Bimlas,
Thanks for this example. In this case you are answering a question I did not yet ask, you are recording the most recently viewed tiddler, and most viewed tiddler.
To me this fits the analytics approach to tiddlywiki we have not yet fully explored , thanks, I love it.
I too have being doing a lot to monitor or allow changes on tiddlers without touching them. For example I have a note button that takes notes relating to a tiddler without changing the tiddler. Especially useful for annotating core tiddlers but I plan to use a Unique serial number to overcome renames. I will publish this soon in a copy of
tiddlywiki.com, for research and notes linked to the documentation, after the next release. Obtaining and setting a unique serial number follows similar logic to your solution.
If we can capture various actions and count them for tiddlers from view, like your example, to edit, but perhaps even when found in search or jumped to in the story river etc... this information can accrue then tell us a lot about our own or other users behaviour.
I also have a tool that lists the history tiddler and this is very effective for a particular session, because the most recent is listed at the top, including the 2nd 3rd ... etc... I hope some day to also allow history to accrue across sessions.
BUT, in this case I am interested in a MRU for values selected and applied to say field selection, or tag assignment. As a result if we do anything if we go to do it again the last and previous values selected will be in a list. I expect some kind of action we can call in the select widget will support this. Then below the MRU list would be other values not in the MRU list which are available for that setting. I will use a filter here, so in some cases we get the values from a preconfigured list, and others we get it from existing values already used eg each field value.
So what I am asking for here is sub tiddler values, setting fields tags etc... although as I said I love your solution for analytics.
I feel strongly that two areas popular in many modern applications we can make use of in tiddlywiki, such as;
- Analytics and "MRU Values" are as yet untapped possibilities. Many of our enthusiasts for example love to find information from references, back links and tags etc... to discover "organic relationships", but the order and patterns of use by ourselves or users can help inform or bring to the top the things of most importance to us, or for that matter neglected items.
- The other popular method involves smart search and suggestions based on related date using knowledge of previous searches and connections, we could call this "little data" as opposed to "big data" that adds a level of machine learning to our tiddlywikis. Just look at Google search and the suggestions, if we could get wikis to respond to search and analytics we could once again help the "knowledge collectors" learn even more from their repositories, to the extent of giving results that look like magic.
- In this smart search approach it is actually possible to do without a TOC etc... because the information appears when it is needed, not that TOC and other structures do not simultaneously have use cases. Imagine an automatic TOC that builds itself from each users or all users activity.
Related ideas
- We can also make use of the existing change count to see numbers of edits in a session, making this also be multi-session, ie carry over values for analytics across sessions.
- Even just counting how often each button is selected will quickly track usage and allow us to show buttons most used and hide others behind more.
Regards
Tones/Tony