ISO Studies of underscores vs MixedCase in Ada or C++

[Andy Glew]

I am in search of any rigourous,
scientific, academic or industrial studies
comparing naming conventions in
C++ or similar languages such as
Ada:

Specifically, are names formed with
underscores more or less readable
than names formed with MixedCase
StudlyCaps camelCase?

....and similarly, any measurements
of programmer productivity, bug rate,
etc.; although IMHO readability matters
most.
* Religion - NOT?!

I understand that this is a religious issue
for many programmers, an issue of programming style.
I am not interested in a religious war.
I obviously have my own opinion, but I am
open to scientific evidence.
* Ada Studies?

I thought that I had seen studies like
this in some of the early design documents
for Ada, but I have not been able to find
such references on the web. Which is not
entirely surprising, since Ada was designed
prior to the web.

The Ada 83 and 95 Quality Guidelines recommend
underscores to improve readability, but provide
no source justifying this statement.
* What such studies might look like

Simple readability and recall:
- present a test subject with
a list of compound words
formed with underscoresand mixed case
- remove the list, and ask test subject
to write it
- score on accuracy

Program debugging
- present programs that are otherwise identical,
differing only in their use of underscores/MixedCase
to test subject programmers (e.g. a CS class)
- program has a known bug
- ask test subjects to find bug
- score on accuracy locating bug

Cruel TA study:
- Two sections of a CS class
- Enforce programming standards,
underscores vs MixedCase
- Pose a programming problem
- Score according to success
completing assignment

Empirical:
- Given version control databases
of large programs, some written in underscore
style, others in MixedCase
- Total bug rates normalized by LOC, name count, etc.
- OR: count only bugs that can be attributed
(after inspection of checkins) to misnamed variables

For that matter, I would be interested in any surveys
folks may have done that count projects and their
coding standards, possibly weighted
- open source (e.g. sourceforge)
- industrial
- textbooks, weighted by sales
- websites of coding standards, weighted by Google score...
Although this is less convincing than a rigorous study.
* Explanation of Newsgroups Chosen

I hope it is obvious why I have chosen these
newsgroups to post this search to:

comp.software-eng, comp.programmin g,
- an issue of software engineering
comp.lang.c++,
- the language I am most interested in
comp.lang.ada
- because I vaguely recall historical work

[Georg Bauhaus]

>>>>> "Martin" == Martin Dowie <ma**********@b topenworld.com> writes:

: "Matt Gregory" <bl************ ****@earthlink. net> wrote in message
: news:BPvfb.2044 5
:: > Agreed! I wish that more languages allowed hyphen use in
:: identifiers. > Dylan is the only one I can think of off the top of
:: my head.
::
:: Lisp and Scheme.

: COBOL

Also a few languages, like SNOBOL4, that allow you to have
any string as a variable name,

$'The Shoemaker - page 3' = 'Once upon a time'

ML allows minus {minus} as identifiers, and also ":->" and the like,
though not ":-)".
Georg

[Leif Roar Moldskred]

"Jakob Bieling" <ne*****@gmy.ne t> writes:

But it is that kind of word-matching I would personally like to see in
more compilers (specifically C++ compilers).

I do agree with Kevin Morenski (news:3f******* *@nntp2.nac.net ) that a
real spell-checker for source code is not practicable. You said that the
spell-checker would just have to ask you whether to ignore it or how else to
proceed. Have you thought about how annoying 100s or even 1000s of those
messages boxes, asking how to proceed, will be when compiling already
existing source with this spell-checker?

Oh, I wouldn't want it to be part of the compiler, or continiously on. Rather,
I'd just want to be able to run a command in my editor to spell-check the
source-code I'm currently working on - much like I today invoke ispell in
emacs for regular text files.

--
Leif Roar Moldskred

[Default User]

James Dow Allen wrote:

(I usually rewind a file with "lseek(fd, 0L, 0)" because I can't
remember if 0 is SEEKSET or SEEK_SET.)

The best choice for rewinding a stdio file stream is rewind().


Brian Rodenborn

[Jakob Bieling]

"Leif Roar Moldskred" <rm******@onlin e.no> wrote in message
news:86******** ****@huldreheim .huldreskog.no. ..

"Jakob Bieling" <ne*****@gmy.ne t> writes:

But it is that kind of word-matching I would personally like to see in more compilers (specifically C++ compilers).

I do agree with Kevin Morenski (news:3f******* *@nntp2.nac.net ) that a real spell-checker for source code is not practicable. You said that the
spell-checker would just have to ask you whether to ignore it or how else to proceed. Have you thought about how annoying 100s or even 1000s of those
messages boxes, asking how to proceed, will be when compiling already
existing source with this spell-checker?
Oh, I wouldn't want it to be part of the compiler, or continiously on.

Rather, I'd just want to be able to run a command in my editor to spell-check the
source-code I'm currently working on - much like I today invoke ispell in
emacs for regular text files.

