Kernel prepatch 3.11-rc4 [LWN.net]

Kernel prepatch 3.11-rc4

Posted Aug 6, 2013 9:59 UTC (Tue) by khim (subscriber, #9252)
In reply to: Kernel prepatch 3.11-rc4 by mpr22
Parent article: Kernel prepatch 3.11-rc4

It's obviously wrong thing to do because code readability is more important then code writability: it's written once by one guy but then read by thousands and sometimes refactored (to refactor something you need to first read old code and then write new code thus balance is unchanged).

Tradeoffs may be different in code which you are writting for yourself to only run it once... but you would not believe how often such code survives for years, so I'm not sure there are places where I would like to see Yoda conditions in practice.

to post comments

Kernel prepatch 3.11-rc4

Posted Aug 7, 2013 9:21 UTC (Wed) by dgm (subscriber, #49227) [Link] (3 responses)

What's the value of code that's easy to read, but incorrect?

Kernel prepatch 3.11-rc4

Posted Aug 7, 2013 9:32 UTC (Wed) by mpr22 (subscriber, #60784) [Link] (2 responses)

Low, possibly negative. But once you fix it, it is code that is both correct and easy to read, making it superior to the code that is correct and hard to read.

Kernel prepatch 3.11-rc4

Posted Aug 14, 2013 13:22 UTC (Wed) by dgm (subscriber, #49227) [Link] (1 responses)

For this you first have to spot the error. Using one style you get help from the compiler, while with the other you need angry users and some debugger voodoo.

Kernel prepatch 3.11-rc4

Posted Aug 14, 2013 15:25 UTC (Wed) by mpr22 (subscriber, #60784) [Link]

The compiler can be (fairly easily) modified to alert the user to the errors that the order of comparison operands people like me find natural makes possible. People-like-me's brains cannot be easily modified to find the order of comparison operands people-like-you find sufficiently cautious natural.