Hacker News Reader

Comment by vog | Hacker News Reader

Once again, a great Linus Torvalds statement! I especially like the last paragraph, which has so much truth in it, and can be applied to small as well as large software projects:

Don't just make random changes. There really are only two acceptable models of development: "think and analyze" or "years and years of testing on thousands of machines". Those two really do work.

View on HN

To be fair, the second case really is just a matter of fossilized random changes.

Linus knows that and says so as well. But that's okay, because that works. Starting down that very same path again without a huge reason to do so would be very painful for any mature product.

That is correct, but is based on a philosophy that the longer something has been in use, the less chance it has to demonstrate behavior that has not been previously observed — he even explains that in more detail in further posts where he discusses his desire to not cause regressions.

Yep, that's how the evolution works. Understanding stuff is how the progress works.

Doesn't that mean that while it's actually happening, it's not, in fact, an acceptable development process?

I've worked with programmers who also would have argued with Linus, but there's no way I would have been able to be as direct as Linus has.

This definitely struck a chord, though, and I have my own saying for silent retorts:

"It's a computer program, not a #$%^ing Advent calendar."

The point being, I want to see analysis and process, not just the right numbers popping up in the right boxes.

Or, in slightly different words, "understand it" or "test the hell out of it".

Reply on news.ycombinator.com