Software Release Methodology [Paperback]

In each case the author explains both the "what" and the "why" behind the strategies, giving practical advice for their deployment in the real world, and practical experience from having already done so.

The only things that bothered me about the book were that (1) the author claims to be introducing a new field "software release methodology"; (2) the back of the book claims most of strategies in the book have never been published before; and (3) there is no bibliography or references section of any kind.

As someone who has been a software configuration management professional for the past twelve years, it is my understanding that release classification, numbering, distribution, and management have all long been considered part of the release engineering process. The other areas of source-control, builds, integration, and defect-tracking & change-control have long been considered part of software configuration management (SCM).

I dont understand why both CM and SCM fail to appear anywhere in the book (especially the index), since the first 7 chapters are precisely the stuff of SCM. I do know that release engineering is often considered part of SCM as well, but I agree with the author that it has been underemphasized for too long. It's good the author remedies this, but failing to even mention SCM, and claiming he's introducing a brand new field stretches reality a tad too far.

Similarly, claiming all or even most of these techniques have never been published before is also too grandiose a statement. I have seen most of them published before, in various books and journals and conference proceedings. The author has done a *great* service by collecting them together in one volume and making them more accessible to the practitioner in the trenches, but he's not the first to publish them (and he doesnt have to be in order for the book to be of value).

I can understand the author making a case that he's creating a "new field" by combining SCM and Release Engineering in a certain way (though I still think its a stretch); I can even understand claiming to be the first to publish these strategies in a single accessible volume (but not the first to publish them ever); What I cant understand is why there is not one single citation or bibliographic reference anywhere in the entire book.

To me, the lack of any bibliographic references seems inexcuseable, and combined with the aforementioned mistatements wreaks a bit too much of hubris/arrogance for my taste. By failing to provide published (as well as online) references, the author gives the reader no way to verify the book's claims of original work, nor to seek out further background material, nor demonstrate that the author performed the requisite due dilligence to comb the literature before making such claims.

It was unnecessary to claim originality for this work to be of value. A modicum of modesty should have been called for here. These three failings which amount to only a few pages of text leave a very sour taste on what is otherwise a very fine, substantive and insightful contribution in the other 250 or so pages of the book.

If not for those ever so brief yet overinflated claims, I would have given the book five stars instead of four. It really is a *must* *read* for anyone who is less than expert in the fields of software configuration management and release engineering. Just keep in mind that the material is not quite as inventive or innovative as those few pages would lead you to believe the rest of the book to be. Fortunately, this in no way detracts from the usefulness of the proven practices described therein.