Re: Why (or not) use single letter variable names?
From: Shmuel (Seymour J.) Metz (spamtrap_at_library.lspace.org.invalid)
Date: 12/18/04
- Next message: Shmuel (Seymour J.) Metz: "Re: WAR IMMINENT betweeen Mathematicians and Physicists!"
- Previous message: Shmuel (Seymour J.) Metz: "Re: Existence of countable Hamel basis"
- Maybe in reply to: Ben Pfaff: "Re: Why (or not) use single letter variable names?"
- Messages sorted by: [ date ] [ thread ]
Date: Sat, 18 Dec 2004 18:01:14 -0500
In <yI6dnXJ9Mp_Hfl_cRVn-3A@rcn.net>, on 12/17/2004
at 01:08 PM, jmfbahciv@aol.com said:
>Heh. It's been my experience that people like you created bigger and
>more messes because of an inability to see a complete computing usage
>picture. ;-)
Really? Seeing a complete picture requires thought prior to coding,
which was precisely my point.
>Your thinking is IBMish, mine is DECish.
DEC also did things like requirement analysis. I don't know whether
that remains the case now that they're Hewlett PacPaq.
>You clearly do not understand the issues from the point of the very
>beginning of a project to the point where the code is cycled into
>maintenance mode while being merged into the next development cycle.
I've been involved in all phases, and not just with IBM hardware or
software.
>Not at all. I belong in the don't waste the coders' time with
>useless and long meetings and paperwork.
Then it's a good thing that I didn't suggest useless and long meetings
and paperwork, isn't it. But I understand that if you can't rebut what
I wrote, you are reduced to rebutting things that you invented out of
the whole cloth.
>This shows that you don't what you're talking about. There is never
>time to do it right;
Perhaps not where you're worked.
>the code would never ship if that were the exit criteria of field test.
Oddly enough, I've been on projects where we had code and design
reviews yet managed to ship ahead of schedule.
>Not at all. You seem to think that code developers have to do all
>the work involved with starting a project and finishing that
>project.
Sorry, Uri Geller, but you're wrong again. What I think is that iif
the coders aren't involved in the design then you're going to have
problems.
>About 5% of the manhours needed to do a project is coding;
Water is wet. Your own figures should tell you why keyboard speed is
not an important factor.
>I know you know that.
Indeed, but the above was the first sign I saw that *you* knew it.
>This tells me that you know better; "acceptable cost" means that
>there is a compromise between correct code and working with what
>you've got running.
That compromise is quite remote from optizing for coding speed or run
time.
>The coders I'm talking about used mostly MACRO-10 which is an
>assembler.
Then it's likely that your standards derived from the word size on the
PDP-6, which just happens to be 36 bits.
>yes, and all of a sudden
FSVO "all of a sudden" that is truly bizarre, given that the subject
of this thread is "Why (or not) use single letter variable names?"
>and all of a sudden you're redefining "long" to be 2<x<6.
No, I'm just noting that in this thread the only Mathematical names
mentioned other than single-letter variables were short function
names, e.g. cos.
>We were talking about variables that have enough characters to
>traipse across a listing page.
In your dreams. Had you said that by "long" you meant 40 characters,
then I would not have defended bah-long variable names.
>Oh, honey. Now you're in putdown mode to save face.
No, I'm in put-down mode because that was the mode that *you* chose to
operate in. Had you politely disagreed then I would have phrased
things differently. However, I see no reason to be polite to rude
fools with delusion of adequacy.
>Do you have any idea about how people read?
Yes; poorly, in many cases. In your case, for instance.
>Then you don't know your job well enough.
You're clutching at straws; you don't have the remotest concept of
what I was developing or how well it turned it. You are, of course,
wrong yet again.
>Most variable names are surround by constructs of the language. So
>that implies that a line of code will be the variable name length
>minus 30 characters or more.
It imples no such thing. Perhaps you don't understand what the word
"implies" means.
I've seen plenty of code where the syntactiv overhead was far smaller
than your 30 character value.
>If the variable name lengths are greater than six, the lines of of
>code become too long, produce wraparound, and are simply unreadable
>without a straightedge to match the first character of the coding
>line with the last character of the coding line.
What are you smoking? I've seen plenty of programs with 10-20
character names where the lines of code where not too long and were
perfectly readable without a straightedge. As for wrap-around, that
can make the program more readable rather than less, assuming that you
indent appropriately. If you're talking about random wrap-around, then
you have bigger issues than variable length.
>The code becomes a spaghetti mess if more than one of these
>long-length variables show up in one line of code.
FSVO long considerably larger than your six characters.
>We were talking about lengths that fully documented the purpose of
>the variable.
What do you mena "we"? I was talking about namesthat gave a brief
reminder of the purpose of the variable.
>I sure hope you get to this last part before you killfile
>me...again.
Thanks for the reminder; my standard timeout of one year is too short
for hypocritical fools like you.
*PLONK*
-- Shmuel (Seymour J.) Metz, SysProg and JOAT <http://patriot.net/~shmuel> Unsolicited bulk E-mail subject to legal action. I reserve the right to publicly post or ridicule any abusive E-mail. Reply to domain Patriot dot net user shmuel+news to contact me. Do not reply to spamtrap@library.lspace.org
- Next message: Shmuel (Seymour J.) Metz: "Re: WAR IMMINENT betweeen Mathematicians and Physicists!"
- Previous message: Shmuel (Seymour J.) Metz: "Re: Existence of countable Hamel basis"
- Maybe in reply to: Ben Pfaff: "Re: Why (or not) use single letter variable names?"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|
