Via Ongoing, I read some interesting discussions of programming languages — mainly Python vs. Ruby, with most people happily dumping on Java.
Steve Yegge, in particular, argues that language success is based mainly on marketing, and that Python is doomed to obscurity because of the community’s lack of marketing savvy.
The programming language cycle
While I agree that Python probably is doomed to perpetual obscurity at this point, I think that Yegge’s focus on marketing is oversimplistic; instead, I’d argue that there’s a self-perpetuating cycle at work for successful programming languages:
- Elite (guru) developers notice too many riff-raff using their current programming language, and start looking for something that will distinguish them better from their mediocre colleagues.
- Elite developers take their shopping list of current annoyances and look for a new, little-known language that apparently has fewer of them.
- Elite developers start to drive the development of the new language, contributing code, writing libraries, etc., then evangelize the new language.
- Sub-elite (senior) developers follow the elite developers to the new language, creating a market for books, training, etc., and also accelerating the development and testing of the language.
- Sub-elite developers, who have huge influence (elite developers tend to work in isolation on research projects rather than on production development teams), begin pushing for the new language in the workplace.
- The huge mass of regular developers realize that they have to start buying books and taking courses to learn a new language.
- Elite developers notice too many riff-raff using their current programming language, and start looking for something that will distinguish them better from their mediocre colleagues.
You’ll notice that there’s no step here called “marketing”; instead, there are several distinct stages of evangelization and community building. Major vendors (other than the language’s owner, if it’s a vendor) will start to notice the language once the second wave (sub-elite) developers arrive, and IT managers will notice it because of books, magazine articles, and pressure from the high-end developers. Some — possibly a lot — of marketing will come out of those steps, but it is as much a result of the language’s success as a cause.
Points of failure
In this cycle, there are a few highly probably points of failure:
- Timing: A new language might not be at the right stage of development (too raw, or too stale) at the time when elite developers decide to make a mass migration.
- Features: If the new language’s features don’t answer the elite developers’ annoyance list, not enough of them will migrate to it.
- Openness: Elite developers are used to having a lot of influence, and if the new language’s development process does not allow them sufficient say in the new language’s evolution, they will leave before they attract enough sub-elite developers.
- Tools: Sub-elite developers might find the language unsuitable for day-to-day production use, especially if enough basic tools are not available (libraries, testing, debugging, GUI tools, performance measurement, etc.).
- General acceptance: Regular developers might object to the new language and sabotage projects using it, either by producing poor-quality code or by missing deadlines (and blaming the new language in both cases).
Most programming languages stumble over one or more of these — it’s as much luck as clever design when a language like C++ or Java makes it past the hurdles and into the workplace. Success tends to draw more success, money draws more money, etc.
The final and most important point here is that a programming language’s perceived coolness will always suffer from its success. Java cannot possibly still be cool when there are thousands of regular developers slaving away in the bowels of ACME Widgets using it to write enterprise applications. If, in fact, Ruby displaces Java in the enterprise (which may not happen, since Ruby has no advantage over Java to match Java’s memory-management advantage over C++), it will suffer precisely the same fate, and we can expect Bruce Tate to write a book Beyond Ruby in five years or so.
By that measure, Python’s very failure is a kind of success — as long as it never really becomes takes hold in the workplace it will always carry a small degree of distinction with it, and at least a few elite developers won’t feel pressured to move on. Like a movie or band that never becomes too popular, Python will hang onto its snob appeal.