When will we support Rongo-Rongo?
by Michael S. Kaplan, published on 2006/07/27 09:47 -04:00, original URI: http://blogs.msdn.com/b/michkap/archive/2006/07/27/680232.aspx
A few years back, John McConnell gave a day 2 keynote at the 26th Internationalization an Unicode Conference, entitled The Windows Language Roadmap or When Do We Get Rongo-Rongo?.
The subtitle, in a bold tradition that was subsequently taken up by this very blog you are now reading, had little to do with the actual presentation, but provided an interesting title and a fun story that cannot be found in the slides (leaving people who did not attend the talk wondering what it was all about, just as with the moose at the end of the presentation!).
(He did give a slightly longer version of the talk at the 2004 Global Development and Deployment Conference, where the advantage of a video version of the presentation online exists for your enjoyment. :-)
Anyway, for your reading pleasure I will (with John's permission) provide the transcript of the story below, but it is right at the beginning of the video and definitely worth listening to in John's unique storytelling style if you have the time (since I did not include a laughtrack it's the only way you can find out where the crowd was amused!).
I've had several people ask me about the title of this talk "When do we get to Rongo-Rongo?". Some people thought I made up the name. I'll explain, it has a little bit of a personal history.
One of the very first projects I had when I was still a developer involved in globalization was back in the mid-80s. It was for a very large customer whose name I can't mention, but they're in Langley, Virginia.
The assignment I had was to support bidirectional text; technically the documents supported left-to-right, it did not support bidirectional. So I understood and worked with people who understood bidirectional text and I was able to work that out.
But being the ambitious little nerd that I was, I went off to a library and I decided I would find out more about writing systems. Because I knew vaguely that East Asian text was written vertically and I thought, 'well maybe I should generalize my code so I can support vertical writing.'
So the library was a wonderful resource. I found out about ancient Greek writing, which (I'll probably say this wrong) Boustrophedon, where they would write one line going one way and then the next line would start there and go back. And that was very appealing to me.
But then even better was Rongo-Rongo, which it sounds like it's made up by teenagers or something, but it was a language used on Easter Island, or I shouldn't say language, a writing system on Easter Island.
I believe there's only like 120, some small number of samples. They are on these large round disks. It has never been fully deciphered.
But the thing that was really wonderful about it is it's written sort of like Boustrophedon, but when you get to the second line, rather than just going backwards, it actually turns upside down.
So this really put me into a fever, writing the code.
So, unfortunately in that particular coding assignment I ultimately concluded that I couldn't support Rongo-Rongo -- the performace hit was just a little too great.
And so, when I delivered the software to the salespeople they said "What languages does it support?" and I said "It'll support anything except Rongo-Rongo."
I said this as sort of a joke, but about a month later we had the version two requirements, which said that "Version two must support Rongo-Rongo."
So ever since that experience it's been the goal at the end of the rainbow, it's where we will eventually get to before I retire....
The full presentation talks about ELKs and LIPs and lots off the other things I talk about here, and is worth a listen, in my opinion. :-)
So here is a quick and dirty Q&A:
Q: What company was John working for back in mid 80's?
A: He was working for DEC at the time, though the contract was for that customer in Langley, Virginia.
Q: Does Unicode support Rongo-Rongo?
A: Rongo-Rongo is not yet encoded in Unicode.
Q: Does Vista support it?
A: The first step that Windows requires when it comes to language support is support within Unicode (after that we can get into fonts and shaping engines and such), so given the answer to the first question, the answer to this one would also be no.
Q: Will Microsoft ever support Rongo-Rongo?
A: It is worth noting that John has not retired yet, so who knows what the future holds? It is still at the end of the rainbow....
This post brought to you by ༃ (U+0f03, a.k.a. TIBETAN MARK GTER YIG MGO -UM GTER TSHEG MA)
# Maurits [MSFT] on 27 Jul 2006 12:57 PM:
The alternate-line-directions scheme makes a lot of sense to me. Why should your eye have to do a full carriage return across the width of the text? It's all too easy to skip lines, or read the same line twice.
I wonder if that's why newspapers use columns, to keep the width down?
# Michael S. Kaplan on 27 Jul 2006 4:02 PM:
Well, that is an argument for the Boustrophedon scheme, though. The Rongo-Rongo one requires you to read upside down or move your whole body 180 degrees! :-)
# evilgwyn on 27 Jul 2006 4:54 PM:
That's why it was printed on the large discs.
