On Wednesday, August 27, 2025 8:40 PM, Jon Perryman wrote:

> The transition to mixed case was painful for everyone except Gil, who
> seems to have loved it.
Speaking only for myself, I loved it as well.  I didn't participate in any 
self-study, but I *feel* it makes the comments easier to understand, and both 
code and comments easier to read.  For new files, I write both code and 
comments in mixed case; uppercase is obviously still required for operands such 
as text strings that are destined for a console, etc.  When I update an 
existing file, I do not change existing comments to mixed case, and I decide on 
uppercase vs. mixed case based on scope:  if I add a self-contained subroutine 
then it will likely have mixed-case comments, but if it's just a few lines 
within an existing routine then I use whatever style is currently in use.

> Strange that no one mentions how they solved half a line in uppercase
> and switching to lowercase for the other half. Do people hold the shift
> key for half the line?
Editor macros.  I type everything in lowercase, operands in one string with no 
blanks, followed by comments, and my formatting macro splits operands into 
multiple lines if necessary, formats comments with appropriate capitalization 
(based on a reference file of acronyms), and splits them across multiple lines 
as necessary.  I didn't bother going so far as to parse quoted strings that 
include blanks, so I generally need to hand-format those lines (which are few). 
 The need to "massage" any resulting output lines is rare.

I started writing code like this between 20 and 30 years ago.  Nobody has ever 
complained, either in-house or customers, although I have had a couple of 
people remark "I didn't know you could use mixed-case like that."

- mb

Reply via email to