TiddlyWiki and Hyperstructures - "zzstructures, mSpaces, and Polyarchies"

[Steven Schneider]

This is a bit out there, for me at least, but this article ("A Comparison of Hyperstructures: Zzstructures, mSpaces, and Polyarchies") most has got me thinking: Is TiddlyWiki a zzstructure, an mSpace or a polyarchy? Check out the figures: I think we could build tiddlywiki navigation that was illustrative of these cases.

Just curious to see if anyone thinks in this group is thinking along these lines...

//steve.

Available: https://eprints.soton.ac.uk/259230/1/mSpace_zzStructures.pdf

Formal citation: 

Michael J. McGuffin and m. c. schraefel. 2004. A comparison of hyperstructures: zzstructures, mSpaces, and polyarchies. In Proceedings of the fifteenth ACM conference on Hypertext and hypermedia (HYPERTEXT '04). ACM, New York, NY, USA, 153-162. DOI=http://dx.doi.org/10.1145/1012807.1012852

https://dl.acm.org/citation.cfm?id=1012852

[Ste Wilson]

Sounds interesting... What's the tl:dr? ;)

[TonyM]

Steven

Some of us are always thinking about conceptual models, Jeremy has led this approach and that is why I believe tiddlywiki is as adaptable as it is.

Some lines from the abstract suggest to me what they are looking for is what we already have in tiddlywiki.

Over the past few years, there has been work both in the hypertext and interaction communities that has considered how to represent smaller-than-page-sized amounts of data to deliver more flexible ways to interact with information, and to maintain context of that information. 

Further, these contextually associated pieces of information are themselves actively available to the user: the user can move immediately from their current focus of interest to the new focus of interest, without losing how that new information is related to the previous focus,

I will read further on this abstract but I am not expert in Set or graph-theoretic methods.

Reading so far, however, does stimulate my mind to an idea I had previously that a database made up of objects whose attributes are only ever relationships to other objects, would ensure all contextual information is  "complete". With a waver that fuzzy relationships should be available, eg; Your may want an address for me, but so far you may only know I am in Australia, the model should accept this less accurate data.

To me TiddlyWiki is an ideal playground for investigating these subjects and in fact already satisfies some much dreamed of qualities. For example every database structure or model I know about can be represented in TiddlyWiki.

Regards

Tony

[PMario]

On Sunday, March 11, 2018 at 9:32:43 PM UTC+1, Steven Schneider wrote:

This is a bit out there, for me at least, but this article ("A Comparison of Hyperstructures: Zzstructures, mSpaces, and Polyarchies") most has got me thinking: Is TiddlyWiki a zzstructure, an mSpace or a polyarchy? Check out the figures: I think we could build tiddlywiki navigation that was illustrative of these cases.

Hi Steve,

Very interesting article! ... But I think your question is misleading: >>Is TiddlyWiki a zzstructure, an mSpace or a polyarchy?<<

The article talks about information-structures. Starting from "1 dimensional lists" up to "trees" up to "multitries" - polyarchies - mSpaces - zzstructures. It discusses the last 3 in detail.

... and I think, TiddlyWiki has all of them - at the same time. ...

So I would rephrase the question: "Is it possible to represent tiddlywiki content using "zzstructures, mSpaces or polyarchies?"

My answer is yes. ... BUT at the moment we are not able to conveniently use them. ... 


A second question I would raise is: "Can we use the information presented in the paper, to improve TWs functionality"?

Yes ... reading the paper, I had several ideas already ...

eg: In TiddlyWiki using tagging we automatically create "multi-trees". Which means: "one tiddler can be part of several trees".

Like this:

   AA  A
    \ /
     B  
    / \
   C   D

One possible flow is:   A - B - C
Second possible flow: A - B - D
3rd: AA - B - C
4th: AA - B - D

Now this creates a "navigation problem":

 - Starting at A we click the "Next" button
 - B is selected -> OK
 - we click the "Next" button
 - C or D are possible -> "problem"

Using some insight from the paper imo makes it possible to solve this problem! ... That's cool :)

A related problem "Sending a link to B looses context"

 - We send a link to tiddler B to a different user
 - If S/he clicks "Previous" button
 - AA or A ??? -> problem

I think, if we would disentangle this structure using the mechanisms described in the paper, we could create a "new type of permalink" that would allow us to preserve the context of the "link-author"

I think this would be extremely powerful. ... Just think about it.

-------

I also had an idea, to improve the TOC-macro ... The only problem is TIME :/

have fun!
mario

[PMario]

Edited the tree to make it consistent with the text.

[Steven Schneider]

Mario, thanks for the insight.

I like this q: Is it possible to represent tiddlywiki content using "zzstructures, mSpaces or polyarchies?

I'll continue to share here (and elsewhere) as I progress in this work...