David Starner wrote:
>On Thu, Feb 14, 2002 at 08:56:25PM -0500, Patrick Andries wrote:
>
>>>They are already encoded in Unicode, using two or more Unicode
>>>characters... using a colon and a closing parenthesis (I personally
>>>prefer the version with a "dash" nose) is all you need.
>>>
>>Methinks «We know what you need» is a bit patronizing.
>>
>
>That doesn't mean it's not right.
>
Neither that it is.
>There's a lot of absurd solutions created by people with problems, and a lot of solutions to problems that
>don't exist.
>
Absurd... Missing a smiley or two here, right ?
> There are a couple of "real" smileys too, but some modern emailers
>actually recognize the regular form
>
>>the « regular »... the contrived way you mean.
>>
>
>The regular way; the most common way; the way people actually use.
>
Well, because there is no other way with a keyboard. But what do people
do with a pencil ? What is the way people actually draw smileys then ?
Tilted 90° ?
>
>
>Unless Unicode is willing to dedicate several hundred characters to
>these, there will be many similies that will be unencoded.
>
Which is obviously an argument to encode none (or only those that are
"legacy"). Now, granted the problem is to determine what is the set that
could be encoded and here ISO/Unicode hasn't got its work cut out for
itself : there is no prior approved set.
> And unless Microsoft is willing to add it to their keyboards, most people won't be
>able to use it directly.
>
I admit that there is a practical limitation as far as inputing these
characters is concerned, but then how many Unicode characters has
Microsoft (?) added to its [US ?] keyboard.
> So once most systems support it - in what, 4-5 years? - programs may autoreplace the smilie.
>
They already do. I'm not really sure I understand you. Are you aware
that I didn't need to use the «regular way» to get ☺ and :-) ?
> So IM's will send 3 bytes across the net to replace three byte-sized ASCII characters, with the
>same net effect,
>
Are we really obsessed about byte size ? The effect is not net : you
would now have characters which can take different appearances (font
variants if you want). They can then be straight up (normal instead of
tilted), coloured or even animated.
> but having succesfully broken backward compatibility
>with anybody using older hardware or software.
>
Hardware ? Nothing less.
I wonder sometimes if the largest obstacle in the encoding of smileys as
characters is not the "universal" normalization process itself. Had
they been invented a few decades ago and encoded "locally" in some kind
of popular font/encoding (the Netscape font for example that could have
the iconic :-) :-( ;-) :-P :-D :-[ :-\ found in Messenger)
they might have been included in Unicode without much further ado. I
personnaly see them as punctuation mark (albeit not of "metaprosodic"
nature).
Now, I believe no one feels the hindrance of the « regular way » is so
great thatthey are willing to fight about it and be pointed to the FAQ
without any smileys. Not me in any case.
P. Andries
This archive was generated by hypermail 2.1.2 : Thu Feb 14 2002 - 22:19:02 EST