On Compact (esp.Kifish) languages and writing

>From: Jo Jaquinta <jaymin@maths.tcd.ie>
>Subject: Cherryhlist/catching up
>Date: Fri, 3 Sep 93 13:12:16 BST

(...)

Does anyone speak Kiffish?
The Traveller Role Playing Games has fairly nifty language generation tables in it which syllibacally construct words. Delighted with this construct I wrote a program several years ago that would take a sample text and perform structural analysis on it eventually producing a traveller like table. I used it to produce tables for English, Irish, German, Greek, Latin, Indo-European, Elvish, Maori and Hawaiian.
I have recently found this program again. If anyone would like to type in all the examples of Kiffish (even repeated phrases, that helps the frequency distribution) and Kiffish ship names I can run this on it and produce a chart for people to construct kif-like words.
I have discussed Kifish grammer with Lesley. She's done some linguistic courses and maybe she'll post about our conclusions. Personally I've always though it a bit like Akkadian. Am I correct in remembering that the script is largely dots and triangles?
The other day I upoaded a TrueType font for Klingon and installed it in my Windows system. Seeing all the control panels and so forth come up in that script was fun. I'd love to do a Kiffish font :-)

(...)

Jo

view full message

>Date: Tue, 07 Sep 93 01:44:01 -0400
>From: "Nancy Silberstein" <silbersteinn@a1.mscf.upenn.edu>
>Subject: Cherryhlist\Pride

(...)

Jo, I will endeavor to list as many kif words as possible, including ship's names and individual's names.

(...)

njs  6-SEP-1993 21:13

view full message

>Subject: Re: cherryhlist -- take 2
>Date: Fri, 10 Sep 93 10:35:39 CDT
>From: goldman@orac.cray.com (Goldman of Chaos -- postmaster CRI-US)

(...)

> Does anyone speak Kiffish?
> 	The Traveller Role Playing Games has fairly nifty language
> generation tables (...)

I'd love to get my hands on your program and your various tables. As long as I'm at it, I'll type the Kiffish in (once I find my copies of the Hani series.)

(...)

Matt
(...)

view full message

;From: Jo Jaquinta <jaymin@maths.tcd.ie>
;Subject: Cherryhlist/Names
;Date: Mon, 4 Oct 93 8:42:09 BST

Did anyone type in the Kif vocabulary? I need it in the next day or so.

(...)

Jo

view full message

;Subject: cherryhlist
;Date: Wed, 6 Oct 1993 09:14:36 +0100 (MET)
;From: Onno Meyer <Onno.Meyer@arbi.informatik.uni-oldenburg.de>

(...)

> 	Did anyone type in the Kif vocabulary? I need it in the next
> day or so.

I have a growing list of kifish (I'm listing characters and history of the Compact like I did for Alliance/Union, too), but that is far from complete.

(...)

Onno

(...)

view full message

;From: Jo Jaquinta <jaymin@maths.tcd.ie>
;Subject: Cherryhlist/Compact Language Generators
;Date: Tue, 12 Oct 93 8:47:16 BST

I've typed in two books worth of compact languages (Chanur's Legacy, and Chanur's Homecoming) and ran my digester on them. I had to hand-process the kif. My program doesn't handle more than two letter consonents. I've now got reasonably good tables for the Kif, Hani, Mahen, and Stsho. I tried t'ca but it broke my program :-(. Here are some samples:

KIF

sukkaok sstaktikk' khiskkokkit kakkif skkanou ukifkakk titiap't sga'kt utk hasit akkhtisakkeg shasa sfokokkhtakt 'kk sa fur hu nokkmikkan skkukosik fikko k'ktmukkpumit akakt aktnakk kakkir foakkutut sak kota totir kukt aktef nasamaktinkt riktanthiskki saghnikkhti taktsi hartut hukke khaktkikkatkk gkihtakkt ethemmsskki ho o ikkifakkt skkatmesikkt

