Re: Non-linear / full-2d writing systems?

Reply to Remi's points with some of Sai's replies to my former reply to
Remi - if y'all see what I mean  :P

On Friday, May 13, 2005, at 01:31 , Remi Villatel wrote:

> Ray Brown wrote:
>
> Note: 2DWS = 2D writing system. (My fingers refuse to type it once more.)
So NLF2DWS anyone?     :)

> 	[---CUT---]
>>> Yes and no. I said "A few words one above the other" because it's a
>>> convenient way to write them but that's of no importance. What matters
>>> is
>>> that these few words describe a mental image *together.* That's why I
>>> used a framing around my semagrams;
>
>> Like a cartouche around a proper name   :)
>
> Exactly but a colored background would have the same effect and it would
> remove the "cartouche's effect". Whatever way to separate groups of words
> would fit in.
Yep - but the colored background is a cartouche by another name, isn't it?
  I can see that a cartouche or something similar may be useful in certain
types of linear scripts (whether written along rows or columns or, indeed,
  a 2d mix of rows & colums). But within the context of a non-linear system,
  I agree more or less with Sai:
"Cartouches: they just strike me as aesthetically displeasing, and
wasteful of time."

The aesthetic bit is of course subjective, but they strike me as redundant
in non-linear set up.

>> But if the _words_ describe a mental image when taken together is it not
>> necessary to know how the words relate to one another, otherwise we could
>> finish up with ambiguities like the infamous "little girls' school"?
>
> Ambiguity is context related. If there were a context telling us what kind
> of girls you are talking about, there wouldn't be any ambiguity. Let's
> hope
> a 2DWS would help us clarify our mind and get rid of ambiguous
> expressions.
Of itself it would not.

To the above Sai replied:
"Er? But yes, /me reiterates suggestion. :-P"

But it was because Remi seemed not to take your suggestion on board that I
asked the above question. As I see it, Remi is still using linear elements
within a frame/ cartouche. I do not like leaving things like "a little
girls' school" to context to disambiguate. There are plenty of natlangs
that do not allow the phrase to be ambiguous (and plenty that do). A
NLF2DWS should not do worse than many natlangs on this!

BTW - for any readers who have forgotten Sai's suggestion, here it is
again:
"If you're going to treat them as a whole, why not fuse them 2d, or even
in a completely overlapping / integrating fashion? Why keep the bounds of
the constituent atoms and then string those along?"
AMEN.

>
>>> I think that losing data is the only purpose of a 2D writing system:
>
>> ?? Loosing _data_ doesn't seem a good idea to me. Nor have I considered
>> this the purpose of 2d writing.
>
> Ach y fi! 2 o's in "losing". I guess it's a typo.  ;-)
Not in the sense of typical typing error - just careless spelling error.

In English we do have a a verb "loose" /lu:s/ and also "lose" /lu:z/. But
we have "choose" /tSu:z/ (present) and "chose" /tS@ws/ (preterite) - ach!
what stupid spelling!!

>
>>> To lose datas
>
>> Ach y fi! Either a double plural or a pluralized count noun or,
>> hopefully,
>>   a typo  :)
>
> Grumble... Grumble... It's just a french word. At least, I never write
> "data" without an "s" in French. Grumble... Grumble...  ;-)
Right - I forgot French has these 'double plurals' like "les spaghettis".
I remember we has a young French teaching assistant lodging with us for a
year and we tried very hard to explain to her that we do not say "The
spaghettis are cooked"   :)

>>> that are unncessary for the reader to understand. Our linear writing
>>> system is so much full of superfluous data that we can get bored by a
>>> text
>>> on a subject that however interests us.
>
>> That sounds to me just like plain bad writing! If the data is not
>> necessary and, indeed, superfluous, then IMO it should not be there quite
>> irrespective of whether the writing is 1d, 2d or 3d.
>
> What I meant is that a 2DWS could enable you to ignore details that you
> don't need to be able to understand. In the linear writting, you must read
> from the start to the end before to be able to say "I already knew most of
> it". Jumping over a half sentence isn't practical.
Ok - I understand - not superfluous for the beginner, but superfluous for
an individual with some knowledge of the domain. Yep - I've been in that
position. But if I find this happening a lot, I conclude that the book is
not written at the level I want, put it down and look for something
'meatier'.

> A 2DWS should enable you
> to say "I know enough about this, I prefer to learn more about that".
OK.

> 	[---CUT---]
>> But I thought Sai wanted a writing that did _not require_ a reduction to
>> speech. So what has speaking got to do with it?
>>
>> Indeed, didn't Sai refer to the linearization required by speech as a
>> "bottleneck"? I quote:
> 	[---CUT---]
>
> I didn't read or I forgot. This thread is very dense, to say the least.
> Besides, that's nobody but Sai himself that brings back the subject of
> "linearization".
There's me   :P

It now seems a fairly important to me also.

> And what's the point of having a text that you can't speak
> about?
You can speak about anything. But I thought the whole point of a NLF2DWS
was to create a graphic form in which a person could record their thoughts
*without* recourse to the limitations of linearization.

> Sooner or later you'll meet somebody that read the same 2D text or
> somebody you wish he reads it to be able to talk about it.
Indeed - but the other person can still see the full text and see the data
that has become lost or mangled in the linearization process.

> Unless you want to start a cult of people who communicate only through 2D
> texts.
Nah - just those who communicate by telepathy    :P

> This confirm my idea that a 2DWS can only have an indirect connection with
> speech.
I agree. I don't see what advantage 2d has for recording (linear) speech -
it may be novel, intriguing, aesthetically pleasing etc - but advantageous?

> Now, excuse me but I have a bunch of ideograms to create until june 30th.
Bonne chance!
------------------------------------

Sai replied to the above point about linearization thus:
> So perhaps I should revise my hmmm above to read "not linearizable
> without loosing damn near everything in the process in a way that's
> cognitively irrecoverable".
Yes - I guess that's better.[snip]

> IOW, can you use the nonlinear-2d to encode something that is
> amazingly more intuitive / easy to grasp / dense than linear?
I would not have thought so.

Ray
===============================================
http://home.freeuk.com/ray.brown
ray.brown@freeuk.com
===============================================
Anything is possible in the fabulous Celtic twilight,
which is not so much a twilight of the gods
as of the reason."      [JRRT, "English and Welsh" ]

Replies