Tatoeba m'indique dans un premier temps que la phrase "N'interrompez jamais un ennemi qui est en train de faire une erreur." (citation de Napoléon Bonaparte) n'existe pas encore dans son corpus mais quand je veux l'introduire, il prétend que ma phrase ne peut être ajouté au prétexte qu'elle existe déjà (introduite par Voltaire (robot du Projet Voltaire).
Que faire parce qu"on n'a pas accès à cette phrase ?

I had a discussion at #13684027 , and found some problem that is universal to the corpus, so I want to discuss here.

An inactive non-native speaker wrote this sentence #12888798: "Do you want a electric gaming machine?"

I personally think that sentence is odd, not used, but it's completely grammatically correct. But we cannot infer that if the author of the sentence meant "consoles" or "video games". How should we deal with it? Should we keep it as "Do you want a electric gaming machine?", or should we change it into "consoles" or "video games", or should we delete it?

建議在 [詞彙請求] 中優先顯示例句數多的詞彙項，後顯示例句數少的詞彙項。

比如，包含「學生」的句子可能有 5 個，包含「大學生」的句子可能有 0 個。由於 [現有的詞彙請求] 中優先顯示例句數少的詞彙項，所以大部分人會優先新增「大學生」的句子。當包含「大學生」的句子達到了 10 個，包含「學生」的句子就達到了 15 個。但包含「學生」的句子大多數都是「大學生」。這樣對語料庫的詞彙豐富度是有害的。

I suggest that [Vocabulary Requests] prioritize displaying vocabulary items with more sentences, and show items with fewer sentences later.

For example, there might be 5 sentences containing "學生" and 0 sentences containing "大學生". Since the [Existing Vocabulary Requests] prioritizes items with fewer sentences, most people will add sentences for "大學生" first. By the time sentences containing "大學生" reach 10, sentences containing "學生" reach 15 — but most of those "學生" sentences actually contain "大學生". This is harmful to the lexical richness of the corpus.

In another post - https://tatoeba.org/de/wall/sho...#message_41768 - I offered to @LeviHighway to publish some small programs I'm using to extract greek sentences and roughly sort them by difficulty, based on word counts.

Meanwhile they are online. You can find them at https://github.com/kumakyoo42/tatoeba_tools

The sorting algorithm is mainly designed for languages with spaces between the words. I added support for counting symbols instead of words. With this, I hope at least, sentences of other languages might also be sorted roughly by difficulty. On a whim I checked this with Japanese sentences. At first glance it looks like it might work. The easiest one was "はい" (yes). But I don't know enough about Japanese language to judge the more complicated sentences.

Corpus Maintainer Candidate for Chinese (Mandarin);
Kandidato por la bontenado de la ĉina korpuso;
Candidat pour la maintenance du corpus chinois (mandarin) ;
Кандидат в ответственные за корпус для китайского (мандаринского) языка;
Korpuspflegerkandidat für das Chinesische (Mandarin).

LeviHighway: https://tatoeba.org/user/profile/LeviHighway

🇬🇧 As usual, please feel free to send us a private message to share your opinion (click on the link below).

[epo] Kiel kutime, ne hezitu sendi al ni privatan mesaĝon pri via ĉi-rilata opinio (alklaku la ĉi-suban ligilon).

🇫🇷 Comme d’habitude, n’hésitez pas à nous envoyer un message privé pour nous faire part de votre opinion si vous le souhaitez (cliquez sur le lien ci-dessous).

🇷🇺 Вы можете, как обычно, отправить нам личное сообщение, чтобы поделиться своим мнением (кликните по ссылке ниже).

🇩🇪 Schickt uns wie immer gerne eine Privatnachricht, um uns eure Meinung mitzuteilen (auf den Link unten klicken).

https://tatoeba.org/private_mes...sichbaeumchen.