[khiqxj]

> single letter variables

i hate when people put descriptive names on variables that dont need it. when im tired i end up reading the name over and over as if it has meaning (it doesnt), and subconsciously ponder the meaning for up to a minute before snapping out of it.

they actually dont even serve a purpose. its because we are reading code in text form that they are needed. conceptually, most variables are just something flowing out of one function to the next which would be easily repesented by lines in a diagram or many other ways that dont use names. you're meant to only give names to the rare case where it would truly make the code more understandable. this shit doesnt: thingManagerProxyGenerator = createThingManagerProxyGenerator()

[preommr]

I usually don't like commenting on people's choices for variable names since they're usually scoped and less important than the interface. However, I'll give my two cents because I remember stuff like this being confusing when I was younger.

There's definitely a balance to be struck - one letter variable names are horrible outside of single lines (e.g. a one-line anonymous function), extremely long names are better, but still bad. I say better because it's higher in comprehensibility - which is more important - if not very readable. While it's been well documented that shorter information is easier to remember, it's important to remember that it's usually limited to brevity that still provides context. Very short names e.g. (pg for proxy generator) are bad because minds require an additional step to "unravel" the meaning.

IMO it's usually best to limit to variable names to one or two words. Something like 'proxyGenerator', for the given example. Or even 'generator' if the parent block is extremely small (e.g. 5 lines). Most readers perform word chunking such that something like 'proxyGenerator' reads as two units without the indirection that an abbreviation or unrelated symbol would require.

[nvrmnd]

This is also a very interesting concept/question when it comes to AI writing software for practical applications.

Would a large language model benefit from descriptive variable names when it comes to code understanding, or would this just be for human benefit? My first thought was that a computer (e.g. compiler) does not care what the token is named. But a large language model may actually benefit from it, certainly chatGPT is sensitive to it.

But then, doesn't it mean that it could be fooled by misleading variable names (as a human would), something we would criticize any system for. "Alpha*** is easily fooled by changing variable names, making it completely unusable for blah blah".

[krzat]

This could be a cool task for AI: hover on variable and have it's role at a higher level than just finding all references.