Transliteration of ì & ä

Started by Sìkat, January 22, 2012, 03:30:46 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

Sìkat

I've been going through the lessons in TEWTLNE and had a question, apologies in advance if this is the wrong place for it and/or has been previously asked & answered: how would one represent ì & ä on a device that can't show those characters?

Is there an "official" or commonly used transliteration of these two letters, like ss/sz for ß (German hard S)?

Tirea Aean

On SMS which normally only supports ASCII, I use

I* i*

And

A* a*

With everyone I txt nìNa'vi

Ftiafpi

The convention I've always used is to just substitute the normal i and a and let it be defined by context.

Tirea Aean

That to me is less precise. But as I said, the community has not considered a globl notation for this which escapes utf8 characters.. We just made ä and ì the standard here.

Shall we have the community agree on a ASCII alternative via poll?

Sìkat

#4
A community poll would definitely be a good idea.  My suggestion: ä = ah, ì = ih, based on the alphabet in MIPP's Na'vi for Beginners, but I'm wondering if those might present a collision problem...

Ricardo


Ftiafpi

Quote from: Tirea Aean on February 02, 2012, 04:37:54 PM
That to me is less precise. But as I said, the community has not considered a globl notation for this which escapes utf8 characters.. We just made ä and ì the standard here.

Shall we have the community agree on a ASCII alternative via poll?
A poll would be a good idea.


Ningey

Quote from: Ricardo on February 02, 2012, 06:08:49 PM
ä --> ae
That's the German way.

That's true, but it inevitably would invite collisions nìNa'vi. Please keep in mind that ae is a valid sequence of syllables there so things would become ambiguous quickly.
I, for my part, would consider i* and a* to be the best option here (although I seem to remember that ext ASCII - that is, ASCII codes 128 to 255 - contains both ä and ì).


"Sawtute ke tsun nivume - fo ke kerame!"
-- Neytiri te Tskaha Mo'at'ite

"There are two things that are infinite: Human stupidity and the universe. However, I'm not yet sure about the universe."
-- Albert Einstein

"He who gives up freedom for security deserves neither and loses both."
-- Benjamin Franklin

Sìkat

Extended ASCII does contain ì and ä, but they're not in the same location depending on the code page.  For CP437 (MS-DOS), ì is at 141 (0x8D) & ä is at 132 (0x84), but the same symbols are at 236 (0xEC) and 228 (0xE4) in Unicode itself and Windows-1232.


Ftiafpi

Right but we would still run into trouble on a phone using SMS.

Tirea Aean

Quote from: Ftiafpi on February 03, 2012, 11:07:30 AM
Right but we would still run into trouble on a phone using SMS.

Which is why my vote is on a* i*

Ftiafpi

Quote from: Tirea Aean on February 04, 2012, 08:45:14 AM
Quote from: Ftiafpi on February 03, 2012, 11:07:30 AM
Right but we would still run into trouble on a phone using SMS.

Which is why my vote is on a* i*
Works for me, easy to use and obvious.

Kalti*, 'i*'awn, sa*r.

Sìkat

I checked the database used by the LN iPhone app.  The digraph 'ih' is not used, but 'ah' is used in 6 different words. 

I* and A* have my vote.

`Eylan Ayfalulukanä

I have been giving this idea some thought, and I will throw out a couple proposals.

The first and most radical would be to use use an x after the marked letter. So, you would have ax and ix. The same thing could be done with the letter q. While being very 'alphabetic', it might look too much like an ejective consonant.

So, a;lthough I am not real fond of the asterisk character, it seems to be one of the better solutions presented so far.

That said, TA and I have exchanged texts with the right characters showing up. At least the Droid X allows the entry of the marked vowels with a simple mechanism.

Yawey ngahu!
pamrel si ro [email protected]

'Oma Tirea

How about a: and i:?  Kaltxi: ma oeya: 'eylan...

hi:'i, ka:, va:ng... kea ekxan, kefyak?

[img]http://swokaikran.skxawng.lu/sigbar/nwotd.php?p=2b[/img]

ÌTXTSTXRR!!

Srake serar le'Ìnglìsìa lì'fyayä aylì'ut?  Nari si älofoniru rutxe!!

`Eylan Ayfalulukanä

I like the : better than the *. It is also narrower when proportional fonts are in use.

Yawey ngahu!
pamrel si ro [email protected]

Sìkat

Quote from: `Eylan Ayfalulukanä on February 04, 2012, 10:03:59 PM
The first and most radical would be to use use an x after the marked letter. So, you would have ax and ix. The same thing could be done with the letter q. While being very 'alphabetic', it might look too much like an ejective consonant.
I ran those four digraphs (aq, iq, ax, ix) through the database, and came up blank for all four.  They're definitely usable. 

Quote from: `Eylan Ayfalulukanä on February 04, 2012, 10:03:59 PM
So, a;lthough I am not real fond of the asterisk character, it seems to be one of the better solutions presented so far.

That said, TA and I have exchanged texts with the right characters showing up. At least the Droid X allows the entry of the marked vowels with a simple mechanism.
Likewise, iOS devices have a very simple mechanism for using the marked vowels regardless of the type of keyboard being used. 

I've since discovered that the basic GSM 7-bit alphabet (which is required to be supported by all handsets and networks per GSM 03.38) has entries for ä (0x7B) and ì (0x07), so as far as texting is concerned, "escaping" the marked vowels with another character shouldn't be necessary unless the phone makes it inordinately difficult to access those vowels. 

For communications in which 7-bit ASCII is the lowest common denominator (i.e. email), though, the issue still stands. 

Tirea Aean

In my experience, two andraid phones are able to read and write ä and ì correctly, and two iPhones are able. However between an android and iphone:

iPhone sends "nì'Ìnglìsì ätxäle si"
Android sees "n ' ngl s  tx le si"
Android sends the same phrase
iPhone sees (iirc) "n?'?ngl?s??tx?le si"

Or something like that.

I like colon idea except Wordfinal colons will look like punctuational colon.

Yawne Zize’ite

#18
In days of old, when Unicode wasn't yet invented, a common workaround was to use "prefix notation" or "postfix notation." That is, find the closest punctuation mark on a US keyboard to the accent you want and put it before or behind the letter, like so:

Kaltxi`! Oeya" tstxo lu Yawne. Ngeya" tut?
Fko oeru syaw T`ing Eywatik`ite'e. Nga l"apivawk ko!

I find this more readable than using colons or asterisks since the mark suggests the diacritic. I noticed that my phone doesn't have the grave accent ` and ' would surely produce collisions, so maybe i^ or i* for ì along with a" for ä.

I don't have anything against ax, ix or aq, iq, but those are much less intuitive to the uninitiated. Which looks better for "eĥoŝanĝo ĉiuĵaŭde" to you, "ehxosxangxo cxiujxaude" or "eh^os^ang^o c^iuj^aude"?
For what it's worth, online Esperantists chose the former.

Ftiafpi

Oh, interesting about Esperanto users choosing the former. After seeing your comparison I agree that the latter is definitely easier to make out, though they are both hard to instantly recognize the original word. I would definitely support i* or i^ and a", I like that idea a lot.