Should an Arabic compound given name keep the article that completes it? (Nour El Houda Haddou gives family 'El Houda Haddou') #1378

Since #1257 el is read by default as al is, an article that may be a given name. Algeria is the one population measured where that breaks more names than it fixes: ORCID's Algerian researchers 37 fixed to 135 broken, PSC's Algerian filers 3 to 10 (config_diff.py score, 2026-11-12, against Morocco's 2,002 to 26, Egypt's 286 to 20 and Lebanon's 432 to 6). Every Algerian break is the article inside a compound given name, which the particle join takes into the family. #1257 recorded this as an accepted cost under rules/P2; this issue would lift it. The same shape costs Hebrew-script אל about 6 JRC-Names spellings, and Latin abd before El:

Proposal (vocabulary only, no new rule)

1. The second halves as rules/P10 words (after_particle on their given readings): houda, hoda, amine, islam, islem, din, dine, zahraa, zahra, batoul, imen, bachir, habib…
   • When: a one-word given name, the article, a listed word, then a family name.
   • Then: the article and the word are a middle name (as María del Carmen García reads); two-given-name handling (rules/O7) takes it into the given name.
   • Because: the word after the article is a given name's second half.
   • Accepted: a family name opening with the same word reads as the given name's half, the cost P10 already accepts for Carmen.
   • Open: P10's threshold (75% of moved names landing in the filers' boxes, ten or more) per word, pooled or by country; on by default or under an Arabic choice only.
2. abd, abdel and Hebrew-script עבד with rules/P12's completed_by for the names after the article (El Kader, El Aziz; אל עזיז, אל חלים, אל כרים, אללה), joined as one given name.
   • Open: which names; whether Latin abd takes the same list as Arabic عبد.
3. Arabic-script ايت and ولد (Should Ait, Nait and Ould be family-name particles? (Youssef Ait Taleb gives middle 'Ait', family 'Taleb') #1249's last question): the spellings of ait and ould, 5 and 4 names among 394 + 117 Moroccan parliamentarians. Is ben/bin/בן a particle, a patronymic marker, or a given name? (Karim Ben Salah, בן כספית, Youssef Ait Taleb) #1257's rule 1 would decide them; ولد is also the noun "boy" and needs its bearer check.

Questions

• Does (1) lift Algeria enough to drop the Accepted line from rules/P2?
• Is it on by default or under an Arabic choice only?