Post-CGDC Ramblings

I still don't have a clever name for the column, so you'll just have to put up with something like "The Book Review Column" for a while longer.

So, my first vacation from Rocket Science! I figured I needed to get out of town briefly before the Christmas rush, so I took a week off and we went to New Mexico for 9 days. We spent the time wandering around six or seven pueblos and mesas, visiting shops and museums, and eating way more than we should have. One of the pueblos we visited was Acoma pueblo, one of the two oldest continually inhabited cities in the United States, if not in the New World. Acoma is built on a mesa a couple of hundred feet above the valley floor, and is such a superb natural fortress that Spanish soldiers with firearms took days to take it against Indians armed only with rocks.

So what does a 900 year old fort and a bunch of Indian villages have to do with computer books that will help you make games for this Christmas? Or even next Christmas?

Well, Acoma reminded me that environment can be everything. The narrow trail (in some places more like a ladder) that leads up to Acoma is an attacking soldier's nightmare. One or two people at the top with a big pile of rocks (and rocks are about all that Acoma has in abundance) and Rambo isn't going to make it up even with an AK-74. It's important to choose your environment carefully. I've been working on this at Rocket Science. Trying to get my engineers (and the rest of the creative people) the best possible environment to work in. Now I haven't been able to get the company to move to Maui (yet), or even change buildings or build in offices. But I'm working on a couple of things (notably quiet time) that come from the first three books I want to talk about.

The first one is easy, because I've already reviewed it. But it deserves another mention, just to keep it in your mind. It's Peopleware, Productive Projects and Teams, by Tom DeMarco and Timothy Lister, published by Dorset House Publishing; 355 W. 12th Street; New York, NY 10014; $25. If you are a manager, or you have a manager, or you have to do real work, you want this book. You want your manager to read it, too.

It was a great thrill to hear Larry Constantine speak at the 1996 CGDC (and not just because I spent most of the conference sick in bed, either). I've been reading On Peopleware (Constantine's column in Computer Language, now Software Development) for years. I even did the groupie thing and asked him to autograph my copy of Constantine on Peopleware, by Larry Constantine, from Yourdon Press; Prentice Hall Building; Englewood Cliffs, NJ 07632; $20.

Constantine on Peopleware is a collection of Constantine's columns (not a big surprise, that). As the title suggests, he writes about the human element in software. But he writes about much more than that. Constantine is something of a renaissance man — his degree is in management, his thesis was on the psychology of computer programming, he spent 10 years as a family therapist, and he is now a consultant in software engineering. He also wrote one of the seminal works on structured programming techniques. He's neither your average techno-dweeb nor the stereotyped "touchy-feely" MFCC. He writes and speaks very much to the point that, because people build software, you must understand people in order to understand the process of building (and designing) software. And like most things, you can build software better if you understand what you're doing.

Among the thirty or so columns included are "Irksome Interruptions", which explores the communication skills of programmers as well as the productivity issues involved in having an office-mate (or a cubical environment). In three pages Constantine presents a reasonably complete protocol for office communication designed to reduce interruptions (and very funny besides). Also in here are "In-Time Delivery", which will strike a chord in any games team that's survived a Christmas death-march; "Decisions, Decisions", "Consensus and Compromise", and "Negotiating Consensus", all of which consider the group process of coming to a decision; and an entire section on "Process Improvements", all of which are worth reading.

The last book I held up repeatedly at my Roundtable on Managing Programmers and Game Development at CGDC '96 was a book whose title just made me want it. Managing Software Maniacs by Ken Whitaker; Coriolis Group, John Wiley & Sons, Inc.; 605 Third Avenue; New York, NY 10158-0012; $30, turns out to be just about what it sounds like — a romp through real life with a bunch of crazy people turning out cool software. The only problem is that it's not a game company, so their idea of cool and crazy isn't our idea of cool and crazy. For example, Whitaker makes the point that beer can be useful in team building activities (and it seems that is unusual in his world). Duh. EA figured that out years ago.

But there's a lot of good stuff in here, even if it is culled from the "Real World" where you have to interact with Marketing and talk to Customers and all that. The anecdotes, like Dilbert cartoons, strike home. And the Real World ideas represent a different way of looking at things. In particular, there is a delightfully strange (that is, unusual) emphasis on having the documentation people be part of the development team. This is something we used to do a long time ago which seems to have (in my experience) changed along the way -- people who write manuals now tend to be 'hired guns' brought in during the final stages of Alpha and Beta to knock out a 32 page manual for a few bucks and be on their way.

The emphasis on QA being part of the process from the beginning is also unusual and I think we can learn something from it. The idea that QA people know lots about the market and have information on how well your previous products (or releases) have done (by evaluating Tech Support calls) isn't new, it's just that I haven't seen a lot of attention paid to it in the games business.

Besides the overarching themes, each chapter has a set of "Rules for the Unruly", which summarize the major points that managers should take away from each anecdote. Not all of them hit the mark in our business, but enough do so that Managing Software Maniacs is well worth the time to read it.

One more quickie. Since we're engaged in developing product for Windows 95, I've had a need to become familiar with programming Windows, something I thought I'd left behind in my previous incarnation as a programmer of desktop publishing software. So naturally I started looking for books that would help. This is one of the biggest fields of books available today, and I relied upon recommendations to get a start. I hope at some point to do an entire column on the Windows programmer's bookshelf, but for now this will have to do:

A good introductory book on programming games for Windows is Black Art of Windows Game Programming, by Eric R. Lyons, published by Waite Group Press; 200 Tamal Plaza; Corte Madera, CA 94925; $35. It isn't Windows 95 specific, it doesn't talk about DirectX, and the game example with it is programmed in Visual Basic (although the code samples in the text are C or C++), but it's still quite useful. It gives a good overview of the nature of Windows, talks about WinG (the precursor to DirectDraw), provides useful information about the oddities of Windows color palettes, and generally gives you a nice start on working in Windows.

As usual, email me if you've got something you'd like reviewed, if you want to take issue with something I've reviewed, or if you just want to talk about making games (or even other software).

Evan Robinson is the Director of Games Engineering at Rocket Science Games, Inc. No, it isn't your father's Rocket Science, and yes, I've heard all the "it's not Rocket Science" jokes in the world (anyone who sends me a new one gets a prize -- there is a time limit on this offer). Last year, he was a consultant for companies like Illusion Machines, Inc. (hi Ellen & Steve!), 3DO (hi R.J. & Chris!), and The ImagiNation Network (hi Sue & Gano!). The previous year, he served as a Technical Director at Electronic Arts (hi Tim & Dave!), widely regarded as the reason he went crazy and decided to enter management. In his first 10 years in the business he did products, code, and support for a variety of publishers and developers. He lives with his beautiful partner and her wonderful two children in a small valley without broadcast TV, cable, or a satellite dish.

Copyright © Evan Robinson. All Rights Reserved.