I was standing at the refrigerator looking at the ketchup bottle, when suddenly I thought, "Ha! I bet the Etruscan 'K' is backwards! That would be awesome!". And I checked, and it was. And it was awesome.

Cause, with the broad pen held at the Roman angle, diagonal strokes down to the right are heavy, and diagonal strokes down to the left are light. But if the 'K' was backwards, the top stroke would be heavy, and the bottom stroke would be light and the letter would fall over (that is, be aesthetically problematic, by looking top heavy).

This is a perfect example of what I've been talking about from my lettering class: some genius of a Roman graphic designer took the Etruscan alphabet and changed it up so all the letters looked pleasing and balanced when written with a broad pen.

Just got word that a second piece of my art is hanging, framed, in someone's living room :) (I have art hanging in houses in multiple states! :p )

Admittedly, I provided both pieces, pre-framed, but both couples saw fit to hang them.

Summary:
Left the house Friday, early afternoon, took the bus to downtown, then to San Jose. Got picked up, over to sarah's house. Then a bit later over to mike and shannon's. Game till, what was it 1-ish, 2-ish, not too late as game goes. Back to sarah's. Up early, early in the morning, around 11. Meepa came over. Three of us drove up to San Francisco to kellan's. Four of us walked out to lunch, etc., walked back, drove out to see Bujold at Booksmith. Drove back to kellan's, he packed up, drove back down to sarah's (necessary stop to re-unite meepa with her car). Discussion of Vorkosigan books moved into sarah's living room for a while and additioned sarah's mother. Eventually managed to stop talking long enough to get meepa back to her car, and the three of us to sarah's car, and drive back down to santa cruz. Up even before 11 sunday morning, out to the theater to see Ratatouille, for the beginning of birthday celebrations, then me and my mother dash off after the movie up to San Jose for the first ever face to face meeting of the Gang of Five (warcraft dungeon group), then back down to Santa Cruz for dinner and ice cream.

( Bus trip up )

Game:
I can't remember much about game, it not being my game, i mostly sat in a corner and took more Knight's Tale screencaps and filled in incomplete tables (of the Sêrela script) on sedesdraconis.com that matt complained about.

( San Francisco )

( Her Ladyship )

More discussion about the Vorkosigan books the whle drive down and into the living room. Literary theory and gender-coding (Miles is classically female-coded), sociobiology, feminist utopias and Ethan of Athos. You know, the standard stuff you talk about about military space opera.

Next morning off to Ratatouille. It was good.

( Gang of Five Meeting )

( Bens )

Back back to Santa Cruz, where I'd left Sarah. Birthday dinner and presents and ice cream.

Post script, extension of the back and forth. Sarah then of course had to drive backbackback to San Jose. About the time she got there she realized she'd left her computer here in Santa Cruz. She came backbackbackback monday afternoon to retrieve it.

wheh.

Yesterday i worked on Sêrela in Illustrator for 12+ hours. after some number of hours, I finished making all the glyphs, and started putting them together into words. I decided I wanted to put together the set of 3 names that I use as the example text on the old site, but newer and better.

I start to put it together. I've got several of few of the words done when I realize that I'm grouping the objects wrong in Illustrator. I've been grouping them by word, subgrouped by glyph type: a group for the stem of the word, a group for the vowels of the word, etc. I need to be subgrouping them as syllables, each subgroup containing the stem, vowel, coda, and tone for one syllable. This is not only theoretically better, since I'm working with a featural syllabary here, it will also save much effort when I go to change the morphological endings, such as converting names to genitives (which are used in other peoples' names).

So I re-group all my words I've done for a bit, then start building more names. I'm getting towards the end when I realize, I've made my vowel glyphs too small from the start. So I go back to my glyph file, make the vowels bigger. Methodically go through the file i'm working on and delete all the vowels and replace them with newer, bigger ones, and re-position them.

