Customer Reviews

Programming and Deploying Java™ Mobile Agents with Aglets™

5 star:		(1)
4 star:		(2)
3 star:		(0)
2 star:		(2)
1 star:		(1)

 

 

The IBM Aglets Software Development Kit is, to me, the most promising leap to intelligent and mobile application development yet. I applaud the research and development by IBM, Danny B. Lange, and Mitsuru Oshima.

But this book creates more questions than answers, and with IBM's lack of current up-to-date documentation, I cannot see the java developer community embracing this NOW as the next logical java leap; which I think it is.

My initial excitement waned after I purchased this book. Many of the examples did not work. Written to the JDK 1.1.5, no comparison/contrast with ASDK 1.0.3 and 1.1.b2 is mentioned. Furthermore, the Tahiti server (Aglet Environment) is not even explored. To me this is a fatal oversight because an aglet cannot exist without an environment. That environment at least initially is Tahiti which comes with the ASDK the book mentions.

In short, this work is excellent in concept and poor in real-world implementation. The authors should work on a second edition. Is it worth buying? Easy. For understanding the Java Aglet concept, yes. Implementation, no.

The content covered in this book is both extremely useful and well written. The basics of mobile agents and why you would want to use them, the security issues involved and the fundamental issues facing someone wishing to implement their own system using Aglets are all examined. The code examples are all easy to follow, gradually grow more complex and demonstrate the power of the Aglets package. The latter part of the book covers more advanced topics such as design patterns found in mobile agent based applications and an in depth look at the Aglet Framework, the underlying structure behind the programmers Aglet API.

If you are currently, or are thinking, about doing anything with aglets then this book is a MUST read. Even battle hardened agleteers can learn the best way to build and optimise their aglets.

If you simple want to find out more about mobile agents and see them working within a successful framework then this book is also for you.

This is a great book about an exciting technology. The concepts are presented in a clear and concise manner. Unlike many computer-related books, it is not at all repetitive and doesn't dwell on minor issues. The programming examples are reasonable.

This book falls short in terms of its coding examples, none which worked. The book presents the coding examples in a very confusing manner. There are some errors in the book (particularly in the usage of URL on page 48, for example). The "dispatch" context is no where to be found in any of the aglet jar files I downloaded from their website. The programming examples are not explained very well and there is no discussion of what code goes with what in order to execute the mobile code correctly. As I examine the code and attempt to use it, I get the impression that the authors didn't check to see if the code worked properly. This book is a total waste of money, and my advice to the authors is to take there time and re-write the book to make sure that the code works properly, discuss specifically what "jar" files are needed for the classes used in the book, and to explain in a more thorough way with their code examples how to construct mobile agents. Also, the authors need to provide a website to download the code examples as opposed to forcing the readers to write the code from scratch. This method of obtaining code has become a standard for all the books I've bought, and saves time and effort, and allows the reader to think about how the code is working. In general, I was very frustrated with the examples in the book, none which were consistent with the aglet jar files I downloaded from the Tokyo IBM website. I hope another addition is in the works.

The book is an excellent read. When I started the book, I knew every little about Agents. By the time, I was done I knew much more. The writing is extremely clear, interesting and understandable. Unlike many of the Agent fanatics, Lange and Oshima carefully explain the problems with Agent technology as it stands today.</p>

All of this translates into a book that is useful to read even if you do not care about the Aglet technology. It is a good book for anyone who wants to understand the state of the art of Agents and see how one particular Agent implementation works.</p>

After searching a while for a reference on the Aglets SDK I was quite happy to see that this book was on it's way to be published. But I was surprised to see that the book, while covering the main aspects of the ASDK, does not cover the most important issue it shuld do : creating Aglets applications. It does have a lot of references on topics (like proxies, messaging, etc.), but it does not cover for starters the Tahiti server, the source code is confuse (not very well structured, very messy) and does not contains the imports that are necessary and package declarations, so beginners with the ASDK will be very frustrated when trying to compile the code or seeing it not working when they manage to compile but don't know they need to declare the Aglet code as a package. It's a solid reference about the ASDK, but not a good guide on HOW TO develop a simple application. I whished that the book was better than the docs provided by IBM with the ASDK, but now I know I wished it were A LOT better. I hope someday SAMS publish something like "Teach Yourself Aglets in 21 days", then starters will have a chance with Aglets, but for now, this book helps only who does have an idea about what he/she's going to work with and is patient enough to find out what's wrong with the source code at first, and manages to somehow understand all the security issues on Tahiti from the documentation alone. The motto of this review is "teach not only what you have in hands, but WHY you need it".