-2000 Lines of code (2004) (folklore.org)

Article summary

In 1982, the Lisa software team at Apple tracked engineer progress by the number of lines of code written per week. This metric led to inefficient coding practices. The article highlights the problems with measuring productivity by lines of code. A notable example from Apple's history illustrates the issue.

Main themes

• Code Quality
• Software Metrics
• AI-Generated Code
• Market Stratification
• Productivity Measurement
• Efficient Coding Practices

What commenters say

• Measuring software development progress by lines of code written can lead to inefficient coding practices and poor quality code.
• The use of AI-generated code may exacerbate the problem of excessive and low-quality code.
• Some argue that the market will stratify, with high-quality, expertly crafted code commanding a premium price, while low-quality code will be cheap and abundant.
• Others believe that end-users do not care about code quality as long as the software works, but some argue that they do care, even if they cannot pinpoint why their computer is slow or buggy.
• The contrast between high-quality and low-quality code will become more extreme with the use of AI-generated code.
• Some commenters have personal experiences with refactoring large amounts of poorly written code into much smaller, more efficient versions.
• There is disagreement over whether AI-generated code can produce extremely poor results, with some arguing that it is rare and others claiming it is common.