The I go back to putting together the couple names. Finally I finish that file around 01:30:


(hmm, that looks kind of washed out to me. not the rich black it is in illustrator. i should poke it sometime, see if i can get a richer black.)

I haven't started working on anything today, cause I needed to prepare a demo for ASL class, now done; and then go to class. And it's too hard to stop working.

Working on the page for Sêrela now. Classical Sêrela is a wacky language with 8 vowels and 4 tones.

The vowels are
i y u
e o
ɛ ɔ
a

Which I have been transliterating from Sêrela Script as
i y u
e o
ê ô
a

'cause ê is my standard transliteration for /ɛ/

but since I'm already using diacritics for basic vowel quality, I was left with superscripts for the four tones, getting things like
thwi²sên³

And I don't think there are any other vowel like characters that come with enough diacritics. æ only has 2, and ø doesn't have any. w has enough, but I need it as a consonant. I supppose I could use v for the consonant and w for the vowel, but that would be ugly as sin. Even aside from the question of what vowel I use w for.

but I noticed I could use an 8 way diacritic distinction as such:


i believe that exact set is the only 2x4 diacritic distinction available for e and o. So, the question is, is that better than superscripts? yeah, it probably is.

oh, wait, there's also ẻ ể ơ ổ, so there's a 2x5 distinction available. I think I'd hold off on the hooks if I only need 2x4, though.

oh2, and I can get y in all 4, all 5 in fact, of the diacritics in the necessary dimension: y ỹ ỳ ý (ỷ)

Having finally finished individually uploading and describing the 170 images which appear on the Kedan page of the wiki (most are around 400 bytes), I, of course, immediately set about obsoleting the entire set by starting to re-do them illustrator.

The good news:
*I wont have to re-describe the images, I can upload them into the existing Image: page
*They look better
*It will be much easier to put them together into words, and have them look much, much better as such.

When I've got them made in Illustrator I could theoretically upload them as scalable vector images of some sort, presumably SVG. However if I did that, I would have to re-describe each one while uploading the new format. Also, taking advantage of scalability would be problematic, as mediawiki's image mark-up wants you to specify a width to scale an image, which is the most variable dimension of these images. Even scaling by height would require a bit of extra care in creating the images. So, SVGs probably aren't interesting in the near future.

had been using images for IPA. I think I should probably be able to use unicode these days and assume nearly everyone will be able to read display it, right?

No, as it turns out. The ipa unicode support is not acceptable in firefox.
ɑ
that's supposed to be a script a! script a! what part of script don't you understand? the turned script a works fine, though: ɒ

the triangular colon length glyph has got spacing issues, too. I can just use a colon for that, though I think. I'm not using ipa in page names or anything, i use own transliteration and transcription schemes for page names, which would use a macron, acute, or double letter for length before ever going into colon-type notations.

Tiebarred characters are only acceptable in default serif:
a͡i a͡ɪ ə͡i ɑ͡i
(though the "script" a there messes it up, because the glyph is messed up to begin with)
not default sans-serif (which mediawiki calls, and I'm happy for it do so):
a͡i a͡ɪ ə͡i ɑ͡i

growl.

Also, wikipedia lists SVG as its preferred format for "drawings and line art illustration", and I'm thinking it would be a really good format to do my writing systems in (though converting over would take a lot of work). However my install of mediawiki does not support .svg files. Also, I'm unclear on how well supported SVG is.

I should get all my scripts into .svg anyway, or at least some vector based format, and export to .png for now. I have many years of bad habits regarding file formats.

In Imirn Script, many glyphs are composed of vertically combined shapes. There are three basic vertical slots, the base slot, which sits just above the baseline, the one above the base slot, and the one below the base slot. You can see all three used in /pɛlʂuta:/ :


Every glyph fills the base slot. But glyphs can fill the base slot and above, or the base slot and below, which theoretically could create a confusion about whether the baseline is running below the glyph, or through the middle. In fact, due to some systematic holes in the kinds of glyphs there are, plus the rules about how primary glyphs are composed versus how vowel diacritics are attached, almost any glyph will unambiguously establish the baseline.

