Out of curiosity, why exactly was this name chosen? Simply because it's not Greek? The Persians do not particularly stand out this alpha, they didn't even get a proper wonder (unlike the Gauls).

In rP24787#48512, @Nescio wrote:

Out of curiosity, why exactly was this name chosen? Simply because it's not Greek? The Persians do not particularly stand out this alpha, they didn't even get a proper wonder (unlike the Gauls).

There is no particular reason to point out, except that the team voted in favour of it by a large margin. I had shortlisted Xšayaṛša, Xenophon and Xanthos, because those seemed the most relevant options — I think, had the Han been included, that Xiangnu would have most likely been picked.

I believe there was no particular 'civ-orientation' for this alpha, and X was rather limiting our options. I personally voted for Xšayaṛša largely because I found it more interesting and unique than the other two.

Looked around, it does seem more accurate in the English transliteration, though it's apparently not used in every language.

Given that we are in string freeze, I'll update this manually for the English language, and I'll see about the others.

I guess we should have just gone for the cuneiform and called it a day :p

While it is true different transliteration conventions exist (e.g. x, kh, ch, ḥ, or h for 𐎧; or ā, â, å, or aa for 𐎠), the Old Persian cuneiform is quite clear, the last vowel is explicitly long:
𐎧𐏁𐎹𐎠𐎼𐏁𐎠
x-š-y-a-r-š-a
Xšayaṛšā (or Xšayaạršā or Xšayāršā, depending on how you want to render the vocalic r)

There are no inscriptions where Xerxes does not end with a long a, I checked:

• Rüdiger Schmitt Die altpersischen Inschriften der Achaimeniden : Editio minor mit deutscher Übersetzung (Wiesbaden 2009) 151 ff.
• Prods Oktor Skjærvø An Introduction to Old Persian (Boston, MA 2016)
• https://www.livius.org/articles/person/xerxes-i/ s.v. sources

I'm not sure what sources you used, but vowel length matters and this is something that must be fixed.

In rP24787#48541, @Nescio wrote:

While it is true different transliteration conventions exist (e.g. x, kh, ch, ḥ, or h for 𐎧; or ā, â, å, or aa for 𐎠), the Old Persian cuneiform is quite clear, the last vowel is explicitly long:
𐎧𐏁𐎹𐎠𐎼𐏁𐎠

I agree, my point is that the transliteration of "long a" can vary between languages, and in particular I don't believe French uses it for Old Persian (I based this on the wikipedia pages for Xerxes and for "a macron"). I took the spelling from the suggestion thread (e.g. here) and I suppose my double-check took me to French wikipedia, and I incorrectly didn't assume that the English variant could differ. I should have checked better, admittedly, but it so happens that we were rushed for time since we had declared string freeze without picking the release name (an unfortunate mistake).

Anyways, the question of how this mistake was introduced is rather irrelevant, my point is that I will fix this manually after pulling the translations.

Never blindly trust Wikipedia (or a forum post).

in particular I don't believe French uses it for Old Persian

Here's a page from Pierre Lecoq Les inscriptions de la Perse achéménide (1997):

Anyways, the question of how this mistake was introduced is rather irrelevant, my point is that I will fix this manually after pulling the translations.

Good! For all languages?

In rP24787#48543, @Nescio wrote:

Good! For all languages?

Unless I have a good reason to not change it in one (the fact that these are translated at all seems rather odd to me). I'll make the diff then, I suspect you'll want to review it :)

(the fact that these are translated at all seems rather odd to me)

Keep in mind not all languages use the Latin script, e.g. Bulgarian, Serbian, Russian, Ukrainian use the Cyrillic alphabet.

In rP24787#48545, @Nescio wrote:

(the fact that these are translated at all seems rather odd to me)

Keep in mind not all languages use the Latin script, e.g. Bulgarian, Serbian, Russian, Ukrainian use the Cyrillic alphabet.

Yes, my expectation is that the translators might use specific transliterations there. It seems Arabic did in the past (from a very quick check). We'll see.

Couldn't you just use the specific name string from the hero_xerxes_i.xml template?

(Also, the "AINames" in the pers.json file are an inconsistent mess. Something for the next alpha.)

I think it won't work because translations get one complete sentence for ALPHA XXIV: Xšayaṛša if they could be split then yes. eg

setStringTags(translate("Alpha XXIV: ") + translate("Xšayaṛša"), { "font": "sans-bold-16" })

I suppose the space would get squeezed though and it would break.

Not sure how it would work on the XML end.

grep -r Xšayāršā shows you need to look in the *.public-templates-units.po files to find out how various translations handle this specific name (though I guess most leave it untranslated).

Anyway, if someone ask you how to pronounce Xšayāršā, keep in mind:

• x is pronounced as in Scottish /loch/ or German /ach/ (IPA /x/) and not as in expert (IPA /ks/)
• š is pronounced as in English ship (IPA /ʃ/)
• ā is twice as long as short a

@wraitii, another page from Pierre Lecoq Les inscriptions de la Perse achéménide (1997) with the French (generic) and transliterated (specific) names of the Achaemenids:

(There is a typo in Xerxes' name, though, the s ought to be a š.)

Might be an oversight, but I've never seen it written with it, always a "normal" s.

Might be an oversight, but I've never seen it written with it, always a "normal" s.

No, I meant the first s in the transliteration, it has Xsayāršā [sic] instead of Xšayāršā.

https://trac.wildfiregames.com/wiki/SpecificNames#OldPersian

Solved by rP24808.