted to the theory behind building a windows debugger in Python before going to discussing about PyDbg and I mmunity debuggers.

Examples of usage of techniques such as soft and hard hooking,dll and code injection and fuzzing are explained in simple ways. After discussing the basics of fuzzing and the sulley fuzzing framework, Immunity debugger is used to demonstrate windows driver fuzzing. One of the important tips from the book is to do the fuzzing demonstrations or experimentation on virtual machines so the appearance of the “blue screen of death” is not a significant issue.

One of the last topics of discussion revolves around using Ida Python in the disassembler IDA Pro 5.2 with examples of scripts. The book ends with description and usage of PyEmu- a pure Python IA32 emulator that allows a developer to use Python to drive CPU emulation tasks.

This sums up the review of this concise book of barely 190 pages covering usage of python for reverse engineering and hacking. With a bare-bones approach for the topic, the book does enough to get the interested novice up to a decent level of expertise. There is also lot of information useful for the advanced professionals looking for some fresh approach. The interested readers would do well to have sound theoretical concepts to put the tips and examples in this book to good use.

Book Review: Nagios - 2nd Edition

Ganadeva Bandyopadhyay — Fri, 14 Nov 2008 02:52:39 EST

The book comes across as a wonderful companion for utilizing Nagios- an open source system and network monitoring tool. There are twenty six chapters covering a lot of depth and variety with respect to Nagios.

There are five main sections in the book, viz. Source code to a running installation, In more detail,The web interface and other ways to visualize Nagios data, Special applications and Development. Some of the more unusual topics worth mentioning is the configuration for external notification via SMS and via email, monitoring room temperature and humidity, monitoring SAP systems via plug-in check_sap.sh and via SAP's own monitoring system CCMS and monitoring oracle database with oracle instant client.

The chapters are very concise and readable especially for the system, network or other infrastructure administrator already hard-pressed for time. One of the important facets of this book is that although there is an attempt to present the useful information, it also motivates the reader to go further and explore based on the suggestions and hints that is provided in the book.

To summarize, a highly recommended book for interesting and very useful topics in present-day IT infrastructure.

Book Review: Javascript: The Good Parts by Douglas Crockford

AJ — Fri, 7 Nov 2008 06:28:37 EST

The phrase "JavaScript master" and Douglas Crockford are considered synonymous in the web development world. When I heard that Crockford was writing a book on JavaScript, especially a guide to the better features of one of the most maligned but popular languages in the current web development industry, I was sure I wanted to read that book. I opened the book with very high expectations and unsurprisingly, I was not disappointed one bit.

With the recent explosion in the usage of JavaScript, the interest in JavaScript is at an all time high. When Netscape, which created JavaScript, released the specification of the language in the mid-nineties, it was unable to define a robust and complete specification for the language due to pressures of rushing out a production release. As a result, fair chunk of the language is not well thought out which contributes to bad programming style and promotes some bad programming practices. It is not the programmers but the language which causes this. Programming models based on Global variables, JavaScript eval, inconsistencies in variable scope, and confusion regarding how objects are created and handled in JavaScript can all be the sources of programming errors and give rise to bad programming practices.

This book, as its name suggests, focuses on the "Good Parts" of the JavaScript while cautioning the readers against the "Bad Parts" of the language. All the above mentioned "bad parts" and many other programming constructs are cautioned against in a two-part appendix.

Two other appendices also touch on JSLint, the powerful JavaScript syntax and program correctness verifier and JSON, the preferred and increasingly popular text data exchange format. These two chapters give a taste and a starter for two very important support tools for JavaScript.

However, the meat of the of the book focuses on the better parts of the JS language. In ten chapters, Crockford explains why features like - JS inheritance model, prototypes, objects, arrays and how the language handles regular expressions - are very useful and make JavaScript a fairly powerful language in its own right. Object Oriented programming in JS, how methods and the prototype chain is handled and can be used to write clean and powerful code are all a must read for advanced JS programmers.

The language of the book is very simple and sprinkled with illustrative source code which makes understanding the concept in discussion easy to understand. That said, this is not a beginners book. This book is aimed at those who have programmed in JS and have a working knowledge of the language. Nevertheless, it is a highly recommended book for anyone looking to get into better and more powerful JavaScript programming.