Are there any useful rules of thumb for optimization when using native TW components? Perhaps also concerning html+css.
For example; - In a *filter*, does it make significant difference to put the cruder operators+operands first/leftmost? - And particularly, does a filter with one compound expression ever stop mid filter e.g if the expression that far results in empty and the rest of the expression operators need input to be meaningfully run? I think this would require a kind of meta evaluation first to decide what type of an expression the argument is. - Does a ViewTemplate tax the system more than the average tiddler considering how viewtemplates are transcluded into all other tiddlers? - Or does a viewtemplate only cost when a tiddler is actually open (and in viewmode)? - ...and does the viewtemplate only cost in that particular tiddler or does it cost globally? For example, if I open a *stylesheet *that is NOT tagged $:/tags/Stylesheet, then these styles are active globally for as long as the tiddler is open. A macro on the other hand, has to be called (and I must assume it doesn't cost anything more than plain text before that). - Do local style blocks cost differently than dedicated stylesheet tiddlers? - Are wikitext solutions different in cost from their eventual widget/html/css counterparts? (e.g curly bracket transclusion vs transcludewidget, pipe characters vs html table, @@ vs wrapping in a span to style...) - Do the cost for keeping (plain text) tiddlers open increase linearly or perhaps exponentially? - Tiddlers in view mode vs in edit mode? For example, why do things slow down substantially when the vanilla stylesheet is in edit mode? Is it the mere size of it? Does an editview tiddler affect the actual tiddler content which slows rendering? - And apropos (something like) the vanilla stylesheet - if it were split up into multiple tiddlers, e.g one tiddler per style definition, how would that affect wiki performance? I keep hearing about TWs with X thousand tiddlers and "it works fine", so is tiddler *length *a sensitive issue? - What tiddlers or *types *of tiddlers, if any, or what type of content, is constantly taxing the system? Pagetemplate tiddlers? Macros? Anything tagged X? - Are there ways to ensure lazy loading of stuff? For example, it is known that long lists in the sidebar tabs can cause slowness (...or is it only for some of those tabs?) - but to prevent it, you can simply switch tab to a short list... so does that mean that content in non-active tabs are "turned off"? - ...same question for reveal widgets; Is their content inactive when it is not revealed? Or does it perhaps keep on running until tiddler is closed? or even after that? I'm guessing the common answer to all the above example(!) questions is "it depends on [specifics]" - but, I'm asking for general rules of thumb. Thank you! <:-) -- You received this message because you are subscribed to the Google Groups "TiddlyWikiDev" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/tiddlywikidev. To view this discussion on the web visit https://groups.google.com/d/msgid/tiddlywikidev/c9ab977d-00ac-4c8b-9dd6-1e91666b4a3c%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
