i18n: intl-pipeline translations (all languages)

[wackerow]

Automated Translations

This PR contains translations managed by the intl pipeline.
Each run appends a summary below.

---

Run: 2026-04-18 21:01:13 UTC

• Languages: ar, bn, cs, de, es, fr, hi, id, it, ja, ko, mr, pl, pt-br, ru, sw, ta, te, tr, uk, ur, vi, zh-tw, zh
• Files: 46 (0 MD, 46 JSON)
• Mode: auto
• View workflow run

[netlify]

✅ Deploy Preview for ethereumorg ready!

Name	Link
🔨 Latest commit	8c207e5
🔍 Latest deploy log	https://app.netlify.com/projects/ethereumorg/deploys/69e3fac0d77c540008ea9a92
😎 Deploy Preview	https://deploy-preview-17992.ethereum.it
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.
Lighthouse	7 paths audited Performance: 56 (🔴 down 3 from production) Accessibility: 93 (🔴 down 1 from production) Best Practices: 100 (no change from production) SEO: 98 (🔴 down 1 from production) PWA: 59 (no change from production) View the detailed breakdown and full score reports

---

To edit notification comments on pull requests, go to your Netlify project configuration.

[myelinated-wackerow]

Translation Quality Review

PR: #17992
Languages: 24 (all supported locales)
Files reviewed: 46 JSON
Glossary source: ETHGlossary v1 (507 terms/lang)

Language	Quality Score	Critical	Key Warnings
zh	9.5	0	Minor "只是" editorial addition in ownership description
es	9.5	0	NFTs/los NFT pluralization waver; Bienvenido a Web3 gendered vs rest of file
pt-br	9.5	0	Pronoun concordance (Eles recebem for singular antecedent) at L46
uk	9.4	0	маркетплейс vs NFT-маркет split within sim-cw block
ar	9.3	0	Modal label collision: modal-institutions = المؤسسات vs modal-enterprise = للمؤسسات
de	9.3	0	Werte vs Vermögenswerte inconsistency for "assets"
pl	9.3	0	ICU plural concerns; masculine-default past-tense verbs
cs	9.2	0	Pluralization risks for numeric ICU values (days/minutes/years)
it	9.2	0	Benvenuti vs Benvenuto register split; azienda agricola aziendale tautology
ru	9.2	0	NFT-маркет/Маркетплейс NFT split; кликайте/нажмите inconsistency
zh-tw	9.2	0	Pronoun 你/您 switches mid-file at L102
hi	9.2	0	स्टेबलकॉइन्स plural vs glossary स्टेबलकॉइन; उद्यमों/एंटरप्राइज़ split
vi	9.1	0	Mid-sentence Title Case on Ví/Tài khoản/Chuỗi khối in sim-cw block
sw	9.0	1	sarafu dhabiti vs glossary sarafu thabiti
te	9.0	0	L161 ముద్రించు register break; L101/L102 verb mismatch for "received"
bn	9.0	0	stablecoins left Latin where glossary prefers স্টেবলকয়েন
fr	8.8	0	NFT/NFTs pluralization inconsistency within sim-wallet file
id	8.8	0	DeFi expanded to keuangan terdesentralisasi (DeFi) where EN has bare brand
tr	8.8	0	sabit coin vs glossary sabitcoin; mid-sentence capitalization of Cüzdan/Hesap/Blokzincir
ur	8.7	1	Arabic kāf (U+0643) in لامركزی at page-index:3 — cross-script contamination
ja	8.4	0	送信/送金 split for "send"; fullwidth/halfwidth paren mix in toast strings
ta	8.3	0	token transliterated (டோக்கன்) instead of glossary calque வில்லை across ~10 strings
mr	8.3	0	रिकव्हरी फ्रेज raw English transliteration × 7 strings; assets inconsistent
ko	8.2	0	귀하 (business-honorific) vs 여러분 mid-file register split

Critical Issues (2 total)

Both were independently verified with Gemini and will be addressed via a follow-up patch on this branch:

1. ur — src/intl/ur/page-index.json:3 uses the Arabic kāf (U+0643) in لامركزی where Urdu kāf (U+06A9) is required. Genuine cross-script contamination. Patch replaces with glossary-preferred غیر مرکزی (which uses the correct Urdu kāf).

2. sw — src/intl/sw/page-index.json:46, 98 uses sarafu dhabiti (Swahili dhabiti = firm/strong) where the glossary specifies sarafu thabiti (thabiti = stable/steady). Patch swaps to the glossary form.

Observations Across All 22 Substantively-Reviewed Files

Structural integrity is uniformly perfect. Every ICU placeholder, HTML-like tag (<strong>, <em>, <nft>, <web3>, <stablecoinsLink>), emoji, and ticker is preserved exactly. No JSON parse issues, no mangled placeholders across the PR.

"Recovery phrase" vs glossary "seed phrase" tension. ETHGlossary v1 has only seed phrase; English source uses "recovery phrase". Languages diverge — some translate literally (cs, pt-br, mr, ur), others adapt the seed-phrase glossary entry. Likely a glossary coverage gap rather than a per-PR fix target.

Parenthetical brand expansion. Several languages (hi, id, it, ja, ko, vi, de) expand bare DeFi → <native term> (DeFi) where English source is terse. Matches glossary canonical form; deviates from source fidelity. Harmless pattern.

Non-Latin plural morphology. hi, mr, bn add English -s to transliterated nouns (स्टेबलकॉइन्स, টোকেন্স etc.). Warrants a glossary policy decision.

Mid-file register shifts. ja, ko, zh-tw, vi, tr all show register inconsistency within a single file (formal/informal swaps, or mid-sentence capitalization drift). Suggests the pipeline isn't locking register per-language.

Verdict: Approvable once the two critical patches land. All 46 files pass structural, placeholder, and tag checks cleanly.

---

Reviewed by Claude Code

[myelinated-wackerow]

Approved

The two critical issues flagged in the earlier review have been addressed in 8c207e5:

• ur: src/intl/ur/page-index.json:3 — لامركزی replaced with glossary-preferred غیر مرکزی (fixes Arabic kaf U+0643 cross-script contamination)
• sw: src/intl/sw/page-index.json:46, 98 — sarafu dhabiti replaced with glossary-canonical sarafu thabiti

Post-fix verification:

• Urdu kaf audit on ur/page-index.json: 0 Arabic kaf (U+0643), 264 Urdu kaf (U+06A9)
• No other dhabiti remains in files touched by this PR

Approving. Remaining items in the earlier scoring table are polish-level and non-blocking.

---

Reviewed by Claude Code