Keyboard Mapping
This 1999 posting gives the history of the unigraphic keyboard positons that
are still used by most of the Shavian fonts.
Here is a 1999 web page, created around the time of this posting, which
includes
clickable sounds files
http://victorian.fortunecity.com/vangogh/555/Spell/saunds-eng2.html
I had forgotten about this page. It also includes ANJeL.
The issue was making a unigraphic representation of English speech sounds fit
the available keyboard and still be recognized as English words and easy to
remember.
The desire to limit the keyboard to ASCII characters and to have enough
keyboard positions to handle over 40 sound signs led to a mixed cap notation.
John Malone pioneered this approach in 1950 with the development of Unifon.
[www.unifon.org]. Since you don't have the font installed, this graphic will
display as keyboard unifon. Display Unifon is a jpg graphic.
Unifon
one-soundDa kwik brqn foks jumpt Ovcr EK Tin dxg, lCk qt, I Sqtcd tU hur, fOr
hEz fQld yU cgen.
Keyboard Unifon, like Keyboard Shavian, was not particularly concerned with
the readability of the keyboard positions. The symbol system was readable for
consonants and for ten long and short vowels. 32 symbols were readable
because they generally matched the traditional assignments but there were 7 more
phonemes that needed to have assigned keys. These assignments were arbitrary.
Shavian compounded the problem by adding 8 more sound signs for sound
combinations.
ENgliS, with its schwa-a, is not significantly better than Unifon.
Da kwik brown fcks jvmpd Ovar EC Tin Dog,
lwk owt, I Sowtad, for hEz foild U agen.'
Cap S and C are not automatically identified with the digraphs Sh and Ch
so it takes some time to recognized unigraphic words.
Shavian would be keyed in the following way
H kwik br
If this correspondence chart does not show up, it can be found at
http://www.foolswisdom.com/~sbett/spanglish-ita-red-16.gif
--- paul vandenbrink <pvandenbrink@...> wrote:
> From: "Lionel Ghoti" <ghoti@d...>
> Date: Wed May 19, 1999 4:23 pm
> Subject: [shavian] Re: Keyboard Shavian
>
> Lionel Ghoti [pseudonym] wrote: The keyboard mapping pattern for the
Shavian
> fonts was devised, as far I know, by Ross DeMeyere
> (http://www.demeyere.com/Shavian/info.html), and was first used
> in his Shaw Rough and Shaw Gothic fonts. These were the first
> Shavian fonts to be widely used, and the first Shavian content on
> the web used these fonts. For compatibility's sake, all subsequently
created
> fLG: onts have used the same mapping pattern. When I made my Lionspaw font,
I > considered using a different mapping
> pattern, and I came up with a new pattern which I think featured a
> few improvements over the old one; but I still couldn't find a way
> to avoid having some arbitrary phoneme-grapheme correspondences, and
> I didn't feel that the few improvements the new pattern could make
> would justify making the change and causing all the biliteral
> problems that would follow.
> LG: I think your Unigraf mapping pattern is the best I've seen, and I found
that I could read your sample text very quickly. I find that its Roman text
can be read much more intuitively than the current "keyboard Shavian"; and I
think it would be much easier to learn to type using your system.
> LG: However, I fear that some of the new correspondences which appear
> intuitive to me (since I'm familiar with the International Phonetic
> Alphabet) might seem overly arbitrary (perhaps more arbitrary than
> their values in current keyboard Shavian)
> to someone who doesn't know the IPA. For example, you use an X for
> the /dh/ phoneme.
SB: [X was used because it is as close as we can get to the crossed-D <ð>,
a character in Old English, Icelandic, ITA and PMF and one that I think
should be revied.]
> I find this easy to remember because it looks like the top half of
> the IPA symbol for that phoneme (like a lower-case d, but with its
> ascender curling over to the left and crossed with a small diagaonal
> line). But someone who has never come across the IPA would probably
> find it somewhat easier to remember the current key combination, H,
> since an h is always present when the /dh/ phoneme is represented in
> the Roman alphabet.
> I think we should keep in mind that the Shavian alphabet is about
> alphabet reform, not mere spelling reform: Shavian text is meant to
> be read using its own letters, not those of the Roman alphabet.
SB: Shavians often write as if there is no way to salvage the Roman alphabet.
I think there are several ways this can be done.
ð ónly wûn ðat is not týpograficly chalènjd iz wûn ðat yúzez díagrafiks.
> The only real difficulty I have in communicating in
> Shavian is _typing_ it using my QWERTY
> keyboard, trying to remember which key corresponds to which Shavian
> letter. I put this down to lack of practice
The only sound-signs that I have trouble with are the arbitrary ones.
yY xX P D, etc. There is a logic to Read's shapes that is lacking from the
keyboard verison.
> If there didn't already exist on the Internet a fair amount of
> Shavian text using the current key-mapping, I'd readily change the
> mapping of my font to use your Unigraf system, and I'd urge all the
> other Shavian font designers to do likewise. At this point in time,
> however, as the Shavian alphabet is only just beginning to set down
> its electronic roots on the Internet, I think it would be dangerous
> to cause confusion to any new Shavian acolytes by introducing a
> second mapping pattern. If someone unwittingly tried to read some
> pattern-x text in a pattern-y font, the experience might put them
> off Shavian for life.
SB: The spirit of Shavian involves making available a parallel system of
sound signs. This would include having a parallel keyboard version as well, I
would think.
> LG: I don't think all of the Shavian content providers would agree by
> consensus to change the ASCII-coding of their Shavian text. Only a
> few months ago, it was suggested that the "huNG" and "Ha-ha" glyphs
> had mistakenly been switched in the production of the alphabet table
> of _Androcles and the Lion_, and that they should be switched back
> immediately in all of the current Shavian fonts before anyone else
> noticed. The online Shavian community rebelled, and a terrible rift
> threatened. Tears were shed, catty things were said, and a punch-up
> was avoided only by the fact that none of us knows where anyone else
> lives. The general consensus was that the alphabet should be left as
> it has been since the Sixties.
> Granted, you're not talking about changing the alphabet itself, but
> I would guess that the general consensus would be opposed to any
> sort of change which would affect the readability of the existing
> Shavian content on the web.
SB: The goal, of course, is to improve the readability of keyboard Shavian.
A standard is important but this does not preclude having options.
There should be no change in Classic Shavian or Classic Unifon
but we should experiment with possible improvements.
> ----- Original Message -----
Steve wrote:
> Sent: 12 May 1999 12:41 am
> Subject: [shavian] Keyboard Shavian
> > Where does keyboard Shavian come from? It is used on all of the
> > Shavian fonts that I am aware of.
> > I think that keyboard Shavian, the ascii text that appears on the
> screen before the font is applied, can be improved.
> > The question is whether or not the improvement is significant
> > enough to be worth the problem of chancing existing font development
> > conventions.
> > 90% of keyboard Shavian is intelligible. The other 10% appears to
> be rather arbirary phoneme-grapheme correspondences.
> > An improvement in the rationality and readability of keyboard
> Shavian is probably limited to about 5%. The R-combinations will
> always remain arbitrary. However, D and P could be replaced with ar and or.
> > I would also like to add ligatures in Shavian for syllabic LMN in
> > addition to R as in "hR litL colM cAm fRst."