From thinkit8@lycos.com Tue Dec 04 18:50:56 2001
Return-Path: <thinkit8@lycos.com>
X-Sender: thinkit8@lycos.com
X-Apparently-To: lojban@yahoogroups.com
Received: (EGP: mail-8_0_1_2); 5 Dec 2001 02:50:57 -0000
Received: (qmail 61260 invoked from network); 5 Dec 2001 02:50:57 -0000
Received: from unknown (216.115.97.167)
  by m3.grp.snv.yahoo.com with QMQP; 5 Dec 2001 02:50:57 -0000
Received: from unknown (HELO n21.groups.yahoo.com) (216.115.96.71)
  by mta1.grp.snv.yahoo.com with SMTP; 5 Dec 2001 02:50:56 -0000
Received: from [216.115.96.162] by n21.groups.yahoo.com with NNFMP; 05 Dec 2001 02:45:03 -0000
Date: Wed, 05 Dec 2001 02:49:24 -0000
To: lojban@yahoogroups.com
Subject: Re: thoughts on numerical language
Message-ID: <9uk1vk+4o87@eGroups.com>
In-Reply-To: <13f.59d807a.293ee25f@aol.com>
User-Agent: eGroups-EW/0.82
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Length: 1700
X-Mailer: eGroups Message Poster
From: thinkit8@lycos.com
X-Originating-IP: 216.26.3.195
X-Yahoo-Group-Post: member; u=71054096
X-Yahoo-Profile: thinkit41

--- In lojban@y..., pycyn@a... wrote:
> In a message dated 12/4/2001 6:05:56 PM Central Standard Time, 
> thinkit8@l... writes:
> 
> 
> > In the end, the quickness of expressions determines what gets 
> > expressed, too (isn't that Zipf?).
> > 
> 
> Not really. At most it would be that what is most shortly 
expressed gets 
> said most, but even that is not quite right -- and is wrong 
way 'round.
> 
> <Viewed numerically, a typical phonology can be 
> thought of as a mixed base number perhaps>
> 
> As can anything with a bit of ingenuity. What is the point here?
> 
> <I think it will differentiate itself when you start looking at 
> things that are just too cumbursome that they are never expressed 
in 
> a human language. For example, in a binary language it's easy to 
> imbed something like a bitmap to directly describe a flat picture 
> (or indeed any flat bit string, like a DSD sound).>
> 
> Well, now we have gone beyond language to including the thing 
itself (yes, I 
> know that the picture, nor the jpeg (or whatever) of the picture 
is not 
> strictly the thing itself but it fails to be in a rather different 
way that a 
> linguistic reference or a linguistic description fail to be -- and 
a way 
> closer to the thing). I don't mind illustrated text, but I think 
calling the 
> illustration a part of the text is pushing terminology a bit too 
far.

I'd think of it as a superset. If you define something numerically, 
you can do both the language stuff, and the illustration, which is 
not a "linguistic description". Sure you can do that now, but you 
get rough boundries between them. A rough example is binary 
encoding in newsgroup texts...it's a hack at best.