Saq, tremendously thankful for the detailed response. It was clear and very helpful in furthering my understanding of how TW5 works.
I do see (and had read in other threads on Stroll) the difference between plugins and editions and the complexity that offering an out-of-the-box edition brings. As best I can tell from my limited experience with it, the plugin route should most often be preferred, since it aims to reduce dependencies and modifications to core elements. To that regard, I personally prefer living in the constraints to search with Streams and other issues over trying to develop a specific version that removes some of those only to have to deal with issues of extensibility further down the road. Though I see the benefit of a vertical edition such as Stroll, my comment about linear guides was less about making a specific edition of TiddlyWiki and more about creating a step-by-step walkthrough (like an article or how-to guide) for constructing a TW5 instance from available plugins and themes that are known to work together well. I'm thinking about a tutorial that would get someone up and running for a common use case (notetaking and research), without hunting through the threads for different pieces to bolt on. For instance, at this point, I've discovered a couple of themes (Whitepace and Krystal), your Streams and Stories plugins, the Refnotes plugin, and the Crosslinks plugin for backlinks and freelinks all work well together to create a solid notetaking instance. Had I been able to locate a tutorial that explained TiddlyWiki setup for one notetaking instance and turned me on to that set of themes and plugins, I would have been up and running much faster. I was merely suggesting it might be helpful for the community to come up with the 3-5 most common uses and some solid walkthroughs for how to build a good instance out of available themes and plugins. That is why I pointed out Anne-Laure's work on her site. In fact, I think it better to provide a walkthrough than provide a packaged instance of TW5 with the themes and plugins already together, since a walkthrough has the added benefit of teaching just enough about customizing your TiddlyWiki in order to add the plugins. It's the right kind of friction, but it walks with someone instead of leaving them in a wide open field. Of course, this information would be ever-changing as new plugins are developed, so it might not be a feasible approach to onboarding. Concerning the export, that's great news. I assumed there was some option, and I just wasn't sure how to accomplish it yet. I'll have a go at a request here to get that started. Thanks again for your response, and for your work on TiddlyWiki. It's been refreshing having this at my disposal of late. I look forward to any further feedback you may have concerning Streams. And again, I'm happy to help and answer any questions you may have concerning my use case. Thanks, Keelan On Tuesday, August 18, 2020 at 11:48:04 AM UTC-5 [email protected] wrote: > Hi Keelan, > > Apologies for the late response, I wanted to wait to reply to your > feedback until I had the time to do so properly. I am always happy to > discuss at length when the feedback is both detailed and pertinent, as > yours has been. I am going to going off on what seems like a bit of a > tangent at first, to properly reply to some of the points you raised. > > Firstly, I'd like to explain the concept of a TiddlyWiki edition, > sometimes also referred to as a vertical edition. Simply put it is a > TiddlyWiki meant to serve as a starting point, with a set of plugins and > tweaks preconfigured that work together well with no clashes, to provide a > specific set of features or workflow. The best example of this currently is > Stroll from David Gifford, which encourages a notetaking style dependent on > backlinks. > > The key part that is relevant here is that with an edition, you have the > ability to make sure that all the plugins play together nicely and don't > clash, and complement each other with their functionality. When writing a > plugin you have to try your utmost not to make changes to the default TW > behaviour beyond what you absolutely need to, to minimize the chance of > clashes with other plugins. > > As an example consider Streams. If it were to be extended to tweak the > default search results, it would clash with any other plugin that extended > the search results or the search interface. With a pre-configured set of > plugins, this isn't an issue. Furthermore, the more changes you make to the > defaults in TW, the greater the chance the plugin will need updates when > the TiddlyWiki core is updated, increasing the maintenance burden of the > plugin. > > So at this point, you would be right in thinking that editions is > basically what you alluded to when speaking of linear guides or "choose > your own adventure" use cases. There are some significant challenges > inherent in creating editions, amongst them: > > - the need for extensive user studies and testing, to design, test and > refine a workflow and featureset. This requires a significant amount of > time and effort from both the developer and the users working together. > Furthermore it requires developer(s) and users working together towards the > same end goal in an effective and efficient manner. A common shortcut that > is employed is for the developer to also act as the user, which works in > situations where the development is driven by the developer's own needs. > This, by the