As the author says, “this book is about mixed language programming for Windows.” It was written to help Windows users develop new applications. They can use various languages, each one with its strengths and weaknesses. Some part of an application will be easily written in one language, while other parts will be tedious to produce in that language and would more easily be written in another language. Why not mix parts written in both languages? The author remarks that a lot of energy is wasted in rewriting existing programs. Therefore, he has created a system to solve these problems, and written this book to document his system.

The book first describes languages frequently used for  current  applications: Visual Basic, Borland C++, and Visual C++. For each language, the main features are presented, a relatively large example is given, and its strengths and weaknesses are discussed, helping the reader grasp the domain for which the language is best suited. The fourth chapter is about Microsoft Word, because it contains a lot of code that could be of interest in many applications, to produce reports: this code should be used without rewriting, which is what mixed programming is about. The chapter tells how to call Word from another application, and how to create Word files from other programs that generate ASCII files.

Chapter 5 is on methodology. It gives some simple, interesting rules, such as designing the application right the first time, doing the least amount of work necessary, and using standard format when one exists. Telles gives some valuable precepts about the user interface such as “avoid screen clutter,” “do not mix and match fonts on the screen,” and “do not mix colors in a single dialogue box.”

Chapter 6 starts with a nice remark about examples in programming books: “One of the most difficult things about learning a new concept from a book or technical article is that the authors tend, for reasons of space and simplicity, to choose examples which are not applicable to the real world.” Telles presents the overview of a project. Then he gives a methodology for program development: interview the customers to learn what they really want; collect the interviews to state what you think they want; return to the users and present your project with screen prototypes and report layouts, so that the users agree with your project, even if in three months what they want will not resemble the original design; decide what will and will not be in the final release of the product; write down a formal (or at least semiformal) specification of the project; and, finally, write the software. The chapter ends with the definition of the example project.

The following chapters describe the other steps of the project development: defining the entry screens and the database (chapter 7); editing data in the database, and getting records back from it (chapter 8); and reporting and printing, using Word facilities (chapter 9). The last chapter deals with interaction with the Windows program manager. AppendixA is a user’s guide to ISAM, the system produced by Telles, and Appendix B is a reference guide.

The author says, “if you are a programmer, you will find thousands of lines of source code that you can borrow for your own applications.” This is true, but thousands of lines of code are not easy to read. They are not explained in English, so that you could skip them and nevertheless learn what they do. I am a programmer, and interested in the possibility of transforming an ASCII file into a Word file. I tried to read the 35 pages of code: although there are a lot of comments, this is not easy. According to the author, “the format for Word for Windows is not easy to obtain.” I was hoping to at last know this format. I gave up reading those 35 pages. I can say the same thing for nearly all the examples given in the book. Telles criticizes those who choose simplified examples in their books, because they are not realistic. His examples are more realistic, but they are too complicated to be easily read. I am not convinced by those examples. Telles also includes comments on transparent code: here are more complex program pieces for which I would have appreciated some comments. The best feature of this book is its effort toward promoting good programming methodology. The precepts for programming are worthwhile. Programmers who read them and put them into practice should become more efficient. The programs are well structured, and generally well documented. I recommend this book to apprentice programmers. From what I can see, there are still a lot of programmers who need to be convinced to mix languages and reuse code in an application. The book is well written, clear, and enthusiastic in its argumentation, at least for the nontechnical parts. I hope it will convince those programmers to improve their habits.