Update Romanian Translation #7366
Conversation
|
Thanks for this! I'm assuming they're part of the Crowdin but just in case, pinging the two contributors to #6456: @claudiu-cristea & @alextecplayz |
| msgstr "{0, plural, one {o zi} few {# zile} other {# de zile}}" | ||
| msgstr "{0, plural, one {# zi} few {# zile} other {# zile}}" |
There was a problem hiding this comment.
This change is wrong. The initial string was correct
|
|
||
| #: src/screens/Profile/ProfileFollowers.tsx:40 | ||
| msgid "{0, plural, one {# follower} other {# followers}}" | ||
| msgstr "" | ||
| msgstr "{0, plural, one {# urmăritor} other {# urmăritori}}" |
There was a problem hiding this comment.
| msgstr "{0, plural, one {# urmăritor} other {# urmăritori}}" | |
| msgstr "{0, plural, one {un urmăritor} few {# urmăritori} other {# de urmăritori}}" |
|
I didn't cross the whole file but the changes to strings with plurals are wrong and are not accounting for Romania 3 variants plurals |
There was a problem hiding this comment.
Overall, this looks like a full rework of the Romanian localization. I think we should start from existing strings and rules established in #6456
| msgstr "{0}l" | ||
| msgstr "{0}luni" |
There was a problem hiding this comment.
Should keep the short version
| msgstr "Pachetul de început al {displayName}" | ||
| msgstr "Pachetul de pornire al lui {displayName}" |
There was a problem hiding this comment.
It was decided in #6456 (comment) to use "pachet de început". See the glossary
|
|
||
| #: src/view/com/notifications/NotificationFeedItem.tsx:351 | ||
| msgid "{firstAuthorLink} and <0>{additionalAuthorsCount, plural, one {{formattedAuthorsCount} other} other {{formattedAuthorsCount} others}}</0> signed up with your starter pack" | ||
| msgstr "{firstAuthorLink} și <0>{additionalAuthorsCount, plural, one {un alt utilizator} few {alți {formattedAuthorsCount} utilizatori} other {alți {formattedAuthorsCount} de utilizatori}}</0> s-au înscris folosind pachetul tău de început" | ||
| msgstr "{firstAuthorLink} și <0>{additionalAuthorsCount, plural, one {{formattedAuthorsCount} alți} few {{formattedAuthorsCount} alți} other {{formattedAuthorsCount} alți}}</0> s-au înregistrat cu pachetul dvs. de pornire" |
There was a problem hiding this comment.
We are using the informal language ("tău", not "dumneavoastră" or "dvs."). See #6456 (comment)
|
Hi, sorry, I am new to Github and I did not see that PR Overall my translation is based on official translation of Twitter (Using it for a few years their translation is engraved in my mind) and other platforms and companies like Microsoft, mainly fixing incorect terms like: -mute/unmute are better translated as amuțire/dezamuțire, similarly to Discord -handle that cannot be translated into nume de utilizator because "user name" alredy exists and is very confusing to a new user of the app, I wanted to sign up and I was very confused. As a person with a backround in translation, the perfectionist in me was pissed off and wanted to took the matters in my own hands, and it turned into a complete remake. Im used to translations being as faithful to the original as posible, always refer to the user with Dvs., find translations that sound better hence "pachet de pornire". |
|
For example, "# zi" is more correct as # is replaced by a number "1 zi", using "o zi" technically breaks and does not work as it can't show the number 1 because there is nothing to replace. |
Of course they mean the same thing just that is not correct Romanian plural. The Romanian language has 3 plurals (singular + 2 plural variants). The variation is defined by the formula from the Using a single plural form is a relic of old times when there was no implementation of plural variants because 99.99% of the software was English and software localization was not yet an issue. The gettext translation project is aware of this particularity of some languages and introduced the possibility of
Consistent with English? As is a Romanian localization we should be consistent with Romanian language rules.
I don't have a strong opinion, but "dezamuțire" doesn't sound good. I would let @alextecplayz discuss this as she proposed the actual form in #6456 (comment).
Quoting @alexandregz from the same #6456 (comment): "Romanian does not have an equivalent to a user handle specifically, so you'd have to use the Romanian equivalent of 'user name' instead, "Nume de utilizator". The literal meaning of "mâner" is the handle of a door, or the part of an object that you can grab and maneuver, and it certainly can't be understood in the context of a social media app. Don't use "identificator" (identifier", because it's different from the username." But, of course, we can discuss on this. However, my background of software developer, tells me that "user name" (often used as username, in a single word) refers what here is called handle, i.e., a combination of letters and digits, uniquely identifying a user. In the same time "identifier" is mostly used for an internal ID, such as a serial integer or a universally unique identifier. User name is something that the user is choosing to represent them in that virtual space while the identifier is a value assigned by the system.
The original (English) language has only one form of addressing. Romanian (and not only) has two forms: one formal (dumneavoastră, vă rog, etc) and the other informal/colloquial (tu, te rog, etc). Trying to be "faithful to the original" is not possible, you'll have to make a decision and pick one form or the other. Both are correct in Romanian language but the context is important. When addressing an official you'll use the formal form but with a friend you want to use the colloquial form. As this is a social app, I seemed to me more appropriate to use the informal way. We've discussed this in #6456 (comment), #6456 (comment) and #6456 (comment) and decided to continue with the informal form as it makes the app more familiar and friendly.
I'm a software developer and I know how this works. The "#" placeholder is replaced only if exist. We can discuss whether "1 zi" sounds better than "o zi" but the technical aspect should not be part of the discussion. Both forms are working correctly, w/o bugs (the current translation which is deployed shows it) I do agree that the translation can and should be improved, but we need to decide on the process. Here's a proposal:
|
"Amuțire" and "Dezamuțire" are usually related to a person losing the ability of speaking (becoming mute) and regaining the ability to speak, so while I really wish that we could use something as short as this, it probably doesn't work well in the context of social media - the user we're muting is not losing the ability to speak, because we are silencing their replies on our side, for a single conversation or their entire account. "(Dez)amuțire" could work in a social media moderation context, however. Users lose the ability to speak / get shadowbanned, limited, muted, etc for specific actions. And as for the general translation process: as you linked the new release process from the Bluesky team, it seems like they're officially going to use Crowdin at some point (hopefully in the next few month(s)), which would make updates significantly easier since we wouldn't have to create new or re-open PRs for translation stuff - something that IMO should have been present since the beginning of the project, or at least since Bluesky left beta. |
This PR updates all missing strings into romanian and corrects certain words (uses better terminology).
The translation was done on Crowdin and was checked by multiple people.