Re: Non-linear full-2d writing (again)
| From: | Yahya Abdal-Aziz <yahya@...> |
|---|
| Date: | Monday, January 23, 2006, 9:50 |
|---|
Hi all,
1. On Sun, 22 Jan 2006 Sai Emrys wrote:
>
> First off - has anyone made progress on this since I first brought it
> up in Mayish?
>
> Second, a few specific points for pondering:
> * how to have sememes that overlap or intertwine?
> * how to have the 'web' interconnected at higher-order levels (e.g.
> paragraph plus)?
> - would probably require that the subparts be semantically connected
somehow...
> * how to show a conversation or interplay of ideas?
> - how to deal with converting, or obviating, the temporal-sequential
> nature of normal conversation?
> - perhaps have one color per speaker, & some point-counterpoint
> relationships of sememes - so it looks like a normal web, except
> multicolor?
>
> ... Any others that y'all have thought up or philosiphized about since
> the last big NLF2DWS threads?
See (*1 below)
> Personally, I still find it very difficult to think about sufficiently
> to get an idea of what it would look like concretely; I have some
> abstract ideas of how I'd want it to work (still very much like
> Heptapod B in skeleton) but no way of even sketching an example.
See (*2 below)
2. Jefferson Wilson replied:
>
> Sai Emrys wrote:
>
> > First off - has anyone made progress on this since I first brought it
> > up in Mayish?
>
> The Glyphica Arcana is a two-dimensional system, but I've joined
> since May.
>
> > Second, a few specific points for pondering:
> > * how to have sememes that overlap or intertwine?
>
> Sememe?
>
> > * how to have the 'web' interconnected at higher-order levels (e.g.
> > paragraph plus)?
> > - would probably require that the subparts be semantically connected
somehow...
>
> The GA doesn't do this, sentences exist in separate blocks. I'm
> not sure yet how to distinguish paragraphs.
>
> > * how to show a conversation or interplay of ideas?
> > - how to deal with converting, or obviating, the temporal-sequential
> > nature of normal conversation?
>
> Aren't conversations linear by definition?
See (*3 below)
> > - perhaps have one color per speaker, & some point-counterpoint
> > relationships of sememes - so it looks like a normal web, except
> > multicolor?
>
> Interplay of (feedback between) ideas I'm still playing with.
> Nothing has been satisfactory so far.
>
> > ... Any others that y'all have thought up or philosiphized about since
> > the last big NLF2DWS threads?
> >
> > Personally, I still find it very difficult to think about sufficiently
> > to get an idea of what it would look like concretely; I have some
> > abstract ideas of how I'd want it to work (still very much like
> > Heptapod B in skeleton) but no way of even sketching an example.
3. Sai Emrys replied:
>
> > The Glyphica Arcana is a two-dimensional system, but I've joined
> > since May.
>
> Linky?
>
> And what do *you* mean by "two-dimensional"?
>
> > > Second, a few specific points for pondering:
> > > * how to have sememes that overlap or intertwine?
> >
> > Sememe?
>
> The equivalent of a "word", but exclusively depicting meaning (not sound /
etc).
See (*4 below)
> > > * how to show a conversation or interplay of ideas?
> > > - how to deal with converting, or obviating, the temporal-sequential
> > > nature of normal conversation?
> >
> > Aren't conversations linear by definition?
>
> Transcribing a spoken conversation? Yes, certainly.
>
> But I would say that writing the *essence* of a conversation - i.e.
> the interplay of ideas - need not necessarily be linear. (At least,
> for conversations that have actual content.)
See (*5 below)
> Parts that rely on actual sequence of events (or sequence of
> processing, e.g. having understood X only after hearing Y) would
> obviously be a bit more difficult.
See (*6 below)
4. Jefferson Wilson replied:
particularly the translations of The One Ring and Babel text verse 6.
>
> > And what do *you* mean by "two-dimensional"?
>
> "With interpretation dependent on both horizontal and vertical
> relationships."
>
> >>>Second, a few specific points for pondering:
> >>>* how to have sememes that overlap or intertwine?
> >>
> >>Sememe?
> >
> > The equivalent of a "word", but exclusively depicting meaning (not sound
/ etc).
>
> In the GA all symbols can carry a wide variety of meanings, with
> the specific interpretation dependent on grammatical markings.
> For example, the glyph 'dark' can mean "dark" as an adjective,
> "to prevent from seeing" as a verb, "the darkness" as a noun, or
> many other possibilities.
See (*7 below)
------------------------------
My thoughts:
(*1)
> ... Any others that y'all have thought up or philosiphized about since
> the last big NLF2DWS threads?
Philosophy ... ! Maybe the cart is before the horse? What do you want a 2D
writing system FOR? What
is the purpose of such a system? What advantages do you expect from using
two rather than one dimension? Are we just trying to create such a thing
because, like a clever programmer, we can? Or is there a real need for it?
To determine possible uses of a 2D system, let's analyse the strengths and
weaknesses of a 1D system; maybe an extra dimension (or seven) will help
remedy those defects.
Some analysis: A linear writing system allows us to set sensible (usually
visible, but sometimes tangible, as in eg Braille) symbols in sequence.
Those symbols may represent words or thoughts. The word representation may
be phonic or ideographic. If phonic, it may be phonetic, phonemic or just a
phonemic skeleton (typically consonantal), and the sounds depicted may be
current, archaic, or a mixture of both. An ideographic representation may
support one or many different phonic realisations, eg Chinese characters and
the many spoken languages of East Asia that have used it. If the system
records words, the sequence ofthe symbols usually corresponds to that of a
potential or actual spoken utterance.
