Computer languages comparison. What else?
I would rather be quiet about this never ending topic, but after brians interesting post
about comparing percentages I have to contribute my thoughts.
I read some other posts about computer language rankings. Some supposedly based on web traffic
and others on books sales.
All this posts do mention some kind of measurements based on numbers. But I think that there are
other view points or considerations to take into account before asking for a valid computer
In order to ranking things these must be comparable with each other.
Beside technical or linguistic considerations, let's assume as a comparison criteria
the quantitative or qualitative use of each computer language. Let the confidence of the
measurement (if possible at all) of the use as a theme for another post.
Take Java and Lisp. Java is not just a language. It's a technology. A huge one: Security,
Databases, Object-Orientation, Distributed Computing, etc...
Lisp yet as powerfull as Java as a language, is not a technology in the same sense as Java is
and also isn't aimed to business enterprises.
So, any sense comparing Java and Lisp?
It's popular folklore that each languange has its best fit on a specific environment or
application area. There are Web programmers and System programmers. There are Health systems
programmers and there are Financial systems programmers.
Based on the current demands of several kinds of applications, e.g. game, space, health,
communication, financial, and so on, makes it any sense to rank computer languages based on
the actual demands of the specific application areas which they are appropiate for?
Would a financial system programmer switch to Objective-C just because certain mobile apps are booming?
So, I think there is little interest in comparing computer languages based on their utilization.
How much web traffic?
More interesting will be talking about another comparison criteria: the generated web traffic
related to the using of a particular computer language. Again, letting beside the objective or
confident measurement of generated web traffic.
I will consider three points that could influence the generation of web traffic:
- Programmer habits
- Expressiveness and flexibility
Generally, I would say that computer systems which are self-documented (docs included in the distribution/installation) in the right quantity and the right quality will generate less web traffic. Also computer languages with a well organized community will generate less web traffic because no need for long searches. Every user of the language knows where to find useful answers.
There are computer languages where the need of copy and paste components or code snippets are typical. So the user has to search for the right component or code in order to solve his problem at hand. Also, components may need to match specific version in order to combine. The user has to search for some publications confirming the right version combination. So, using a computer language impose certain habits to the user of the language. This habits in turn generate more or less web traffic.
Computer languages differ in their expressiveness or flexibility in their use. The more constraints a language impose the higher the need of looking for a specific way of solving a problem. So, more web traffic generation searching for this specific solution.
Another point which I don't mention is the marketing behind each computer language aimed to spread the use of each language. Some have a stronger marketing strategy than others.
Finally, there is another point which require further consideration. Indeed, it's an important one.
There are suggestive studies about how our use of digital technologies is shaping the way we think and act. In short, they suggest that we are loosing the ability of deep reading and deep thinking. The mastering of a certain computer language like Perl, require this abilities.
But that's belongs to another post...