Gerhard Fiedler wrote:
> Especially in those places, mu should be available. On the Internet,
> all the necessary standards are in place for a long time to represent
> mu correctly and platform-independently. Pretty much all web server
> script languages have ready-made functions to convert user input into
> the appropriate HTML entities. Character sets are well-defined, and
> so are the ways to define which one is used. Browsers and newsreaders
> handle that and display that correctly. It's only not available when
> there's a bug, or an inept programmer :)

I've been watching this debate on the sidelines for a while, but you're
constant complaining about how everyone that doesn't follow the standards is
inept is getting me fed up enough to respond.

Most things that we have today evolved via a long list of individual
decisions that largely made sense *** at the time and context the decision
was made ***.  This process doesn't always lead to the optimal end result.
The end result is merely the result we ended up with, just like with
biological evolution.  And it's also not a final answer as things will
continue to evolve.  Just because it's not what we wish it were or because
we can imagine a better result doesn't make the individual decisions wrong
for the most part (this is not to say there aren't occasional bonehead
decisions along the way, but mostly that's not the case).

Look at the use of the metric system in the US.  I think a lot of people
would agree that metric is a better system on the whole and it would be a
benefit if everyone used it (the english system has some benefits too, but
let's stipulate metric is better for this discussion).  However, that
doesn't make it stupid to buy a set of english wrenches.  My first set of
tools were all english because my first car was a 1972 Ford Pinto which
required english tools.  I bought that car for a variety of reasons, and
whether it required english or metric tools wasn't on the list of criteria.
Meanwhile I have acquired metric equivalents of most dimensional tools
because occasionally I run into metric hardware.

Now think about a mechanic who has 1000s of $$ of english tools because
that's what was required for the cars he works on.  Along come cars with
metric hardware.  Is this guy going to be happy to go out and spend another
$1000 on metric tools?  No, but he'll probably do it while grumbling all the
time.  He probably buys only what he needs in metric.  He grew up with
english and probably thinks in inches and feet.  If he's going to build
something himself, it will probably be in english units.

What about people building a few things on the side in their basement?  They
have a large collection of english tools.  What kind of bolts and nuts do
you think they are going to buy when they get to the hardware store?  I do
this too.  I've got a good feeling what #4 or #10-32 or 1/4-20 machine screw
is.  Yes, I could convert that to metric and find an equivalent metric size,
but when I'm working on something in my basement why should I?  That's what
I grew up with, so that's what I think in, and converting requires more
mental work.  So what are the hardware stores going to stock?  Whatever
their customers demand, right?  So why should I buy a stock of metric
hardware when I've already got a nice stock of english hardware and it's
more available anyway?  (Actually metric hardware is available too, but not
quite as widespread as english.)

So who's doing anything stupid here?  Everybody is just doing what makes
more sense for them each little decision along the way.  Eventually things
will change as more and more stuff gets manufactured in volume using metric.
Cars are just about all metric now.  Kids newly learning auto mechanics will
probably be more exposed to metric and eventually there will be a generation
that thinks in metric because that's what they were exposed to when they
grew up.  It all takes time.  Just about all scientific work is done in
metric, most large industries are metric, the military is all metric,
electronic measurements have always been metric, there's just the minor
problem of 300 million people not having grown up that way.

If you were Outer Vulgaria and woke up at the end of world war II taking a
leap from ox carts to cars in a span of 5 years, it's no big deal to decide
to standardise "about that long" with a meter.  It's a lot more trouble for
a technologically advanced society like the US that already has an
entrenched system and existing infrastructure.

As for capacitance, I've never understood why that was the only value where
milli and nano were skipped.  When I grew up it was all micro or pico, with
pico sometimes called micro-micro.  Rarely I did see mF mean micro-farads,
but I haven't seen that for a very long time.  The old timers I dealt with
when I was learning only ever used micro and pico farads, but I grew up with
enough metric to think that was silly.  I make it a point to use milli and
nano farads whenever appropriate, partly because I've deliberately taught
myself to think that way and partly because I want to get the world used to
it.  I've only ever once run into a problem with this.  There is a 1mF
capacitor in the EasyProg, and the guy who put the kits together took that
to mean 1uF even though there were other caps labeled with uF and that I had
pointed that out to him ahead of time.  He had to retrofit a bunch of kits
with 1mF capacitors for free.  I think of this as one small victory, because
I doubt he's going to make that mistake again.

As for the mu symbol, I've looked upside down, inside out, and sideways at
my keyboard and I don't see one.  Yes, there may be a way to make a mu
symbol with a bunch of keystrokes, but I don't know that off the top of my
head and it would be too much hassle anyway.  Using "u" for mu is a well
accepted and very common practise.  I'm quite sure that when I write "22uF"
every last person reading this knows exactly what it means.  Yes there is of
course a unicode character for mu, and also in the iso-latin-whatever
character set.  But in reality if it's not ASCII it's either not safe or too
much of a hassle.

Unicode is a big pain in the butt because it takes 16 bits per character.
I've got a huge amount of software that uses 8 bits/character.  This means I
intend to ignore unicode to the largest extent possible.  It also means that
new software I write will not use unicode since it will be layered on many
existing utility routines that all use 8 bit characters.  And if I'm adding
to the utility routines, I'm going to stick to 8 bit characters because else
I couldn't make incremental changes.  So my software is going to use 8 bit
characters for a long time yet to come, hopefully that will be good enough
for as long as I'm around.  Since the iso-latin-whatever character set fits
in 8 bits and it's a superset of ASCII, and it contains all the character
that I know how to use, the really really good reason to switch to unicode
is just missing.  Maybe the coming economic pre-eminence of China will
change this.  When writing HTML and I think about it and I have time, I do
use the "μ" construct for mu.  That's OK since it's guaranteed to work in
all browsers (part of the HTML definition), I can remember it, and the
resulting file still uses 8 bit characters.

> You don't like non-standard web sites that work only in IE
> on Windows, do you?

Actually I couldn't care less since I use IE.  Again you are forgetting that
everyone makes individual decisions that are the best for them at the time.

> I'm pretty sure you prefer standard-compliant web sites that can
> be viewed properly on a MacOS X system with Mozilla :)

No, I don't give a crap as long as the web sites look right in *my* browser,
which happens to be MSIE.

However when I write web pages that I want a wide audience to see, I try to
stick to lowest level common HTML.  But that's because in that case it's in
my interest.

Again, my main point is that you are whining about people doing things
different from how collectively you think it should be done, but ignoring
what is in each individual person's best interest.  Yes, it might be good if
we could all suddenly use the same standard, but who's going to pay for
that, and why should I pay if I don't perceive the value to me outweighs my
cost?  Are you willing to go thru a few 100000 lines of my source code and
convert every last character reference to unicode?  I didn't think so, so
stop whining about it already and stop calling me inept for not doing it.


******************************************************************
Embed Inc, Littleton Massachusetts, (978) 742-9014.  #1 PIC
consultant in 2004 program year.  http://www.embedinc.com/products
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist