Hi,
\nI noticed that many of the entries in the eBL dictionary file don't have an English definition given. Would it be reasonable to assume that these words are names? Or would it be better to assume instead that their definitions are unknown?
\nIn the case that their definitions are unknown, if a model were to encounter these words, what would be the preferred way to deal with them when translating to English?
\nAnd if they are names, should they be converted to/from any transliteration format when put into English?
", "messages": [ { "id": "3387701", "postDate": "01/07/2026 13:57:42", "content": "Hi,
\nI noticed that many of the entries in the eBL dictionary file don't have an English definition given. Would it be reasonable to assume that these words are names? Or would it be better to assume instead that their definitions are unknown?
\nIn the case that their definitions are unknown, if a model were to encounter these words, what would be the preferred way to deal with them when translating to English?
\nAnd if they are names, should they be converted to/from any transliteration format when put into English?
", "rawMarkdown": "Hi,\n\nI noticed that many of the entries in the eBL dictionary file don't have an English definition given. Would it be reasonable to assume that these words are names? Or would it be better to assume instead that their definitions are unknown? \n\nIn the case that their definitions are unknown, if a model were to encounter these words, what would be the preferred way to deal with them when translating to English?\n\nAnd if they are names, should they be converted to/from any transliteration format when put into English?", "votes": null }, { "id": "3387710", "postDate": "01/07/2026 14:21:05", "content": "For proper names it's better to use OA_Lexicon_eBL.csv. It has a type PN and GN that will be helpful for you, including multiple variants of the same name. Column norm shows you how it should be translated.
I have heard about PN and PN₂ but who is GN ?
", "rawMarkdown": "I have heard about PN and PN₂ but who is GN ?", "votes": null }, { "id": "3388024", "postDate": "01/08/2026 00:51:28", "content": "Yes, some of these non-dictionary words are names, in which case you should find them under PN in the OA_Lexicon_eBL.csv (and soon we'll add an onomasticon). We also added the full eBL eBL_Dictionary.csv which you can use more broadly for searching, but yes, there are some unknown words still in the texts. \nSometimes you'll see instances of hyphenated words in the translations, such as \"2 salt- \" < 2 ha-we-ru ša ṭá-áb-té-en, for example. In this case haweru/hawiru is an unknown word and the translator used the normalized word in his translation originally: \"2 salt-hawiru\" (AKT 6b 268; Kt 94/k 670)
", "rawMarkdown": "Yes, some of these non-dictionary words are names, in which case you should find them under PN in the OA_Lexicon_eBL.csv (and soon we'll add an onomasticon). We also added the full eBL eBL_Dictionary.csv which you can use more broadly for searching, but yes, there are some unknown words still in the texts. \nSometimes you'll see instances of hyphenated words in the translations, such as \"2 salt- \" < 2 ha-we-ru ša ṭá-áb-té-en, for example. In this case haweru/hawiru is an unknown word and the translator used the normalized word in his translation originally: \"2 salt-hawiru\" (AKT 6b 268; Kt 94/k 670)", "votes": null } ], "comments": [ { "id": 3387710, "author_name": "acraevschi", "author_url": "", "post_date": "01/07/2026 14:21:05", "content": "For proper names it's better to use OA_Lexicon_eBL.csv. It has a type PN and GN that will be helpful for you, including multiple variants of the same name. Column norm shows you how it should be translated.
I have heard about PN and PN₂ but who is GN ?
", "votes": null, "replies": [] } ] }, { "id": 3388024, "author_name": "deeppast", "author_url": "", "post_date": "01/08/2026 00:51:28", "content": "Yes, some of these non-dictionary words are names, in which case you should find them under PN in the OA_Lexicon_eBL.csv (and soon we'll add an onomasticon). We also added the full eBL eBL_Dictionary.csv which you can use more broadly for searching, but yes, there are some unknown words still in the texts. \nSometimes you'll see instances of hyphenated words in the translations, such as \"2 salt- \" < 2 ha-we-ru ša ṭá-áb-té-en, for example. In this case haweru/hawiru is an unknown word and the translator used the normalized word in his translation originally: \"2 salt-hawiru\" (AKT 6b 268; Kt 94/k 670)
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "3387701": "Hi,\n\nI noticed that many of the entries in the eBL dictionary file don't have an English definition given. Would it be reasonable to assume that these words are names? Or would it be better to assume instead that their definitions are unknown? \n\nIn the case that their definitions are unknown, if a model were to encounter these words, what would be the preferred way to deal with them when translating to English?\n\nAnd if they are names, should they be converted to/from any transliteration format when put into English?", "3387710": "For proper names it's better to use `OA_Lexicon_eBL.csv`. It has a `type` PN and GN that will be helpful for you, including multiple variants of the same name. Column `norm` shows you how it should be translated.", "3387815": "I have heard about PN and PN₂ but who is GN ?", "3388024": "Yes, some of these non-dictionary words are names, in which case you should find them under PN in the OA_Lexicon_eBL.csv (and soon we'll add an onomasticon). We also added the full eBL eBL_Dictionary.csv which you can use more broadly for searching, but yes, there are some unknown words still in the texts. \nSometimes you'll see instances of hyphenated words in the translations, such as \"2 salt- \" < 2 ha-we-ru ša ṭá-áb-té-en, for example. In this case haweru/hawiru is an unknown word and the translator used the normalized word in his translation originally: \"2 salt-hawiru\" (AKT 6b 268; Kt 94/k 670)" }, "source": "meta" }