Look into how to use rdf:List.

[candlecao]

• I may refer to https://ontola.io/blog/ordered-data-in-rdf.
• It may be useful to do with the occasion where there are alias for tunes.
• I will see can this be used to deal with the occasion where "many to many" relation occurs instead of using named graph or blank nodes.

[candlecao]

Findings:

In RDF, rdf:List (or simply RDF List) is used to represent an ordered collection of items.(We can ask GPT: "How to use rdf:list?")

So I don't think it's suitable for the case of alias for tunes, unless there is precedence for a group of aliases for a tune.

In addition, for the case of aliases, what you want to refer to may be rdf:Bag used in scenario where you need to represent a collection where the order of items does not matter, and it is important to allow for the possibility of duplicate items:

Considerations:

• It needs to use empty nodes. Eg:

• :Person rdf:type :Individual ;
  :hasHobbies [
  rdf:type rdf:Bag ;
  rdf:_1 "Reading" ;
  rdf:_2 "Cycling" ;
  rdf:_3 "Photography"
  ] .

• Simplicity: In many cases, you might use multiple triples with the same predicate instead of an rdf:Bag, especially when the order doesn't matter and duplicates aren't a concern. However, if you need to group these triples explicitly, rdf:Bag can be useful.
• Tool Support: Not all RDF tools and systems handle rdf:Bag well, especially if processing duplicates or unordered collections is required.
• Clarity: Using rdf:Bag makes it clear that the collection is unordered and that duplicates are meaningful.

[candlecao]

• I will see can this be used to deal with the occasion where "many to many" relation occurs instead of using named graph or blank nodes.

-- This can not be used to deal directly with the occasion where "many to many" relation occurs...

[fujinaga]

Findings:

In RDF, rdf:List (or simply RDF List) is used to represent an ordered collection of items.(We can ask GPT: "How to use rdf:list?")

So I don't think it's suitable for the case of alias for tunes, unless there is precedence for a group of aliases for a tune.

What are "the case of alias for tunes"? I thought we are dealing with list of tunes in an album or a playlist. Ask ChatGPT: "What are some of the ways of representing sets of playlists with RDF, where each playlist have a sequential list of songs and some songs appear in several playlists. Assume each song has an URI."

[candlecao]

Findings:

In RDF, rdf:List (or simply RDF List) is used to represent an ordered collection of items.(We can ask GPT: "How to use rdf:list?")

So I don't think it's suitable for the case of alias for tunes, unless there is precedence for a group of aliases for a tune.

What are "the case of alias for tunes"? I thought we are dealing with list of tunes in an album or a playlist. Ask ChatGPT: "What are some of the ways of representing sets of playlists with RDF, where each playlist have a sequential list of songs and some songs appear in several playlists. Assume each song has an URI."

Hi, Ich. "the case of alias for tunes"--Maybe I mis-memorized or otherwise...
It's OK, you are right by making an example of "playlist". I complete knew your meaning this time.

Let's see a tune_set: https://thesession.org/members/1/sets/1

tuneset_id | username | name | tune_id | setting_id | type | type_wiki -- | -- | -- | -- | -- | -- | -- https://thesession.org/members/1/sets/1 | Jeremy | Tarbolton, The | https://thesession.org/tunes/560 | 560 | reel | https://www.wikidata.org/wiki/Q118120 https://thesession.org/members/1/sets/1 | Jeremy | Longford Collector, The | https://thesession.org/tunes/563 | 563 | reel | https://www.wikidata.org/wiki/Q118120 https://thesession.org/members/1/sets/1 | Jeremy | Sailor's Bonnet, The | https://thesession.org/tunes/570 | 24958 | reel | https://www.wikidata.org/wiki/Q118120

It contains 3 tunes. Yes. The sequence probably counts. Thank you for pointing this out!

[candlecao]

So, in order to render a sequential group of property values, we write RDF like this corresponding to the example above:

<https://thesession.org/members/1/sets/1> <https://thesession.org/tunes> (
      <https://thesession.org/tunes/560>
      <https://thesession.org/tunes/563>
      <https://thesession.org/tunes/570>
) .

@Yueqiao12Zhang Hi, please notice this too.

[candlecao]

So for our current method of CSV2RDF into Virtuoso, it's very important to analyze the data type, redundancy, cardinality constraint ... of the dumped raw CSVs, which is almost about schema of RDB.
Next round, when we are doing with CantusDB, SimssaDB... we can have the schema to check beforehand.

[fujinaga]

Did you ask ChatGPT: "What are some of the ways of representing sets of playlists with RDF, where each playlist have a sequential list of songs and some songs appear in several playlists. Assume each song has an URI." Which solution do you think we should use?

[candlecao]

Yes, I asked and got its reply. Based on that, I am quite confident that the best method is to use the format as:

<https://thesession.org/members/1/sets/1> <https://thesession.org/tunes> (<https://thesession.org/tunes/560> <https://thesession.org/tunes/563> <https://thesession.org/tunes/570>) .

@Yueqiao12Zhang