A 03:57 09/01/98 -0800, Michael Everson a écrit :
>Ar 09:22 -0500 1998-01-08, scríobh Alain LaBonté:
>>A 20:32 07/01/98 GMT, John Clews a écrit :
>>
>>>Don't forget that ISO NP 15924 is NOT an IT standard, so not all
>>>users of this standard will have the opportunity to do such
>>>conversions by computers, even though such codes may well nbe useful
>>>in IT systems. Meaningful alphabetic codes are therefore likely to be
>>>useful.
>>
>>[Alain] :
>>Dont't ignore the fact that for ISO 639 or ISO 3166, which are similar in
>>nature, I often have to go to a list to check what « IL » means for example
>>(Israël, and not Italy) or what « iu » means, and that most Californians
>>think that CA means their state rather than Canada.
[Michael] :
>And you suppose that you will have to go _less_ _often_ to the list if for
>Israël the code « 376 » is used? :-)
[Alain] :
No. You, by chance, arrive on my ground, where I am starting from. It will
force the making of good user interfaces which will be universal and not
biased to make some languages more equal than others. And if you still have
to enter a catalog number, you will be more likely to have a list of
equivalences in your language, whatever that language is. Sensible enough,
isn't it? Simple. Nobody complains much in practice when they have to enter
catalog numbers to order goods, catalog numbers that they will never
furthermore have to reuse. Of course you can always make mistakes (although
there are statistically not so many, otherwise it would not make sense at
all economically and even credit card numbers would not exist any more),
but with codes like "IT" or "IR" in ISO 3166, you can also very easily hit
the wrong key as r and t are adjacent keys on most European-like keyboards.
And if you use numerical codes often you can also remember them. Count the
*number* of *numbers* you remeber, you'll be shocked to see how good your
memory capabilities are. I know that 514 is the telephone area code of
Montréal even if mine is 418 in Québec, I know that in Toronto I have to
use 905 to call the Western part of the city and 416 in others. I also know
that Ottawa is 613 and the Québecer bank of the Ottawa River is 819. And if
I don't know what is 408, I look in my phone book. After 2 or 3 occasions
it's over, I know it. I know when I call France that the country code is
33, I call there often enough to remeber. I also know by heart many
9011331xxxxxxxx 15-digit-long number sequences I have to dial from my
office even if I make calls to Paris only occasionally, but often enough to
remember (and I know my 14-digit calling card number by heart, beacuse I
use it so often, it's like a song, I even know the dial tones by heart and
I react if it is not the correct tune when my computer dials such a number
for me! I never made an effort to learn them, otherwise I would be in
trouble, I also know that! You have to know how the human brian works, it
has wonderful capabilities, like learning your mother tongue in the firts
place, it is almost supranatural [it is, in fact, metanatural, in my humble
opinion (; ]!)
I know by heart that the country code of Switzerland is 44 and Genève's
area code is 22 because I also call there sometimes. I also know by heart
all my all-numeric PINs (I make business with several banks [we say NIP in
French, btw, I have to remember the English acronym to tell you!]), my
email and Internet servive providers' passwords and many other such numbers
(9-digit social security nummer and so on)... I very rarely make mistakes
*with numbers*, in fact I do many more typing mistakes entering more
obvious things, lik ethis email note, because I don't pay attention as much
(my extremely fast dyskinetic fingers believe they are as fast as my brain
and they seem not to care mucha bout the fact that it is not true! when
they deal with numbers, they have, because I force them to slow down, our
brain is still a despotic master if it wants).
Of course, these memory capabilities are, even more, true for specialists
of any domain who use codes repetitively every day. No question about that.
Do you ever compalin about the myriads of acronyms you have to learn and
about new ones, non repetitive, that you have to suffer thousands of times
a day? I do. They use letters, not numbers, and there are infinite
collisions. Numbers are less hypocritical.
Numbers do not force us to use a hinge language or some hinge languages
that are more equal than others. That is the issue, an issue of
universaltity, of openness. You, in particular, have the same goals I have
in these issues (I have known that for a while, and that's *one* of the
reasons why you're my little brother), so it should not be a problem for
you to undesrtand this (even if you still have the right to disagree, of
course). And if you use numbers you don't have to discuss about going from
2 to 3 letters if you add codes, as numbers should be fixed in concrete for
centuries if we are wise enough. They can be made more stable than country
names. They are even more flexible than clay, the clay which some believe
the world is made out. They are wonderful tools, universal tools, absolute
tools (they are the only absolute thing the existence of which human beings
are sure).
And machine-wise (that's a side effect, not the main issue here), 16 bits
are enough for roughly 65000 combinations, and 32 bits enough for 4 billion
(four milliard to avoid all confusion with words (: , 4 294 967 296
exactly, for purists, just in case you would be short by 200 million (; ).
That's efficient. Machines do not require to enter 10-character-long
strings, only 32 bits. It can be an issue, and it is one, in informaton
technology, which is used universally these days, but to serve humans, not
to enslave them, of course. They are tools, we should use them as such, and
we do, you do, he does, she does, they do, I do, even Thou (who is also We)
do... (;
Alain LaBonté
Québec
President and founder of AAA, Anti-acronym association (;
______________________________________________________________________
Quote of the day (no joke, this is highly human, not random-generated:
"Abar bs gur glcvat zvfgnxrf va jung cerprqrf jnf vagragvbany, ohg V qvq
abg pbeerpg gurz ba checbfr, sbe vyyhfgengvba."
This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:38 EDT