A couple of years ago I tried to find out whether
somebody has made experiments related to word
separation in names. Unfortunately I do not have
these old e-mails available, but my conclusion
at the time was that no such experiments or
psychological tests have been made.
In my books I have used underscores to separate
the words of the names in example programs. I have
justified this on page iv of my Java book that
can be found at
http://www.naturalprogramming.com/javabooksamples/some_reasons_to_choose_javabook.pdf
To test whether names like integer_from_keyboard
or names like integerFromKeyboard are more readable
is not an easy task because most programmers
already possess strong opinions related to these
naming styles. One possibility might be to use
non-programmers as subjects. This might mean that
a group of 'ordinary people' would read a text like
ThisIsAPsychologicalTestToFindOut
ImportantThingsRelatedToImportant
Matters. ItIsVeryImportantThat ...
Then another group would read the same text that
is written in the following way
this_is_a_psychological_test_to_find_out
important_things_reated_to_important
matters. it_is_very_important_that ...
Then these groups would be tested with some
clever questions.
Best regards,
Kari
Mr. (Dr.) Kari Laitinen
Oulu Institute of Technology, Finland
http://www.naturalprogramming.com/
Richard A. O'Keefe wrote:
This sounds like one of the things that *must* have been studied
properly by someone, but I have been unable as yet to find any
published empirical results:
Are multiword identifiers easier to read if the words are separated
(by hyphens as in COBOL or lisp or underscores as in Ada or C)
or if they are run together in baStudlyCaps style?
I have searched the PPIG mail archive, so I am aware that there has been
work on *typography*, which comes close but doesn't ring the bell, and I
have found published work about using whole words or abbreviations, which
again comes close but doesn't ring the bell. I am also aware of the
advice in the Ada Quality and Style Guidelines (which has no reference to
any published work on this particular topic) and in 'man perlstyle', and
in Betrand Meyer's OOSC book. I've also spent a couple of hours trawling
the web.
Sufficiently old studies may fail to ring the bell for another reason:
the readability difference, if any, between ATROPHY and A_TROPHY is
arguably different from that between aTrophy and a_trophy.
----------------------------------------------------------------------
PPIG Discuss List (discuss@ppig.org)
Discuss admin: http://limitlessmail.net/mailman/listinfo/discuss
Announce admin: http://limitlessmail.net/mailman/listinfo/announce
PPIG Discuss archive: http://www.mail-archive.com/discuss%40ppig.org/
----------------------------------------------------------------------
PPIG Discuss List (discuss@ppig.org)
Discuss admin: http://limitlessmail.net/mailman/listinfo/discuss
Announce admin: http://limitlessmail.net/mailman/listinfo/announce
PPIG Discuss archive: http://www.mail-archive.com/discuss%40ppig.org/
