Wikidata:Property proposal/text of verse

From Wikidata
Jump to navigation Jump to search

text of verse[edit]

   Not done
Descriptiontext string of verse item
Representsverse of a text (Q3556186)
Data typeMonolingual text
Domainverse of the Bible (Q29154550), verse of a text (Q3556186), ayah (Q188219)
Allowed valuestext strings
Example 1John 3:16 (Q397709) → verse of text → For God so loved the world, that he gave his only begotten Son, that whosoever believeth in him should not perish, but have everlasting life.
Example 2Al-Baqara 256 (Q1550546) → verse of text → Let there be no compulsion in religion: Truth stands out clear from Error: whoever rejects evil and believes in Allah hath grasped the most trustworthy hand-hold, that never breaks. And Allah heareth and knoweth all things.
  • has edition (P747)Abdullah Yusuf Ali (Q1077843)
  • language of work or name (P407)English (Q1860)
  • Example 3John 1:1 (Q2538046) → verse of text → In the beginning was the Word, and the Word was with God, and the Word was God.
  • has edition (P747)King James Version (Q623398)
  • language of work or name (P407)English (Q1860)
  • Source
    Planned useto create items of the type verse of the Bible (Q29154550) and upload their texts
    Robot and gadget jobsI will import text strings using OpenRefine's Wikidata interface


    I've seen attempts to add the text of verses to Wikidata items with varying degrees of success, some in the item's description (see John 1:1 (Q2538046)), and some in native label (P1705), which is not extensible to other languages(see John 1:1 (Q2538046)). I've also seen motto text (P1451) used to record text for an item, as in "Abandon all hope, ye who enter here (Q17562638)", but this is clearly a misapplication of the property. I think that it's reasonable to include the text of a verse somewhere on an item's page. If I point to Matthew 2:10 (Q6789898), I should be able to retrieve the substance of that verse--its text--and I think a specific property is a good place to do it.

    I think this property would be useful in other religious and literary contexts as well, to allow for the Q-id and label to stand in for the language-neutral conceptual abstraction of the verse, with the substance of the verse in whatever language in its own property.

    Language of the string would be implicit in the has edition (P747) statement (e.g. the King James Version is a translation of the Bible into English), but ideally it would be mandatory to indicate the language of the text with language of work or name (P407) in a qualifying statement.

    • Symbol neutral vote.svg Neutral @LesserJerome: While I do not see a problem with creating dedicated items for verses in religious texts, I have a feeling that this property will lead to unnecessary bloat on such items. (There is also the issue of the texts themselves not being 'structured data', on which basis some people will object to this property, but the issue I bring up with bloat is independent of that point.) It may well be better to include information about the content of the verses using other properties (such as 'figures mentioned', 'Torah portion', 'cross-reference to other verse', 'Quranic maqra', etc.; some of these properties may need to be proposed), so that the information about such verses can be considered structured data (in how many verses of Book of Genesis (Q9184) is Enoch (Q213027) mentionedm for example). Mahir256 (talk) 23:50, 12 June 2019 (UTC)
      • @Mahir256: Hi Mahir, thanks for your input. I definitely do want to allow for more data to be structured around the verse items as you've mentioned, but I think having the text of a verse is still an important part of the foundation. For instance, in your hypothetical query, in how many verses of Genesis does Enoch appear, the results would be an amount of verses, a list of the verses, and maybe a graph indicating other relationships and hierarchies, but the one thing missing would be what those verses actually are. You'd then have to go somewhere else to actually find the text of the verses. The text strings may not be pieces of structured data themselves, but situated within a network of structured data, you can interact with them in new ways. I also take your point about bloat--there is that potential here, as there are many translations into many languages, and that's just counting the Bible--but if the alternative is not including the text at all, I think it's a missed opportunity. For my part, I am not looking to add all translations or even multiple (though I'm aware this might open the door for others to do so), and I think one or a few authoritative editions has some really cool potential for digging into this textual data on this platform. LesserJerome (talk) 16:27, 14 June 2019 (UTC)
    • Symbol oppose vote.svg Oppose I don't think Wikidata is the place for this. These texts are already available in Wikiquote -; I would support figuring out some way to link the Wikidata items to the relevant text(s) in Wikiquote, as the current sitelink mechanism does not work for this. ArthurPSmith (talk) 13:26, 13 June 2019 (UTC)
      • Symbol oppose vote.svg Oppose I agree with ArthurPSmith Infomuse (talk) 03:41, 14 June 2019 (UTC)
      • @ArthurPSmith: Hi Arthur, thanks for commenting. My idea in adding the text to each verse is to facilitate a way to interact with the text in a more complex fashion. The text of the Bible and many other works this property might apply to are already available in many places online, so I'm not just trying to put it in one more place. Texts in WikiQuote aren't structured in any really meaningful way, and texts in WikiSource are pretty flat, for lack of a better word. By attaching the text of these verses to the Wikidata items, you will be able to see their context in the greater hierarchical structure of the whole, and they would be query-able through the powers of SPARQL--you'd be able to do things with the text here in Wikidata that you could not do anywhere else. I think structuring and recording a text like this in Wikidata is an exciting, dynamic way to engage with it. LesserJerome (talk) 16:27, 14 June 2019 (UTC)
    • Symbol oppose vote.svg Oppose This would encourage too much copyright violating items. ChristianKl❫ 11:26, 27 June 2019 (UTC)