h a l f b a k e r y"More like a cross between an onion, a golf ball, and a roman multi-tiered arched aquaduct."
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Please log in.
Before you can vote, you need to register.
Please log in or create an account.
|
We are currently expected to learn more than two dozen letters, various punctuation symbols, arithmetic operator symbols and so forth. This is unnecessary.
Instead, just use three symbols and either RPN or prefix notation, thereby making brackets and priority superfluous. Introduce / or a white
dot for truth, \ or a black dot for falsehood and | for the Sheffer Stroke. Express all conjunctions reducible to logical forms through the NAND operator, so "but", "however" and "and" are all the same, as are "then" (much of the time), "therefore" and "so". Express arithmetic operations in the same way, but taking recursion into consideration due to the recursive nature of binary adders - no, i don't know how. Use binary codes for letters from ASCII or Unicode.
The problem of learning is then deferred from memorising dozens of symbols to applying them. It's to our current notation as ours is to Chinese. Chinese has thousands of symbols, we have less than a gross in common use, this would employ three. To further reduce overheads, introduce a convention to define macros.
This would be for visual communication rather than spoken, which would continue as before.
[link]
|
| |
Then watch our children get outsmarted by the Chinese, who are required to take the opposite approach. |
|
| |
Erm, don't we already globally manage rather well with only 2 symbols for everything? NAND can be encoded perfectly easily with a series of 0's and 1's, as can everything else - as this annotation proves by its very existence. |
|
| |
No, we get machines to do that bit. |
|
| |
[Spacecoyote], not sure how much time we waste
learning the alphabet and so forth when we could be
doing something else. |
|
| |
We will bun this because is it as step in the right direction (which as [zen_tom] points out is pure binary). |
|
| |
Binary is just a representation of the presence or absence of electrical current. So, if we pursue this line of thinking to its logical conclusion, then what we are really arguing for here is communication by cattle prod, isn't it? Do something right, no current. Do something wrong, bzzzzt! + |
|
| |
You don't say why - and I'm not quite sure of the context of the idea. Are we talking about alphabets for interpersonal language or alphabets for machine manipulation etc.? |
|
| |
Binary is great for machines because of the minimal alphabet {0,1} and using only NAND operations you can create any logical output - modern computing allows us to create systems that have billions of NAND operations and therefore lots of potential functions. Problem is - you need loads of operations & 1s and 0s to do this. This concept doesn't work so well for human language... |
|
| |
If you represent the 26 character English alphabet with a trinary system { \,|,/} then you get this kind of affair: |
|
| |
a = /// , b = //|, c = //\,
d = /|/, e = /||, f = /|\,
g = /\/, h = /\|, i = /\\,
j = |//, k = |/|, l = |/\,
m = ||/, n = |||, o = ||\,
p = |\/ , q = |\|, r = |\\,
s = \//, t = \/|, u = \/\,
v = \|/, w = \||, x = \|\,
y = \\/, z = \\| |
|
| |
So, to articulate "hi", we have to use "/\|/\\" |
|
| |
As you can see, we have reduced our alphabet from 26 to 3, but we have increased the length of the word from 2 to 6. It's a trade off. I do not have any trouble knowing the English alphabet, so reducing the alphabet down to 3 does not give me any benefit. |
|
| |
And I've not even got into symbols used in telecommunications...
{Do you send symbols that have 4 levels per unit time (0,1,2, or 3) or send symbols with 2 levels (0 or 1) but send them twice as fast?} |
|
| |
{Ans - depends on the channel that you send them over and the types of error you might get!} |
|
| |
Well, they could be in parallel rather than series.
Eight by eight grids but joined up like Marain's
three
by three ones. You stack up the variables,
constants
and/or literals you need, then operate on them
with
your stroke, or do it the other way round. |
|
| |
The context is textual, on the whole. |
|
| |
In fact, how about this? Twelve by twelve grid,
sixteen colours (two levels of each three-bit RGB
colour), pack six dozen characters into each
square. It'd take a long time to write though. |
|
| |
Also, the solution to the Sheffer stroke not being
binary can be solved by making it a six-byte long
line of set or clear bits. |
|
| |
Well - if it's textual, like on signage, then one metric of how useful the alphabet is the unit area that characters occupy. That implies using a larger alphabet so that any words are made up of less characters. |
|
| |
//In fact, how about this?// |
|
| |
Whoa! Stop right there. Why?! |
|
| |
//Twelve by twelve grid... six dozen characters// |
|
| |
You've lost me. Is this an exercise in working out the information content? Or a refinement of using "Three symbols for everything"? |
|
| |
You could get by with two symbols: True and Nand. False would be known as "/|/". |
|
| |
[Jinbish], i was mainly trying to see how much i could cram into a small space. |
|
| |
This would be a bit like Old Entish, in taking a very
long time to say anything. |
|
| |