MAHEN

shin josaijas tan mijir mihi shodanoshi awesa fake e a sha muye sho miphasinar ke piran tuhmime herkis shii mor haa me shahishedi sides mimai kau ja ela jajkesil tatemais ikiphin ji mi tushaisusab a mi nenine tas naseahar mer kariluja mi ni me ma motte retas shus a mut nema amirsa ghifit dishushane shoaidit mashina kunon pre a kaimos mira jahir hedosho ite

HANI

tarun fonan nauhnar ruhas heanyarnhyur chauhn rhoru audris mursar i hea vras charena tao harnau un hamhus arin unhar kahar buniunllun ararkur karuau hanur marhir fulus meni ufyas ranan mansu rha ajir ou sira llunan han fify dia chesmarin asosy tara vanau gfilfun hihnur tusyo hun cgu aparn salun pauhin au garaul ukir fi aura mamur a kara fasa far per rana hanyas furupar shau

STSHO

tin lliste llae o lyilsis mnikfi tlidleshitli lomnis dlinli mni a tlai nost mhi nlaist dlesti mhini stoishan onim i hislyis tlii nos e anli jast i gtotha i dlishelya lle oohon llellas lyeslsim omies ylin enlini itle esttis ee a mni mnejein tlas shastiajas is kfistshasttlol o rlamhi e mna mnost omhe ylesitholi nlost nsimne ais stesti isti tla aitlehi lyis asmnes illellejain

What is really wierd is if you pipe it through a speech synthesiser! Ever hear kiffish in an american accent?

Jo

view full message

;Date: Tue, 12 Oct 93 19:01:59 MEZ
;From: barnard@forge.franken.de
;Subject: cherryhlist

> What is really wierd is if you pipe it through a speech synthesiser! Ever
> hear kiffish in an american accent?

Grin. I tried this, but my Amiga refused to speak Kif; it just spelled the words. Hani and Mahen sounded quite good but Stsho is really wierd :)

	VLV
		Henning
(...)

view full message

;Date: Tue, 12 Oct 93 01:36:00 PDT
;From: seth@cie-2.uoregon.edu (Master Nethacker)
;Subject: C.J. Cherryh List

>What is really weird is if you pipe it through a speech synthesiser!  Ever
>hear kifish in an american accent?

I shudder to the marrow at the notion :)

You know, one of the oddest notions I ever had (here for your amusement, since it's 1:30 AM my time, and I'm almost hallucinating :) was the general distaste the Pride's crew had to knnn singing, and the question: What would a compact species (esp the Hani) make of, say, opera? ;() Pavarotti singing "O Solo Mio," the Comandatore scene from "Don Giovanni?" ;)

(...)

Anyhow, back to Jo's linguistic program... Jo, what sort of algorithm did you use to render that? It was fairly believable, I thought! The structure seems consistent (to my early-morning cursory study) with what I remember of the text ...

... Though I have one complaint... ;) --- The all-important WORD wasn't there, even once, in the kifish!

SFIK! ;)

'hakt!

		Seth

view full message

;From: Jo Jaquinta <jaymin@maths.tcd.ie>
;Subject: Cherryhlist/Kif Writing
;Date: Wed, 13 Oct 93 17:17:38 BST