Strengths: A 1D writing system can -
1. represent a sequence of words EITHER phonically OR ideographically.
2. if phonic, provide enough clues to enable a reader to correctly
pronounce an entirely new word.
3. if ideographic, represent as many different ideas as writer ingenuity
can create distinct symbols for.
4. if ideographic, effectively represent a few similarities between many
different ideas.
Weaknesses: A 1D writing system cannot -
1. represent a sequence of words BOTH phonically AND ideographically in a
clearly structured way.
2. if phonic, effectively represent different spoken dialects occurring
over space and time, the problem being most apparent when significant sound
shifts have happened.
3. if ideographic, represent many different ideas in a few strokes, hence
be written quickly.
4. if ideographic, effectively represent more than a few similarities
between many different ideas.
5. represent digressions, comments, call-outs, hot links, or footnotes with
mechanisms that automatically return the reader to the point of departure.
6. represent simultaneous speeches by two or more actors.
7. represent ideas connected to those presented, showing their degree or
kind of connection or both.
Possible purposes:
1. Remedy weakness 1.
2. Remedy weakness 2.
...
7. Remedy weakness 7.
(*2)
> Personally, I still find it very difficult to think about sufficiently
> to get an idea of what it would look like concretely; I have some
> abstract ideas of how I'd want it to work (still very much like
> Heptapod B in skeleton) but no way of even sketching an example.
Possible means:
1. Use one linear dimension to represent the flow or sequence of events
over time, as at present.
2. Use a second linear dimension to represent a category of information
about a word eg phoneme, suprasegmentals such as stress, toneme, idea,
variant readings.
3. Use the reader's direction of gaze as a means to direct the presentation
4. Use contrast to bring different aspects to the fore- or back-ground
interactively.
5. Use size to represent relevance or importance.
6. Use colour to represent different speakers.
1 is now: 1 D writing.
1 & 2 together make a 2D writing system. There are already limited uses of
this idea in eg (a) ruby text giving the phonetics of Japanese written as
ideographs; (b) tone marking in romanised Chinese as superscript numerals;
(c) tone marking in romanised Vietnamese as diacritics; (d) secondary
articulations using superscript letter glyphs in phonetic & phonemic
transcriptions; and (e) stress patterns using ' and , in phonetic & phonemic
transcriptions. But why not let a complete record of a single utterance
take up a whole page, with the space divided into horizontal bands, one for
each category of interest? We no longer have to try so hard to save paper,
now that we have electronic ink. In fact the usual computer screen
proportions, being those of a landscape rather than a portrait picture,
would possibly be nearly ideal for presenting (say) seven categories of
information about an utterance
in each the bottom and top half of the screen:
. ideograph_1 . ideograph_2 . ideograph_3 . ideograph_4 .
ideograph_5 . ideograph_6 . ideograph_7
. phone phone . phone phone . phone phone phone . phone phone . phone
phone . phone phone . phone phone phone
. tone . tone . tone tone .
tone . tone . tone . tone tone
. stress . stress . stress .
stress . stress . stress . stress
3 is used in place of a mouse pointer in communication software for the
disabled, and (I believe) to augment the abilities of the able eg in
military aircraft.
4 is used by readily available "cognitive map" software for brainstorming,
the name of which escapes me right now.
5 is used now to emphasise titles, headings and advertised brand names.
It's also used in the meta-search engine kartOO (www.kartoo.com) to show
relevance to search criteria on interactive cognitive maps.
6 is a neat way to contrast conflicting speakers or trains of thought that
may happen at once.
(*3)
> Aren't conversations linear by definition?
Not in my house or extended family!
(*4)
> > > * how to have sememes that overlap or intertwine?
> >
> > Sememe?
>
> The equivalent of a "word", but exclusively depicting meaning (not sound /
etc).
So a sememe could be a purely ideographic, linear text. If you want to show
other aspects of meaning, the cognitive map might be the way to go, with
each word surrounded by a "cloud" of similar or related words linked by
lines.
Overlapping meanings are shown effectively by cognitive maps such as kartOO,
using phonemic representations of words, phrases to stand for meanings. The
only real requirement for these maps to be information-dense is that they
use compact representations of meaning; in this regard, a Chinese ideograph
is probably more efficient than a Latin character-set phonemic
representation.
(*5)
> But I would say that writing the *essence* of a conversation - i.e.
> the interplay of ideas - need not necessarily be linear.
Again, a cognitive map seems a good way to go for representing meaningful
content.
> (At least,
> for conversations that have actual content.)
I do hope you're not ignoring that favourite buzz-word of media studies and
artistic criticism: "sub-text"? :-)
(*6)
> Parts that rely on actual sequence of events (or sequence of
> processing, e.g. having understood X only after hearing Y) would
> obviously be a bit more difficult.
Yes ... supposing you're tracking someone's eye movements, or simply when
they've "turned the page", metaphorically speaking. You might decide to
delay presenting some information until after that point was passed.
(*7)
> In the GA all symbols can carry a wide variety of meanings, with
> the specific interpretation dependent on grammatical markings.
> For example, the glyph 'dark' can mean "dark" as an adjective,
> "to prevent from seeing" as a verb, "the darkness" as a noun, or
> many other possibilities.
Jeff, have you thought of responding to the recent thread on
"oligosynthetic" languages? It seems that "the Glyphica Arcana" must be
representing some such language at root.
Regards,
Yahya
--
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.375 / Virus Database: 267.14.21/236 - Release Date: 20/1/06
Reply