From: WarmSpgs@aol.com
Message-ID: <c3.1e890ba5.29ac135f@aol.com>
Date: Mon, 25 Feb 2002 17:23:27 EST
Subject: Re: LF: Spelling of units
To: rsgb_lf_group@blacksheep.org
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

In a message dated 2/25/02 12:25:57 PM Eastern Standard Time, 
j.r.moritz@herts.ac.uk writes:

<< A lot of the variations on symbols for units seem to be due to computers; 
 for instance, most CAD and simulation software can't do Greek letters and 
 is not case-sensitive. The Pspice simulator that I use at work a lot 
 interprets a resistor of either 1m or 1M as 1 milliohm - you have to put 
 1meg if you want a 1 megohm resistor. Confusingly, while 4.7k gives what 
 you would expect, it will interpret 4k7 as 4 kilohms, and ignore the 7. >>

Interesting point, Jim.  I suspect most of that has to do with the computer 
programmer's personal preferences, however, rather than any limitations of 
the computers themselves.  I cannot think of any modern operating system or 
language that is incapable of distinguishing between uppercase and lowercase 
letters in string variables.  While CAD and simulation developers may value 
backward compatibility with early 4- and 8-bit microprocessors having limited 
display capabilities, surely there must be limits to such a noble notion! 
(grin)

Even the 4k7 versus 4.7k distinction reflects the programmer's bias, rather 
than a hardware or software necessity.  It takes no more than half a dozen 
lines in most high-level programming languages to recognize a "word" having 
the form digit-letter multiplier-digit as being equivalent to 
digit-decimal-digit-letter multiplier, and converting it to the same 
numerical value for computation.
 
<< Also, the word processor I use puts nasty red lines underneath 1 
 mA, but seems to approve of 1mA - perhaps it is American influence at 
 work! >>

Actually, we Americans are just following RadCom's lead on this.  :-)

In reality, it's not clear to me why a word processor should observe any rule 
on this at all, but you are right about some of them doing so.  Very odd, 
considering that in the past few days we've seen on this reflector that 
national standards bodies do just the opposite.  Perhaps RadCom's usage is as 
VK7RO says...publications don't want to have the numeral(s) appear at the end 
of a line and the unit at the start of the next line.

Ironically, this is one of the points where a computer program like PSpice 
would find it easier to have no space between the value and the unit.  If an 
application clearly sees "4.7k" or even "4k7" in a list, it can simply 
disregard all blank spaces and doesn't have to employ quite as large an 
algorithm as if it had to also make the determination that "4.7 k" means the 
same thing too.

<< I think the use of * to indicate multiplication stems from many 
 types of computer languages which use this notation to avoid the ambiguity 
 possible if 'x' or '.' were used instead in a text-only display. >>

I would agree that's probably why it is so common.  In typesetting and in 
reasonably advanced maths software, fortunately, we have less ambiguous 
symbols for indicating the product of two expressions.  But since there are 
no ordinary keyboard equivalents, we have to make do with what's available.
 
<< The standards bodies seem to have been determined that we should use the 
 new logic symbols for a couple of decades now - engineers around the world 
 seem equally determined not to use them! The resistor symbol is a bone of 
 contention too, with many still preferring the zig-zag line over the 
 rectangular box. >>

Guilty as charged.

<< Some old American books seem to use 'M' for ohms - so there are lots of 
 50,000M resistors marked on circuit diagrams, which looks a bit strange. >>
 
Hmm.  I'm rather fond of old American electronics books, and can't say that 
I've encountered this peculiarity.  I wonder if it could result from a 
technical draughtsman's misreading of the author's hand-scrawled omega.

Or--could it possibly be that, far enough back in the dim mists of early 
radio, "mille" was an acceptable prefix meaning 1000, before the less 
ambiguous Greek prefixes became the norm?  Is anyone on this list both old 
enough, and willing to admit to it, to remember for certain?

73,
John Davis