Barr begins by describing how he came to work for Microsoft. This is the start of four chapters on Microsoft's recruitment system, covering both the initial selection on campus, the interview system, and the overall effectiveness. There is also an introduction to how work is structured at Microsoft, in particular the division between developers, program managers, and testers. Three chapters then describe Barr's time at SoftImage, a Microsoft acquisition producing digital editing software. Here we are introduced to the different types of "demos" (from carefully scripted sessions presented by special "demo artists" to genuine "hands-on" demos) and the complexities of dealing with third-party hardware suppliers.
Three chapters then present a potted history of computing over the last twenty years or so, beginning with an account Barr wrote as a teenager back in 1982, after visiting ComDex. Barr focuses on evangelism — on the factors that contribute to one platform or operating system winning out over others — and in particular why IBM PC hardware became ubiquitous, why MSDOS beat CP/M-86, and why Windows beat OS/2. None of this is particularly novel, but it's a nice lively account.
This leads naturally to more recent conflicts and debates which pit (as flagship icons) Microsoft against Linux. Again, there is nothing spectacular here, but Barr offers an intelligent, informed, and balanced perspective, coming up with some points that were new to me. Of the claim that it will be difficult to find programmers to do the "unsexy" work with Linux, for example, he writes
"Microsoft, being a company with salaries and a supervisory hierarchy, has the ability to order someone to work on something he or she doesn't want to work on, but I never recall this happening. People worked on things that interested them and projects still got complete coverage. There is no reason that the same should not be true of Linux, especially given the size of the Linux community."Two chapters evaluate attacks against Microsoft, the first addressing popular criticisms and the second the various legal attacks. Here Barr is level-headed, calmly rebutting some of the sillier attacks while accepting valid criticisms.
A major weakness of this material is that Barr only ever talks about "open source" (a development methodology) and never about "free software" (a much broader movement). One major reason for techs ranting at Microsoft is their unhappiness with loss of choice, freedom, and control — and this has been articulated as an ethical and political position by the Free Software Foundation and others. But Barr never considers this argument against Microsoft at all.
A chapter on online community, covering the bulletin board scene in Montreal in the 80s, is really a digression. The final two chapters consider the future of Microsoft. Barr argues that Microsoft should stick to its core PC business and not get distracted by ventures such as the XBox. He ends where he started, arguing that the key to Microsoft's future lies in its handling of employees, in its ability to attract, recruit, and retain good people.
Proudly Serving is nicely laid out and has obviously been carefully edited. Barr avoids most technical details (an exception is some discussion of buses and video hardware in the chapters on SoftImage) and offers separate digressions on Code, APIs, and Middleware. A minor complaint is that the workings of Microsoft stock options are only explained in the last chapter, by which point the reader will either have got the general idea already or decided they don't care.