Kari,

>   However, in my old thesis, in address
>http://www.naturalprogramming.com/to_read/doctoral_thesis_of_kari_laitinen.pdf
>   I list on pages 34-38 many good reasons to believe
>   that using natural names might be a "good thing".

I will leave readers to make their own minds up about that.

>   It is true that I preach about the use of natural
>   names because I have positive personal experiences
>   about it,

My own experience with teaching programming to raw beginners
(which last occurred 25 years ago) is that they tend to use
single character names, single character + digit, or words
like temp.  They do need some instruction on the benefits of
using 'meaningful' names.

> but unfortunately I cannot scientifically
>   prove my opinions. The epilogue of my C++ book

I think you ought to prefix any recommendation about the use
of natural names with this statement.

>   I believe that I am not the only person who preaches
>   against the use of abbreviations in computer programs.

Abbreviations in themselves are not the problem.  The problem
is that there are only a few abbreviations that are widely recognised.
If identifier names are intended to convey information to the reader
then the semantics associated with any words, acronyms or abbreviations
needs to be recognised.  Kari, if you used natural names built from
Finnish words I would consider them unnatural, because I don't
understand Finnish.  The issue is one of communication of
information, not the form of communication (which can include
abbreviations).

>   Some gradual developments suggest that people are
>   finding long and more natural names a "good thing".
>   According to my observations, the names in books
>   about computer programming have become longer during
>   the last decade or so, and the names used in class

This reminds me of a claim to respectability argument made by
proponents of psychic phenomena a few years ago.  The claim
was that publication of a paper dealing with psychic phenomena
in a highly regarded, peer reviewed, journal showed how seriously
the scientific community took this phenomena to be.  The fact that
what the paper said was less than complementary was glossed over.

>   I wish you a nice weekend, and, if you think that
>   this overcomputerized world is sometimes too hectic,
>   you might calm it down by using the truly slow imaginary
>   computer in address
>http://www.naturalprogramming.com/IC8/index_for_ic8.html

I think the use of human language words in identifier names can
provide worthwhile benefits.  However, there are other factors
that need to be taken into account; also the use of natural language
words does have a non-zero cost.

The proof-read version of my own work was not boiled down to
an empty file, i.e., the correct url is
www.knosof.co.uk/cbook/sent782.pdf


derek

--
Derek M Jones                                           tel: +44 (0) 1252 520 
667
Knowledge Software Ltd                            mailto:[EMAIL PROTECTED]
Applications Standards Conformance Testing   http://www.knosof.co.uk


 
----------------------------------------------------------------------
PPIG Discuss List ([email protected])
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/

Reply via email to