This should revolutionise (oops) the printing industry.
I'll stop now.
# Maurits [MSFT] on 27 Jul 2006 5:53 PM:
The end-of-line transformation is jarring either way. That could be eliminated by writing in outgoing spirals, starting at the center, and writing out to the circumference of a circular page.
That would require some serious surgery to the windowing architecture - circles instead of rectangles. But once that's done, reading would be much easier on the eyes.
The eye may be able to adjust to reading progressively rotated text (think Roman numerals on a clock,) or for certain media it might make sense to rotate the media at a user-configurable variable speed.
# Maurits [MSFT] on 27 Jul 2006 5:58 PM:
Like this thing
... but with Evans' original idea of reading from the inside out.
There's a singularity at the center, but that's a good place to put a title or a page number.
# Dean Harding on 27 Jul 2006 8:02 PM:
Wouldn't be easier, since we're using dynamic computer displays here, rather than static pieces of paper, to just have the text on a line and automatically scroll it along, fading things out to the left and fading them in from the right (or the other way around if you prefer)?
Even better if we could get the computer to automatically sense how fast we're reading it (which should be possible to do by looking at the reader's eye movements) and adjust to the reader's speed?
You would never have to worry about windows with only a limited size again! Well, except when you go to print your document...
# Maurits [MSFT] on 27 Jul 2006 10:12 PM:
Like a news ticker? Sure. The only drawback is the drawback of all linear media - it dramatically increases seek time. You can easily flip to a particular chapter in a book or a DVD, but it's much slower to do so with a video cassette, even if you know the exact location from an index.
I think I recall an Asimov story that dealt with a future society that was still using linear media, and during the story the concept of a multi-page "book" was reinvented. Can't remember the title, though.
# Andrew West on 28 Jul 2006 4:31 AM:
"The end-of-line transformation is jarring either way. That could be eliminated by writing in outgoing spirals, starting at the center, and writing out to the circumference of a circular page."
As in the Phaistos Disk "script". Now that Phaistos is going to be encoded, Windows will presumably soon be supporting spiral text layout as well as bidi and vertical ?
Incidentally, I think that U+0F03 is a great choice for this post ... one of the head marks used in secret "discovered" (gter-ma) texts.
# Michael S. Kaplan on 28 Jul 2006 7:46 AM:
Well, notice how I said encoding in Unicode is the first requirement; there are certainly later ones. :-)
# Gia Kvashilava on 4 Aug 2006 6:12 PM:
I decipher a syllabic writing of the Phaistos Disc in full. The text is written in the ancient Aea-Colchian (the modern Mingrelian and Lazian languages), on Cretan dialect of the one. The Disc is the oldest European documented literature, a verse of a hymn to the Palasgo-Colchian Aia-Sun and Chthonian (Colchian: “Aiabzha-Neshkari”/”Nerchi”) “Magna Mater deorum Idaea” (”Dea Dia”), Great Mother Goddess Nana/Cybele (Colchian: “Nana” _ mother; Phrygian: Matar Kubileya).
Colchis was “Coli” people’s (or Gargians, Margians, the modern Mingrelian) family-country, what was unite of races of the present day western Georgia (Caucasus). It was in the ancient geography district of Asia Minor, at the eastern extremity of the Black Sea, bounded on the north by the Caucasus mountains where was chained Prometheus by Zeus, centred about the fertile valley of the Phasis River (the modern Rion). The civilizations of ancient Colchis came to bloom in the Caucasian world. In the end of the 2nd millennium BC and first centuries of the 1st millennium BC (the Late Bronze Age) was the oldest Georgian Kingdom of Kolkheti, where the Golden Fleece was sought by Jason and the Argonauts, in the 6th century BC _ 7th century AD Georgian Kingdom Egrisi known to the ancient Greeks and Romans as Lazica. The capital of Colchis was the city Aia (or Cytaea; Colchian: Kutaia; Georgian: Kutaisi; or Greek: Archaeopolis, Colchian: Jikha-goji; Geogian: Tsikhe-goji, now Nokalakevi, the village of Senaki region in Mingrelia, Georgia). Colchian language were spoken by the pre-Olympic Titans, the Sun-god Helios and the daughter of the Oceanus Perseis’ son and daughters: King Aeetes (Colchian: “Aiathi” means “Aia-Sun family”) of the Aea-Colchis and Thessallia (in the ancient Greece), Queen Pasiphae (Colchian: “Bzhaiphe” means “she who shines for all”), the wife of king Minos of Crete, and goddess Circe of the mythical island of Aiaia. Ayetes was the father of Medea (Colchian: “Madiaia”).