way is not the case for Streams which I personally do not use > and was created as a tech demo for the community. I will come back to this > later. > > - Customizing, simplifying and optimizing the workflow for a specific use > cases often involves making choices that limit or alter functionality. > While the components of an edition can be designed to work together well, > they are likely to clash with other plugins available in the community. > This often leads to discontent from the users that are unhappy with the > perceived lack of flexibility and customization in an edition, considering > that it is still based on TiddlyWiki. > > For example let us consider WYSIWYG editors, something that has been a > frequent topic of discussion in the TW community. Due to the way that > TiddlyWiki widgets, macros and transclusions work, implementing a full > featured WYSIWYG editor would be a very significant, challenging and time > consuming endeavour. There have however been comments in the community that > even a WYSIWYG editor that *only* handled simple text formatting would be > useful for everyday writing, but at what cost? > > I have a prototype of such an editor that only supports text formatting, > but that means giving up custom HTML and widgets and macros in the tiddlers > where you use it. So for instance, in your case to use such an editor you > would need to give up using the refnotes plugin in those tiddlers. This > perhaps hasn't been the most ideal example, but hopefully it serves to > illustrate the point about compromises and limitations imposed when > designing specific features and workflows. In some ways TiddlyWiki's > flexibility works against it in these situations, by setting certain user > expectations. > > - Last but not least is the maintenance burden that comes with providing > an edition, and keeping it up to date with TiddlyWiki core updates as well > as plugin updates. > > I do agree that there is a lot of untapped potential in providing editions > that serve specific workflows, but I am uncertain if a community as small > as this has the resources to tackle this. > > Similarly Streams would benefit greatly from an edition that incorporated > it and addressed some of the issues like search results, backlink results > etc. However, considering both my limited availability and the fact that I > do not use this myself, I am not the appropriate person to do so. > > Regarding Streams and the motivation behind its development: > > Overall I feel that there is a lack of innovation in the TW5 community in > recent years, compared to the peak years with TiddlyWiki Classic (the > previous version), especially in light of its potential. This may be in > part due to a change in the demographic, I think there were more active > developers in the community back then. The availability of more alternative > platforms has probably also played a role. My perception is that there is > far too great a dependence on Jeremy (the creator and developer of > TiddlyWiki) for new ideas and features. Considering how long TW5 has been > around, and how much potential and flexibility it offers even for those > that aren't JavaScript developers, I am surprised at how many avenues are > yet left unexplored. (Note that all of this is my perception only, as > someone that only just became active again in the community recently after > an absence of some years. Others will likely disagree.) > > TiddlyWiki 5 goes to significant lengths to allow things to be adaptable > and customizable without needing to write JavaScript, however this > potential is largely untapped. Consider that the entire user interface of > TiddlyWiki is entirely built with widgets and templates, however we have > seen so little innovation and experimentation with alternative UIs. I would > love to see someone just throwaway the default interface entirely and start > from scratch. It can be done without any knowledge of JavaScript. > > Streams was written as an adaptation of a personal task management > solution, to demonstrate workflows and interactions that were previously > considered difficult or not possible in TiddlyWiki. > > Towards this end the very first demo was written entirely without any > extra JavaScript. As such, it was meant to be more of a technology demo and > reference implementation to inspire other developers, rather than something > for people to use, and this contributes to some of the current shortcomings > in the workflow. I will endeavour to address them to the extent that I can > within both my time constraints, and the scope of a single plugin (and not > an edition). > > Regarding exports: exporting just the content of tiddlers in HTML is > entirely possible via a button in TiddlyWiki. I suggest posting a thread in > the group asking for assistance on this to get you going. You can then use > Pandoc to convert to your format of choice. Note also that my work on the > WYSIWYG editor that I mentioned earlier also involves a limited HTML to > markdown conversion option in TiddlyWiki, that could potentially be reused > to convert wikitext to markdown as long as its just plain text. > > Hopefully this addresses the more general feedback you provided, and I'll > follow up with more specific replies to the Streams related items a little > later. > > Regards, > Saq > -- 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 [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/tiddlywiki/0b916e06-2412-4829-ad46-8f10835805ddn%40googlegroups.com.
