I didn't know what to post about in the UL blog. But I have thought of a topic I've been working on and this seems a pretty good place to post it.

I have been working on an alphabet for my dialect, Andalusian, based on the Arabic script, but not with purely Arabic letters because there are a lot of sounds we have and MSA doesn't.

With Latin letters it's pretty common to re-establish their phonemic value, but with Arabic ones it's not so common, hence I have tried to keep actual values of the letters.

First I will expose the phonemes of Andalusian. Each row are the different sounds for the same phoneme (I mean, there is no phonemic difference between the two sounds of the same row).

VOWELS
[a]
[e]
[i]
[o]
[u]

CONSONANTS
[p]
[b] [β]
[t]
[d] [ð]
[k]
[g] [ɣ]
[m]
[n] [ŋ]
[ɲ]
[r]
[ɾ]
[f]
[θ] [s]
[ʃ] [tʃ]
[ʝ]
[h] [x]
[l]
[ʎ]

I think I'm not forgetting any.
The goal is to assign an Arabic letter to each row.
ISSUES FOUND:
- aspiration because of a former "s" makes some sounds mutate so that there is some phonetic difference. Should this mutation be reflexed 'as is' or should the aspiration be represented by an 'aspiration marker' wherever it appears?
- should sounds that are merged in a big part of the country be represented by the same letter? or should the differentiating dialects be respected, though minoritary.

I will have to think about that.
___________________
The first solution I've given:

VOWELS
ا [a] like in Arabic.
ێ [e] from Kurdish.
ي [i] from Arabic.
ۆ [o] from Kurdish.
و [u] from Arabic.

CONSONANTS
پ [p] from Farsi.
ب [b] [β] from Arabic.
ت [t] from Arabic.
د [d] [ð] from Arabic.
ك [k] from Arabic.
گ [g] [ɣ] from Ottoman and Farsi.
م [m] from Arabic.
ن [n] [ŋ] from Arabic.
ڽ [ɲ] from Jawi.
رّ[r] I'm not sure about this. I think I should come up with a different letter, instead of using a tashdid and the letter for [ɾ].
ر [ɾ] from Arabic.
ف [f] from Arabic.
ث [θ] [s] since both sounds are the same phoneme, I've decided to use this letter because it was supposed to mean [θ] in MSA but is pronounced [s] today, AFAIK.
ش [ʃ] [tʃ] from Arabic.
ژ [ʝ] this letter represents [ʒ] in Urdu and Kurdish, and that's a similar sound, we don't have [ʒ].
ه [h] [x] from Arabic.
ل [l] from Arabic.
ڵ [ʎ] supposedly represents [lˁ] in Kurdish, and this one is also related to [l].
ٔ (hamza) would be the aspiration marker. It's a glottal stop in Arabic, so it has little to do with our aspiration anyway.