THE HISTORY AND DEVELOPMENT OF KIF WRITING

The earlies forms of Kifish communication existed as territorial markings scratched on trees and other surfaces as a warning to others not to tresspass. Thus the original glyphs were representative signs for individuals, names if you would. As their social unit evolved so did the complexity of the signs. Not only did they simply have to denote what belonged to whom but who belongs to whom and what social position they maintained. Thus the original pictograms developed into ideograms that could express more complex issues.
The next leap forward in Kifish writing came with the development from hunting to hearding. With a, relatively, more stationary base fortifications developed to protect territory and with it the ability to work clay and mortar. Having previously used their claws for marking wood and soft material this translated very effectively to writing on clay tablets. With this new medium an additional amount of variety was made possible, namely dots in addition to strokes, and we can date the age of certain word-forms by this.
Writing seems to have been largely in the form of military instructions alternating with a complex arsenal of phrases for intimidation and suplication. The major archeological records for this period consist of courier tablets and engraved stone steles erected in praise of a regional Hakt, possibly at the entrance to the fortress/city.
As Main Kiffish developed the written language became large and unwieldly as new ideograms and logograms were introduced as that culture expanded. Similar symbols had widely divergant meanings over different regions. A certain resentment began to develop toward scribes as professionals now needed to be used for communication and translation. The warrior elite could not afford to be without some literacy as to trust their scribes implicitly would weaken their position. Their fears were justified as there are several incidents told, possibly apocryphal, of warrior-scribes taking the leadership and the ensuing chaos that came of it.
What developed from this was a more widespread use of a technique originally used to record foriegn names. As these names were quite diverse and it was impractical to represent them with unique logograms they were represented as concatenated logograms of Main Kiffish words of single syllables. Thus the name Tikkakkhun was represented by the three logograms for the Main Kiffish words tikk, akkh, and the modifier un.
If the scribe-warrior tales are fictional they can be seen as stories promoting the use of the syllogramatic writing base on a wider scale. In any event it is this form that eventually won through.
The basic syllograms were established at that time and, with a few simplifications and changes of pronunciation, are widely used today for formal Kifish documentation. The latest development in Kifish writing is realtively recent and has to do with its adaption to the digital computer and pixel orientated display.
In their early days of computing the space and processing power of their devices and the resolution of their monitors was low. Too low to adequately express the large number of syllograms in common usage. Originally the basic syllograms, which by linguistic evolution were also the most common, were represented. This saw them through their initial experimental stage but as the use of computers became more widespread it proved insufficient.
In a curious parallel to previous development these most common glyphs were used to represent individual sounds. As this developed into an alphabetic system the syllograms in turn became composed of these individual letter-glyphs. Although that was a transitory stage in their technological development the alphabetic system is still largely used for data entry and storage.
Moving to the letter-glyphs themselves it can be redily seen that they display the general left-to-right and top-to-bottom nature of Main Kifish writing. Horizontal strokes always point right and vertial one always point down. Diagonal strokes usually draw from upper left to lower right but a few go from upper right to lower left.
The least sophisticated data entry and display systems currently in use comprise of a keyboard, with claw-depression keys, consisting of the alphabetic glyps (see below). The display device displays the glyphs as they are entered.
More advanced systems translate the glyphs into syllograms on-screen as they are typed. Thus typing "t" would produce that alphabetic glyph on the screen but when followed by the "ikk" keystroke it would coalesce into the "tikk" syllogram. These keyboards usually have two "space bars". One for explicitly signifying syllable breaks (as with all natural languages there are certain ambiguities that defy computer algorithyms) and another for explicitly speficying word breaks.
More advanced systems consist of a data entry pad where the operator can literally claw out the letter or syllogram. On some a small area is reserved for tapping on to indicate that the character is done. On others it is purely done by time delay. On the most advanced systems the keyboard is both an input and output device changing to meet the data-entry requirements of the software being run.
Here now follows the most basic representation of the Kiffish alphabetical system:

A		AK		AKK		SG
..@@@@@@@@@@..  ..@@@@@@@@@@..  ..@@@@@@@@@@..  ..@@@@@@@@@@..
....@@@@@@....  ....@@@@@@....  ....@@@@@@....  ....@@@@@@....
......@@......  ......@@......  @@....@@......  ......@@......
......@@......  ..@@..@@......  @@@@@@@@@@@@@@  @@@@@@@@......
......@@......  ..@@@@@@@@@@..  @@....@@......  ..@@..@@......
......@@......  ..@@..@@......  @@@@@@@@@@@@@@  ..@@..@@......
......@@......  ......@@......  @@....@@......  ..@@..@@......

[For anyone who is interested in the rest of the alphabet it can be found on the archive in the pub/jaymin/cherryh/sup directory called kifwri.txt. There is also a kif8x8.fnt and kif8x16.fnt font files for the Microsoft Software Development Kit for Windows]