 |
John Vlissides Research staff member "And make it sound natural, like you're explaining something to a friend. Avoid pretentious, stuffy prose. The point is to be understood. " |
On the book: Pattern Hatching: Design Patterns Applied (1998)
In Memoriam
IBM researcher John Vlissides passed away on November 24, 2005
What made you decide to write a book on this topic?
My coauthors on Design Patterns and I had a lot of object-oriented (OO) design experience among us, and we felt a need to write it down. Erich Gamma had already taken a stab at writing design patterns in his doctoral thesis. It was a good start, but we quickly evolved beyond it. At first it seemed like much of what we were describing was well known and obvious. As the feedback arrived, however, we learned that it was anything but. There were many encouraging comments, too, which propelled us as we drove toward the finished manuscript. Its reception has been gratifying to say the least.
Pattern Hatching was meant as a commentary on Design Patterns. I wanted to show people how to get the most out of patterns. But Pattern Hatching almost didn't get published. I had accumulated articles from my column in C++ Report and wanted to publish them in a more archival form with minimal effort. I got the publisher of the magazine to agree verbally to publish them. I had just finished a draft when I learned that said publisher was getting out of the book business. It was only when I mentioned the story in passing to some folks at Addison-Wesley, with whom we had an ongoing relationship, that they encouraged me to resume the project under their aegis.
How did you go about the research necessary to write such a technical book?
We already had a broad collective knowledge of what had been done in OO software design—what worked and what didn't. We also embarked on "archaeological digs"—scouring the literature and the software systems themselves for commonalities from which to distill design patterns.
 |
Finding the time to write. I'm a born editor; I love restructuring text and wordsmithing. Creating prose is much harder for me. It takes a lot of time. I also require a computer to craft my thoughts. Seldom do I write a sentence that doesn't get rehashed and moved around countless times.
What or who inspires and encourages you to write?
I've drawn inspiration from many authors, both technical and non-technical. Don Knuth has been a big influence. I love his soothing style. Other influencers include Scott Meyers, Steve Reiss, Steve McConnell and Fred Brooks. In the non-technical realm, I've been drawn to people who have written books on good writing. The strongest influences here have been Joseph Williams, John Trimble and William Zinsser.
What advice do you have for aspiring authors?
Write about what you love, and keep at it. Don't fool yourself into thinking you can write well in one pass. Most of the work of writing is rewriting, at least for me. Time spent reworking and polishing pays big dividends. And make it sound natural, like you're explaining something to a friend. Avoid pretentious, stuffy prose. The point is to be understood.
Who are some of your favorite authors today?
The folks I mentioned earlier are among my favorites. I also enjoy novelist Jeffrey Archer, political commentators William Kristol and David Gelernter (who's also well-known as a computer scientist), and automotive journalists Jamie Kitman, Ezra Dyer and Phil Llewellin.
What role did books play in your childhood?
I preferred magazines to books, much to my later regret. I would devour pseudotechnical rags like Popular Science and Mechanix Illustrated, as well as car magazines (especially Motor Trend) and newsmagazines (my parents subscribed to Time for many years). The books I read were mostly those assigned at school. It wasn't until later, in college, that I developed an appreciation for real literature.
I always liked writing, though. I still have "reports" I wrote in second grade that I did for fun, unsolicited by my teacher, with titles like "The Sea" and "Clouds of the World." Most of the material was unresearched, even made up. I wrote them mainly to delight in the finished product. I guess I still do!
Other books by John Vlissides
Pattern Languages of Program Design 2 (1996)
Design Patterns (1995)
Object-Oriented Application Frameworks (1995)