If you have more questions, please contact us by E-mail: email@example.com
Mathematician Gia D. Kvashilava
From Georgia, Caucasus
# Gordon Berthin on 18 Sep 2006 1:59 PM:
Re: When do We Support Rongorongo
The upside down Boustrophedron nature of Rongorongo is historically accurate. However, in my opinion it is not necessary to preserve this option when integrating the script into a font system. Rongorongo research is typically carried out by re-configuring the characters from the tablets so that they read in the direction of 'normal' script and never turn upside down.
A more important consideration by far, pertains to the observation (substantially correct) that Rongorongo is comprised of about 120 base units arranged in combinations. Epigraphers are still not in agreement as to what the 120 bases are, and, therefore, we don't yet have a workable alphabet to input.
It is critically important that we get to this point before setting up a font set. In my opinion, we not only want to be able to model the existing Rongorongo texts but we should also have the freedom to COMPOSE new prose in Rongorongo--not just regurgitate from the existing 13,000 character corpus.
In other words, we may find ourselves producing Rongorongo characters that are not known to exist, but are linguistically valid on account of following all of the rules of the old language.
Our best efforts can never produce more than a weak approximation for several reasons, among them:
1. I doubt that there was even consensus among the old Rapa Nui as to those 120 bases and the exact rules for their application.
2. Insofar as a grammar system was established on Rapa Nui, it will be millennia before we figure out all of the fine points such that our bases can integrate for a true, workable, model.
3. I think there is evidence in Rongorongo for 'animations' and morphs. Characters are shown in succession in slightly different poses or artistic renderings so as to express a concept which otherwise could not be handled by the limiited capabilities of the language. (We can get all the old animations and morphs, its true, but there will never be opportunity to create our own new ones even where they may be linguistically required).
True, an approximate system is much superior to scribing characters in Paint or AutoCAD or whatever. Yet, I say, wait a decade before going ahead. I think that there is a good possibility that we will see some strong fundamental linguistic work on this subject within the next few years and that would be an invaluable resource in preparing the Font system. The well known Barthel system is the current standard for glyph classification (see www.rongogongo.org ). Pozdniakov (Society des Oceanistes 1996) 103 (2): pp. 289-303 is, in my opinion, the best published resource showing how Barthel needs to be modified to get us to where we want to be.
# Gia D. Kvashilava on 20 Feb 2008 6:22 AM:
Copyrighted long comment by Gia D. Kvashilava removed
I have snipped the rest of this comment as this is not the place for posting it. Copyrighted works where I am not a/the copyright holder? Not the place, certainly not in long comments.
Please post it elsewhere and then if you want to explain why you believe it relevant and then post a link.
Sorry, but entirely too many people have been abusing this blog as of late....
Jacques B.M. Guy on 23 Jun 2008 1:10 AM:
I discovered the... ahem! ... "seminal" article by Berthin & Berthin yesterday. My immediate thought was "this needs debunking." Then, on second thought: "but wouldn't that bring that article publicity?"
So I am in two minds. To debunk or not to debunk?
As for the gentleman here who has deciphered the Phaistos disc, sorry, mate, John Chadwick beat you to it long ago. And his decipherment demonstrates that the Phaistos disc is undecipherable, as anyone who knows anything about decipherment should know.
Jacques B.M. Guy on 23 Jun 2008 1:17 AM:
As for the Phaistos disk, the direction of writing was independently invented by a friend of mine 40 years ago (Perry Cook was the name, if memory serves). He wrote me a letter in spiral, outside in. In the centre he had punched a hole, with "continue here." On the other side the letter continued, inside out this time. As a post-scriptum: "I was pissed out of my head when I wrote this."
Food for would-be decipherers' thought.
Michael S. Kaplan on 23 Jun 2008 1:52 AM:
Popping the stack a bit to the comment from Maurits, in keeping with the most recent directions in the thread:
There's a singularity at the center, but that's a good place to put a title or a page number.
That is also a nice place to set one's beer. :-)
go to newer or older post, or back to index or month or day