Showstopper: The breakneck race to create Windows NT and the next generation by G Pascal Zachary is an interesting exposition of the team that created Windows NT. NT is the basis of Microsoft’s post DOS operating systems including Windows NT, Windows 2000, Windows XP, Vista and Windows 7. It is the most used desktop operating system in the world. The book attempts to be something like the classic The Soul of a New Machine . The book is not as good a read as The Soul of a New Machine but it is about a subject that was far, far more important than the creation of the computer in the older book.

The lead of the NT program, Dave Cutler is profiled in depth. Cutler is portrayed as a strong, driven engineer who is more than a little rough around the edges. He is not idiolised in the book which is great. His limitations and defects are well described. Cutler was at Digital before moving to Microsoft. At Digital he created another operating system VMS. Cutler’s dislike of Unix is interesting to learn about.

Much of the book is spent looking at the other people involved in the creation of NT. Paul Maritz, who was an execute in charge of NT is described as careful, thorough and critically able to handle Cutler. Michael Abrash the graphics guru appears. Bill Gates is also given an interesting portrayal. A raft of other characters also appear and it’s quite a game to keep track of them. Annoyingly the book does not have an index so that the first references, which are usually descriptive and ground the characters, can easily be found.

The book does give a very interesting view of the interplay between the OS/2 groups and the NT group. Clearly Microsoft’s strategy was fairly ad hoc but wound up being very successful. Microsoft’s ability to hedge their bets, including developing the Windows 95 system after Windows NT enabled them to have different operating systems that could obtain good combinations of speed and performance as hardware changed.

The combativeness of people in the various teams of the 250 or so people who wrote NT is described in some detail. Clearly Zachary interviewed people that disliked some of the people they worked with. The graphics group’s contests with other sections of the team are well described.

The growth of NT and the greater and greater hardware requirements are interesting to see. It’s worth noting that Windows NT would not really take off until Windows 2000 appeared, some 7 years after NT first appeared. The legacy, non-preemptively multi-tasking, shared memory managed dos kernel would labour on for some time until the more stable NT kernel took over.

The role of testers and the build cycle in the development is fascinating to see. In few other books is this critical part of modern software engineering actually considered lest given substantial time. Cutler’s skill in managing and working with the group is praised.

It is very interesting to see how the groups involved in creating new software have expanded. Zachary points out in the epilogue what a big project NT was. Compared the individual effort of creating the Apple II, the smaller team effort involved in creating the Macintosh or the handful of people who created Unix NT is a different breed. Zachary makes an interesting statement in the epilogue about how large organisations are the only ones who can create things like a complex, backward compatible operating system.

The book has its limitations, the lack of an index is infuriating, the technical descriptions are often fairly poor and as a book there are too many characters portrayed. However, it is a thoroughly researched portrait of the people and process that created a hugely successful system.

3.5 / 5


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s