There are only 4 glyphs which do not establish the baseline:

(from right to left, /p(ə)/, /la:/, /ŋ(ə)/, /na:/)

These are called the "palaa" glyphs, after the word palaa (/pəla:/, "starling"), which is written with a sequence of two of the ambiguous glyphs (though, together, they disambiguate each other):


So, there are very few things that can be written without establishing the baseline. Only a piece of writing that used only /p(ə)/ and /ŋ(ə)/; or only /la:/ and /na:/ is ambiguous. This is highly unlikely, since there are few words, much less phrases, which can meet those criteria.

naa (/na:/), "from", could be misread as /ŋ(ə)/. But the latter is not a possible word of Imirn, since it is not a minimal phonological word.
Likewise, laa (/la:/), the question particle, could be misread as /p(ə)/; but again the latter is less than the minimal phonological word.
The word pang (/pəŋ/), an onomatopoeic word for a bell sound ("ding"), could be misread as /la:na:/.

((Now here's the question: is it more interesting if laanaa is a real word, so there is pair that really is ambiguous [only in isolation], or is it more interesting if only one of that pair is real, also? Help me out!))

Rock! I spelled onomatopoeic right on the first try!

When I get home, I have a few things to do. I might post illustrating the "palaa" (or "starling") phemonenon of Imirn writing, and ask for some opinions on an aspect of it.

Here's the word palaa in Imirn, the piece of the illustration which I've already done:

We went to the Ren Faire on Saturday. I'd never been before. Wandered around for a while and it was nice. The really exciting parts were two-fold. First, there were a pair of trolls there. They were beautiful. I've got pictures, I'll upload them soon and link them from here. One of the trolls also took my camera at one point, so I've got a picture of me taken by a troll, too (upside down).

Then there was a booth selling stuff from this guy's workshop based out of Berkeley. They had astrolabes! And pocket sun-dials. In short, a buncha the instruments I read about this summer, but couldn't really learn about, 'casue I didn't have any to play with. I got an aquitaine like this. It's not overwhelmingly great in the accuracy department, I could have gotten a pocket sundial of the more familiar, gnomon-style for the same price that was significantly more accurate. But I know basically how they work. And, anyway, they look like clocks, the aqutitaine doesn't, so it's much cooler. I really want an astrolabe, but they're a lot more expensive.

(At the moment, my web server seems to be down, hopefully it will be back up soon, and then the links will work better.)
I'm almost to presentation stage on a new writing system for Sedes. At first, it was an abugida evolved from a very old form of Kedan with a visual style something like Mkhedruli, but that rapidly became an older form, and a new form evolved that looked more like Burmese script. Which is an abugida itself, so it's not as much mixing of influences as I might of liked. Plus it became more like the featural syllabary style that Serela uses, so not as much of a new system added as I'd hoped. But then I realized that it was probably the link between Kedan and Serela, so it's okay. And I'll probably keep and make the older form, too, but it'll be a lot harder to do well on the computer then the new form, so maybe not soon. Oh, and this new writing system is used in Ladîîmer, for the one or two people in the world that might mean anything to.

I did glyphs for hand-written Kedan a little while back, but so far, they were only visible in the names of pages, e.g. gahijo, Ândhozanirtajaban; but no systematicly. I'd been resisting doing a systematic page, somewhy.

Then I realized that my current page for Kedan has a mixed purpose. It's partly to talk about Kedan, and partly to provide a key for transliteration and pronunciation. So. I'm starting work on splitting up those two functions and allowing each to go, now unrestrained by the other. But it's probably going to take a while, and those two pages, Kedan and Transliteration, may be dysfunctional for a while, not to mention all the pages that link to them. But, enough time waffling. Time to Forge on!