Ah, I did not consider this difference even. *g* My point was, checking
my own identifier names for spelling errors based on natural language rules
is a bad idea. Instead, a special code-spell-checker would be great:

a/ 'Created' identifiers are automatically added to the list of known
words (for the current compilation run only)
b/ When an identifier is used, but not found, the spell-checker tries to
find out what I meant (like any spell-checker)
c/ Comments are spell-checked like in any other word processor, except
that identifier names should also be considered

So let us assume we have this (rather useless) piece of C++ code:

1 int main ()
2 {
3 char* pRok = new char;
4 *pork = 'p';
5 delete pRok; // do not forget t delete pRok!
6 }

In line 3, the spell-checker should not question the identifier name and
try to tell me that I meant 'pork' (see a/ above). Instead, it should tell
me in line 4 that I most probably meant 'pRok' and not 'pork' (see b/
above). And last but not least, in line 5, the spell-checker should point
out the missing 'o' (ie. ".. forget to ..") but leave 'pRok' undiagnosed.

regards
--
jb

(replace y with x if you want to reply by e-mail)

[Frank J. Lhota]

"Default User" <fi********@com pany.com> wrote in message
news:3F******** *******@company .com...

The best choice for rewinding a stdio file stream is rewind().

Nah, rewind() makes me think of tapes, and that makes me feel old.

[John W. Krahn]

Georg Bauhaus wrote:

>> "Martin" == Martin Dowie <ma**********@b topenworld.com> writes:

: "Matt Gregory" <bl************ ****@earthlink. net> wrote in message
: news:BPvfb.2044 5
:: > Agreed! I wish that more languages allowed hyphen use in
:: identifiers. > Dylan is the only one I can think of off the top of
:: my head.
::
:: Lisp and Scheme.

: COBOL

Also a few languages, like SNOBOL4, that allow you to have
any string as a variable name,

$'The Shoemaker - page 3' = 'Once upon a time'

You can do the same thing in Perl:

${'The Shoemaker - page 3'} = 'Once upon a time';

And since perl interpolates in double quoted strings you can include any
8 bit value:

${"\xFF\0\t\cV" } = 'Once upon a time';
John
--
use Perl;
program
fulfillment

[William]

"Jim Rogers" <ji************ **@worldnet.att .net> wrote in message
news:82******** *************** ***@posting.goo gle.com...

"William" <Re***@NewsGrou p.Please> wrote in message

news:<ud******* *************@g iganews.com>...

Talking to myself here... I occurred to me that its syntax highlighting
makes spell checking reserved words less necessary - and the syntax
highlighting can deal with case. -Wm

Even more to the point -- any compiler should be
able to properly recognize reserved words.
Why use another tool to check what the compiler will also check?

Until we got faster machines, one of our builds could take an hour.
No fun to have the compiler detect a mispelled word 50 minutes in.
(Sure, if everyone did an incremental test build, the problem would
be avoided, but...) -Wm

[William]

"Leif Roar Moldskred" <rm******@onlin e.no> wrote in message
news:86******** ****@huldreheim .huldreskog.no. ..

In the same way that spell-checkers for ordinary text today handles names and other words that are correct, but not in the dictionary: When detecting the unknown word the first time, ask the user what to do with it - whether to
correct it, accept this instance, accept all instances in this document or
add it to your private dictionary. (For a spell-checking of source-code we'd probably also want the option "accept all instances with this case.")

Well, they will if you add the entire label to the dictionary, but none I'm
aware
of will break up words based on mixed-case. Ultraedit will break up words
on underscores - didn't know that until I just tried it I can make it ignore
mixed-case words, but not break them apart. (It does support multiple user
dictionaries, so you could have a dictionary-per-project defining all the
approved labels.) -Wm

[Dave Thompson]

On Fri, 03 Oct 2003 15:42:20 -0700, Peter Ammon
<pe*********@ro cketmail.com> wrote:
<snip>

Or my preference if the tool allows

classmethod :
access-specifier
method-type-specifier
method-return-type-specifier
method-declaration
method-body

(In proportional fonts, hyphens are usually skinnier than
underscores and (to my eye) make the text more readable.
It's not as noticable with monospace fonts, but I think the
lower example looks better (read: more readable :-).)

Agreed! I wish that more languages allowed hyphen use in identifiers.
Dylan is the only one I can think of off the top of my head.

COBOL ! Only embedded, though. My compiler accepts -X as minus X with
a warning, but I don't think that's standard.

Also the usual suspects: FORTH allows everything but space (and
controls); and LISP everything but space, parens, period, quote,
backquote, and undoubled slash, if I haven't forgotten anything.

- David.Thompson1 at worldnet.att.ne t

[Robert Stankowic]

"Kevin Morenski" <[e-mail] center.net> schrieb im Newsbeitrag
news:3f******** @nntp2.nac.net. ..

// Number of misspelled words fuond so far
int noErorsInTetx = 0;

I want to spell-check this such that I get notified both on "fuond"
for "found", "Erors" for "Errors" and "Tetx" for "Text". They are all,
after all, words in natural language, and it should be possible to
spell-check them automatically.
Let's say you had a variable named "tHTa," for example. With respect to
your concept, this would be a misspelling of the word "that." Now, a lot

of programmers--myself included--use letters to represent certain things in
variable names. tHTa could mean "type HTa" or anything else a programmer
could think of. How could a program possibly differentiate between
conventions in the naming of variables?

It's much simpler to check the spelling of comments...prog rammers have
developed so many conventions for making their lives easier; a spell checker on variable names just adds one more problem to overcome.

An editor which is capable of finding the definition of variables and
highlighting a name already defined/declared might solve that problem. VB
has that feature to a certain degree - if you use mixed case for names of
variables, functions etc. and you type a name already declared in lowercase
the editor will automatically convert it to the form in which it is
declared. If you know, the name is declared and case does not flip you know
you made a typo. IIRC Keil development system for embedded C programming has
a similar feature.

just my $0.02
Robert