Microsoft BizTalk Server 2004 Developer's Guide

Rating:
Summary: Introduction to BizTalk features
Review: Background: I have been working on BizTalk 2004 from quite some time. I learnt most about BizTalk 2004 from BizTalk community.

Detail:
1. As scott points out in one of his presentations, each feature of BizTalk (might) deserve a book. So putting all those topics into a 700 page book does not make justice to those topics. While going through this book, I never felt that I was learning some thing new.

2. In my opinion its always a bad idea to get those many Authors together for a single book. While it makes sense that the product is new and its not possible for a single person to write a book in the given time frame, its a terrible idea and makes the book boring when those many people are involved.

3. BizTalk is an EAI and B2B tool. It is a huge product and is/will be used in critical scenarios. For EAI or B2B applications having a good understanding of the 'infrastructure' and 'configuration' needed are critical. This book does not provide that. I would like to see an even more in-depth book about BizTalk. And where are case studies?

Good for: New BizTalk developers (Who just started) or for those who are lazy or busy to go through all the other sources available
Not for: Any one with 2-6 months of development experience or who had gone through most of the BizTalk blogs and Scott's Tutorial and fair amount of product documentation.

Conclusion: Good for some one who wants to know what BizTalk 2004 is and its features. Since its just 30 bucks, its not waste of your money. But dont expect a lot from this book.

My rating is based on the content and composition of the book and not on my expectations of the book.

Rating:
Summary: Biztalk Beginner Book-
Review: For a beginner book, the book does have alot of explanation.
The books requires a lot of fiddling to get the examples to work.
So far this is the best book out for biztalk server 2004. The book is very comprehensive and a very good tutorial.
The explanations are a blast!

It would be nice if the following solutions to the below problems could be provided. They are both very important
techniques.

The example submitting to a webservice enclosing an orchetration using infopath does not work p426-428

The example in chapter 19 p598-599 does not work since the information to configure the machine to handle msmqt and msmq
could not be found

Rating:
Summary: Same standard as other Biztalk documentation
Review: Having spent a (painful) month with BizTalk now, I have to say there is nothing in this book that has actually helped me with any of the issues I've encountered.

BizTalk Unleashed covers a lot of ground by not resting in any place long enough. Expression editor capabilites? Half a page won't help. Correlation giving you trouble? Two wordy pages won't yield any greater understanding. Need to know about configuring the native adaptors? The appendix chapter C isn't in the book or online yet. The list goes on and on.

This book is only useful as a feature introduction, and even then the webcasts offer a better use of your time for this purpose. Webcasts; samples; blogs; newsgroups and just old fashioned experimenting are still the only way to get started with BizTalk.

Rating:
Summary: Not advanced topics
Review: I had been waiting this book for a lot of time, but the topics are for introduction to biztalk.I found a lot information (tutorials, hand on labs, presentations, etc.) about advanced topics in internet that the authors didn't go deep.

The Appendix C (online chapter) isn't published at this time.




Rating:
Summary: Grossly over rated
Review: I've worked with BizTalk since its inception and awaited this edition of the "gospel according to Scott" with great anticipation but sadly, this a huge let down. It is back to square one and a beginner's / marketer's / business analyst's view of the world of business process automation but has nothing concrete in it to help make this product perform in the manner it is billed to do. For example, it takes an act of congress to add a custom adapter. It is pure luck if you add the web service adapter and don't fill up your event viewer in 30 seconds or less with WSSLib errors. Biztalk still has problems with large files even though having this product fully .NET framework compliant was supposed to be the holy grail. It dies on a 3MB HIPPA 820 file everytime. Embarassing! Ok, rant complete!

Scott, when will we get a book that is for the folks out here in production fighting the myriad of problems this product has and addresses how we can either fix them or work around them as the emerge from it daily?

Signed,

Sad Sam in Florida

Rating:
Summary: What a mess
Review: The Biztalk team should be ashamed, Biztalk 2004 is the worst documented Microsoft product I have seen in years. The product itself is stable and powerful - but the documentation is incredibly weak. This book provides a somewhat better approach to learning the basics than the product documentation - but thats really not much of a complement. Unfortunately, only the chapters that cover the basics are worth reading. The worst problem with this book is the lack of samples and code in critical areas. For instance, the chapter on developing adapters provides no full example, no sample code in the download, and no strategy for actually implementing an adapter. The chapter on orchestration correlation - no full example or sample code. There's a couple of chapters on "patterns" which offer little more than short scenario walkthroughs - thats not what patterns are. But the biggest insult of all is that the chapter on using the native (pre built) adapters was made an "Appendix" that was available online.. But its not there? Who are they trying to fool? Using the native adapters is critical information and the lack of this chapter should have prevented this book from being published. I realize biztalk is a big, new product - but the poor quality of the documentation, including this book, is going to hurt the products adoption.

Rating:
Summary: positioned to support Web Services
Review: The book describes BizTalk Server 2004 as a crucial enabler for n disparate applications on a network to talk to each other. It acts as a hub in a hub and spoke configuration, where it and the applications implement various standard communication protocols. So that instead of a maximal worse case of the n applications all having to talk directly to each other, which scales as n**2, we have a linear scaling.

So far so good. But you have probably also heard much about Web Services. These are inherently a distributed programmatic interaction between applications. It is expected and hoped that Web Services will one day be huge. Well, Microsoft has positioned BizTalk to take advantage of this, if it happens. The book explains how BizTalk supports the alphabet soup of XML, XML Schema, XSLT, SOAP, WSDL. Plus the various WS-Security and WS-Policy standards. Moving up the logic chain, there is an emerging Business Process Execution Language (BPEL) that describes business computations. Microsoft has said that BizTalk will also support this. Although you should note that BPEL is still being formalised. After which, if BizTalk will conform to that, you will probably be looking at the 2005 or 2006 version of BizTalk.

Rating:
Summary: Fill with errors and lies
Review: Three chapters of the book were made available on Theserverside.net and I read the chapter on Biztalk's rule engine. After reading the chapter, I was suspicious of the description of Biztalk's rule engine. I posted a question on TheServerside.net asking for clarification. The response from the author Scott Woodgate is very revealing. His response stated the Biztalk rule engine implements RETE algorithm by Charles Forgy. When I asked further questions, Mr Woodgate's response revealed Microsoft's claim of an inference engine is a flat out lie. His response was that for small rulesets, they use RETE, but for large rulesets, it uses a proprietary algorithm.

RETE has been proven over the last two decadees to perform better than all other algorithms. It's defining charateristic is the number of rules in a ruleset does not impact performance. The fact that Biztalk's rule engine does not scale as the ruleset increases proves it is not RETE and possibly not even an inference engine. Glaring lies like these in the rule engine chapter make the book flawed at best. I would recommend those who read this book completely ignore the chapter on rules as it is a blantant lie.

Rating:
Summary: A lot more to be done
Review: We also awaited this book with baited breath. We experienced a nightmare in development when Biztalk 2004 was released with no documentation. Microsoft needs to pull their game even the Microsoft employees I spoke to agreed this was not how a product should be released. Biztalk has some very fierce competition in the Enterprise Application Integration space and I've heard of a few companies in NZ (where I'm based) dropping Biztalk 2004 mid development to go to SeeBeyond. Hopefully a third party will produce a better book. I suspect however that Biztalk 2004 will be a short lived product as Indigo is not far away (2 yrs) from release. As Biztalk shares much of Indigo's technology there will need to be a another release soon, bad strategic planning by Microsoft. It is however a very powerful product and I believe a good starting point but please no more painful releases.