Cons of High-Level Languages
Cons of High-Level Languages
How high-level can a language get, and what are the potential problems associated with them? In Star Trek, science-fiction computers communicate with their commanders in an almost human language. Our science fiction tells us that the higher-level a language is, the easier it is to communicate, the better. In real life this isn't the case.
The biggest problem with high-level languages is that they are slower than their low-level counterparts. There is a give-and-take relationship between the speed of development and the efficiency of a program. C is speed efficient because the programmer handles all of the low-level resource management by hand.
Since they aren't as speedy and they handle low-level resource management themselves, high-level languages are not great for engineering systemlevel programs like device drivers or kernels, or other situations in which you need tight control over low-level tasks, like memory allocation. Lack of speed also makes them poorly suited to computationally intensive applications, like those that build data structures and algorithms from scratch. In particular, a low-level language may be more suited to your application if:
-
It needs to implement complex algorithms or data structures.
-
It needs to manipulate large data sets.
-
Execution speed is critical.
-
The functions are well defined and will not change.
The pros and cons of high-level languages are highlighted in Table 1.2.
|
Pro |
Con |
|---|---|
|
Saves human time |
Less efficient during computer runtime |
|
Portable to many platforms |
Specific platforms aren't as efficiently utilized |
|
Modularity and reusability |
Can lead to dizzyingly high number of libraries |
|
Easier to read, write, and maintain |
Loss of some control over code organization |
|
Auto-management of many bug-prone features |
Less low-level control of resources |
|
Easy to learn |
Too many programmers could lower one's salary! |
NOTE
High-level languages are criticized more often for their lack of speed than anything else. But keep in mind that they usually can be compiled or semi-compiled. This can make them much faster than languages like Perl, AWK, or other rivals. Also, today's machines are 500-2000 times faster than their predecessors from the 1980s.
  |