Talk:Q55739694

@Wostr: your edits here establish the class "organonitrogen" as separate (subclass) of "organic nitrogen" compound. You didn't give a reference. I have failed myself to find one. Please give a reference, else I will merge the two. --SCIdude (talk) 15:37, 4 July 2021 (UTC)

@SCIdude: You have one in the item: [1] — organonitrogen compound = Any heteroorganic entity containing at least one carbon-nitrogen bond. Any class that is named as organo... compound have similar definition (including organometallic compounds) — bond between C and another element is required. Examples:

• methyl nitrate (Q425393) and n-pentyl nitrite (Q412350) are organic compound (Q174211) and nitrogen compound (Q3685323), but are not organonitrogen compound (Q55739694) as there is no C–N bond in the structure
• isoniazid (Q423169) and aspartame (Q182040) are organic compound (Q174211) and nitrogen compound (Q3685323), also are organonitrogen compound (Q55739694), because there is C–N bond in the structure.

Definitions such as in ChEBI are frequently ignored by non-chemical sources (medical, pharmaceutical, etc.), so there is often confusion about what compounds should be called organo... compounds. Regarding the second class: we could have subclass of (P279)organic compound (Q174211) + subclass of (P279)nitrogen compound (Q3685323) in methyl nitrate (Q425393) and n-pentyl nitrite (Q412350), but it is more convenient to have only one class (and this class is organic nitrogen compound (Q77899525)). This class does not introduce any new definition, but merely combines two classes for convenience. In some of your edits I noticed that you mistakenly considered organo... compound as every organic compound with a nitrogen atom. Examples:

• sulfur diimide (Q25323761): this class has a general structure RN=S=NR′ — it is organonitrogen compound (Q55739694), but it is not by definition organosulfur compound (Q422785) (there is no C–S bond)
• alkaloid (Q70702): by definition alkaloids are organic nitrogen compounds, I think most of them are organonitrogen compound (Q55739694), but are you sure that 100% of alkaloids have C–N bond? I'm not.
• amidines (Q423088): derivatives of oxoacids (not only carboxylic acids) in which –OH is replaced by –NR₂ and =O by =NR. There may be no C–N bond in many of amidines.
• oxacycle (Q71573538): is an organic oxygen compound, most heterocyclic oxygen compounds have C–O bond, so such compounds are organooxygen compound (Q71574018). However, e.g. 1,2,5-oxadiazolidine is a oxacycle (Q71573538), but is not organooxygen compound (Q71574018) as there is no C–O bond.

I can go on and on, but I think the above examples are sufficient to conclude that these two classes are not the same. Definition of organonitrogen compound (Q55739694) is taken directly from ChEBI (but this is not the only source that points to this definition and the necessity of a specific binding in the structure). I don't think that organic nitrogen compound (Q77899525) require any source, as this class is a convenient combination of organic compound (Q174211) and nitrogen compound (Q3685323), i.e. any organic compound + any compound with a nitrogen atom. In other words, organic nitrogen compound (Q77899525) is not required at all, we can use organic compound (Q174211) and nitrogen compound (Q3685323) for the same purpose. If you have any more questions, I will try to clarify any further doubts. Wostr (talk) 22:31, 4 July 2021 (UTC)

@Wostr: I understand the need for two classes. However, a paper or book source is needed for the use of "organo-" as a prefix (or any other term) to distinguish the two, because I think this usage is not supported. If we can find a term so that the definition "organic compound with at least one covalent carbon–nitrogen bond" can be inferred from it then this would be the second best solution. However, having only the Wikidata definition simply does not suffice. --SCIdude (talk) 06:34, 5 July 2021 (UTC)

@Wostr: Never mind. I found a reference myself: https://www.google.de/books/edition/A_Dictionary_of_Chemistry/UXMqCwAAQBAJ?hl=en&gbpv=1&dq=organo+prefix&pg=PA396&printsec=frontcover. Quote: "organo- Prefix used before the name of an element to indicate compounds of the elements containing organic groups (with the element bound to carbon atoms)". I will use this reference. --SCIdude (talk) 14:27, 5 July 2021 (UTC)

@SCIdude: as I have a shift work (mainly second and third shift) I am available at odd times of the day and can take part in discussions usually only once a day. I'm not sure for what exactly you needed the source. Definitions of organo-X compounds like organonitrogen compound (Q55739694) were inferred mainly from ChEBI which is now our main source for most classes of chemical compounds. For most organo-X compounds classes there shouldn't be any problem, however, there is a problem with organophosphorous compounds and we will have it also in WD — there is not the slightest consensus in the literature as to which compounds should be named so. Even ChEBI states that this term is often applied to compounds without C–P bond. There is one big classification mess in the English-language chemical literature. Such mess is not good for WD, because classes should have clear boundaries. Wostr (talk) 22:36, 5 July 2021 (UTC)

I need a source to add to P31 and P279 statements of these classes. The statements will be the backbone of our ChEBI-independent compound ontology, which I'll soon present on the Wikiproject. Remember, we cannot simply copy part of ChEBI for copyright reasons, the license is not compatible to Wikidata's CC0 license. Regarding the classification mess, I would propose to replace organophosphorous compounds with a term that describes the class exactly---even if that takes a lot of space---and put the organophosphorous compounds as alias. I have changed Q2182492 accordingly. An example, where a respectable database does things similarly, is enzymatic activities, e.g. hydrolase activity, acting on acid anhydrides, in phosphorus-containing anhydrides (Q14908099). --SCIdude (talk) 15:44, 6 July 2021 (UTC)

@SCIdude: regarding the basics of classification of compounds in WD, I tried to write a proposal for this (unfinished version is here: User:Wostr/Problems of classification), however, like with many other things, I don't have much time for this. Maybe there are some commons points. Wostr (talk) 22:09, 6 July 2021 (UTC) PS Regarding copyright issues: simple definitions cannot be copyrighted, so using them in WD won't be a problem, even in many items; it would be a problem however if someone copied a part of ChEBI in an organized way (e.g. automatic, using a bot). I doubt that the copyright issues with the use of ChEBI data may ever be a problem, as WD is a bigger project, even if you only count the chemical compound data. However, it is true that one have to be careful, especially with big data imports. Wostr (talk) 22:29, 6 July 2021 (UTC)