Tones, your description of your hierarchy brought back to mind the PhD thesis work I did on extending conventional data dictionaries to cater for knowledge objects. I created a proof of concept using Hypercard, a precursor of TW, using a data model similar to what you are describing. The data model is [image: Screen Shot 2020-12-06 at 12.15.20 pm.png]
The dual associations between Entity Type and Allowed Relationship and Entity and Relationship are to record the owner and member of a relationship. The top three entities provide a model of the domain at a conceptual level whilst instances are recorded in the bottom three entities. Entries in the bottom three entities must conform to those allowed, ie. those recorded in the top three entities. So, entities must be of an allowed type. Relationships must be of an allowed type between allowed entity types. Attributes must be of the allowed type for that entity type. You can read some articles about the Knowledge Dictionary on my ResearchGate account (researchgate.net) This discussion takes me back many years. Fascinating how things come round again. bobj On Friday, 4 December 2020 at 13:31:19 UTC+11 TW Tones wrote: > Gentlemen, > > I just want to add if there has not being a database model before, > tiddlywiki is an ideal platform to model any relationship. Of late I have > endeavoured in any application to never compromise the ability to add an > additional layer of organisation, an alternate view or a different > simultaneous representation. An old line "not taking hostages of the > future" my father quotes, is reinvented by me to "Not taking decisions > that compromise the future" is an interesting approach on top of tiddlywiki > especially when looking at alternate database or knowledge models. As one > proceeds to "try different systems" on top of tiddlywiki we gain practical > experience with a kind of meta database systems view. > > One Idea of my own that may be of interest, not withstanding Charlies love > hate relationship with hierarchy ,is the following model I am keen to > experiment with. > > - Every object is a tiddler > - Every object is in a hierarchy, even if it begins with only one > - Every attribute is a relationship to an object in another hierarchy > - Hierarchies act as I kind of "fuzzy value" where with more > information the hierarchies go deeper as they grow > - When assigning an attribute a value you do so via a relationship to > a hierarchy if you find it you use it, if not you add it, > - If you do not have a detail ie it is coloured but no what color it > is you point to an item in the color hierarchy such as color - or unknown > colour. > - Should you come across a database of colors you use it to populate > the colour hierarchy, and where possible change items pointing into the > hierarchy you move the relationship to a less fuzzy member of the > hierarchy. > - People, a group, a process can take charge of a hierarchy and do as > they wish as long as the honour or improve the relationships already > codified. > > Just some thoughts > Tones > > -- You received this message because you are subscribed to the Google Groups "TiddlyWiki" group. To unsubscribe from this group and stop receiving emails from it, send an email to tiddlywiki+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/tiddlywiki/e25b0661-72f0-45b6-86c1-86583c5fd833n%40googlegroups.com.
