Empirical proof that underscores are good

[Lost User]

Super Lloyd wrote:

Maybe it's what you argue about when your work doesn't matter, I suppose?!

Welcome to the discussion. :-D

[Kirk Wood]

Empirical study to prove that the person doing the study had no clue. Early in we find that they included programmers and non programmers in the study. Which raises the question, "Why would someone include an audience the code wasn't meant for?" I will give the person the benefit of the doubt that he didn't know any better. At the end of the day, it is for people. And there is benefit to having the code consistent. And there is benefit to having that consistent state follow that which is least annoying to the audience. (And here is a clue: said audience writes code.

[Gary Huck]

Me? I used underScores. And I use pascal_case. And I use CamelCase. I also use Hungarian notation sometimes. I've found over the years and thousands of lines of code in many languages that there's NO ONE GOOD solution ... to just about any question. I do not, however, use tabs :)

[Gary Huck]

nice :)

[GenJerDan]

Underscores are a pain in the butt if you open the code in anything other than a code editor. The spellchecker underlines the "misspelled" word, and the underscores disappear. :p

YouTube and My Mu[sic], Films and Windows Programs, etc.

[Luiz Monad]

This study just consider normal people brain, they used "seven undergraduates" from second year, people that aren't programming for long time, if they even program instead of just frequent classes. I am coding for the last ten years, my neural network has adapted to read CamelOrPascal just as fast as this_mode. But nothing is faster than just space, there is only one way to read faster, is to read less.

[TNCaver]

Loctrice is right: ThisIsTitleCase. thisIsCamelCase.

If you think 'goto' is evil, try writing an Assembly program without JMP.

[Nish Nishant]

TNCaver wrote:

ThisIsTitleCase. thisIsCamelCase.

Camel case has many variants. Pascal case is upper camel case. Title case is CamelCaps. See http://en.wikipedia.org/wiki/CamelCase[^]

Regards, Nish

---

Blog: voidnish.wordpress.com Latest article: C++ 11 features in Visual C++ 2013 Preview

[TNCaver]

So we're all correct. However, from that wiki, "Camel case may start with a capital or, especially in programming languages, with a lowercase letter." See also the article's named anchor #Current_usage_in_computing.

If you think 'goto' is evil, try writing an Assembly program without JMP.

[Nish Nishant]

Yeah typically I'd assume these : camelCase PascalCase (same as TitleCase).

Regards, Nish

---

Blog: voidnish.wordpress.com Latest article: C++ 11 features in Visual C++ 2013 Preview

[Narf the Mouse]

There's one of two conclusions that can be drawn: 1) They tried to improve on the previous study by making a biased study. (The phrase refers to their own study) 2) They tried to improve on a previous biased study. (The phrase refers to the previous study they are comparing and contrasting their own with).

[RafagaX]

I find your lack of faith in camelCase and PascalCase disturbing...

CEO at: - Rafaga Systems - Para Facturas - Modern Components for the moment...

[BotReject]

Underscores may make text without spaces more readable, however, just how long do you think variable names should be? variable names that are too long degrade readability anyway. IUsuallyPreferThisStyle because I find it easier to type.

[Brisingr Aerowing]

Yes. It is used quite a bit in Python. (Seriously. It is.)

brisingr_aerowing@Gryphon-PC $ rake in_the_dough Raking in the dough brisingr_aerowing@Gryphon-PC $ make lots_of_money Making lots_of_money

[jibalt]

From the abstract: "subjects were trained mainly in the underscore style" ... ... which makes this study useless for determining which is better. And any proper study would have to accurately account for the cost of longer identifiers, which can make code layout more difficult, among other things.

[Chad3F]

How about we just simplify things and replace all these nasty and confusing concatenated word styles with something straight forward, like integer based identifiers.. Old:

int calculateFactorial(int num) /* or calculate_factorial() */
{
int value;
int i;

value = 1;

for(i = 2; i < num; i++)
value *= i;

return value;
}

New:

int x9F37A170(int x00000001)
{
int x00000002;
int x00000003;

x00000002 = 1;

for(x00000003 = 2; x00000003 < x00000001; x00000003++)
x00000002 *= x00000003;

return x00000002;
}

Now doesn't that just make things so much cleaner! :omg: And lets not forget the inherent localization benefits. :-D Of course, then there will be flame wars over whether to start it with a lower 'x' or an upper 'X', if the hex characters should be upper or lower, or do we use 32-bit identifiers or 64-bit. ;P