Hi Keelan, The detailed feedback is highly appreciated, no need to apologize. I will reply properly when I have a bit more time, but wanted to drop a quick note that the latest version of Streams should not have the issue you experienced with needing to add a space at the start of a node:
https://saqimtiaz.github.io/sq-tw/streams.html#%24%3A%2Fplugins%2Fsq%2Fstreams Cheers, Saq On Thursday, August 13, 2020 at 3:06:31 PM UTC+2, Keelan Cook wrote: > > As promised, my response to your questions... at length (my apologies). > > *Have you encountered any issues or pain/choke points in your workflow > with Streams?* > > Overall, I'm very impressed with the manner in which Streams does what it > does. So, pain points are slight, but there are some of course. I've been a > long time Workflowy user, and I often find myself wanting it to work like > WF in instances. Furthermore, there is a good chance the pain points are > simply from the nature of TW and more an issue of me having to relearn > behaviors. > > - *Hitting enter in a node with text does not always behave the way I > want.* - I'm very glad you added the ability to split a tiddler into > two nodes by hitting enter. It would really slow down my workflow if that > process was inhibited, as I often write full paragraphs into a node, > realizing only afterward that they need to be separate paragraphs. If I > had > to confirm this change every time, it would impede using Streams for > writing and notetaking. However, I also expect the reverse behavior > because > of my time in Workflowy. If I am at the beginning of a node that needs to > be connected to the one above it, I expect hitting backspace to join the > two nodes. And, of course, Streams doesn't do this, so I find myself > regularly attempting this action by reflex. > - *Nested nodes stay with an empty node when hitting enter instead of > moving with content.* - If I have a node with nested tiddlers and hit > enter at the beginning of the text in that node, my expectation is that > all > the text of that node and the nested nodes under it would move down, > creating a new blank tiddler above with nothing in it. Again, this is a > Workflowy behavior. However, in Streams, this action leaves all nested > information under the blank node created and moves the text from the > parent > node down into a separate tiddler below everything. > - *Ctrl+down does not always behave as I expect.* - If I've collapsed > nested nodes and want to switch between top level nodes. For example, if > I'm writing an outline, I will have headings. If I collapse the > subheadings > under each heading and then attempt to use the Ctrl + down function it > switches down to the nested node that is not visible (one of the > subheadings) instead of switching down to the next heading that is visible. > - *Having to add a space at the beginning of a node.* - Finally, there > appears to be some quirk where I have to enter a space prior to using a > text expander. It will not call the action if I enter the slash as the > first character in a new node. So, I have to hit a space and then put it > in. > - *Search gets confusing.* I've learned to adjust to this, but all of > the streams tiddlers show up in search. Most of the time, I only want the > parent to show and then I will find what I want on that "main tiddler." > However, I periodically have a tiddler created in a stream that rises to > the level of a stand alone tiddler, so I would want it to index. I realize > the dilemma in all of this. So, I think if we could omit all the children > and just index the main parent tiddler for a stream in search, it would be > preferrable, since it is possible to just rename a specific stream tiddler > and then it could index. At current, when I have one of these special > stream tiddlers that I want to also stand alone, the relink gives me the > ability to name it with something outside the stream and keep it in the > stream. That's probably clear as mud. If you can't tell, I'm still > thinking > through this one myself. > > *Are there any features you find yourself wishing that Streams had?* > > - *Easily adjust max width.* - It would be nice when you develop a > configuration panel to include adjustable max width. I've found that line > in the style and can easily adjust that way, but having that option in the > control panel would be nice. > - *Different views would be stellar. *- A Document view, in specific, > would be a great addition for me. As mentioned previously, I use this to > create a source tiddler for literature notes, then I create an outline > tiddler (also with Streams) where I draft the outline of the article I am > writing. I can easily drag my notes from source tiddlers to fill in the > article outline. With a little manipulation, these notes become paragraphs > within the outline. The outline becomes the article. If I could then > switch > it to a document view that removes the bullets and makes nodes headings, > that would be incredible. > - *Conditional styles(?)* - Even while in outline form, I wish there > was a way to make some styles conditional. Though, I have no idea if > that's > even a thing CSS can do. For instance, if I turn a node into a heading > with > !!, it would no longer show the bullet next to it, but still show the > bullets on all body text in nested nodes under it. Another instance where > this *might* be nice is with a block quote. Though, the ability to see > that a block quote is its own tiddler in the stream instead of being part > of the above bullet is important info. > - *Flattening a whole stream into a tiddler.* - Finally, and I've seen > this mentioned in the threads here on the group, the ability to flatten to > an article, conflating all tiddlers into one that can then be exported to > word processing software for publication. I did see that on your roadmap, > and I'm interested in the implementation of that. > > *Overall in your larger workflow in TiddlyWiki, not just related to > Streams, what if any are the roadblocks that slow you down or cause > difficulties?* > > *Export*, but there's a good chance this is mostly ignorance on my part. > I'd love a way to export as markdown, or some easy integration with pandoc. > A good LaTex export would be dynamite. I know I can export as a static > html, but that pulls all of the styles with it from the page, which is > great if I want to display it as an html file in my browser, and sometimes > I do. It is suboptimal if I just want the text portion of the tiddler, > properly formatted, for opening or importing into a word processor or > markdown/LaTex editor. I've experimented some with online pandoc tools, and > they work to a level, but I'd love to click a button somewhere in > TiddlyWiki and make this happen. Again, it may be my ignorance in > capability that is prohibiting this at current. > > *I do not know the horizon of possibility with TiddlyWiki.* There are > reams of content written in this Google group and other places on > TiddlyWiki, so it's not for want of content. However, I do think (as has > been echoed in this group in other threads) that it is a daunting task for > someone new to even realize what TiddlyWiki can do. What is clear from my > last 4-5 months living in these group threads is that it can do almost > anything you want, if you have the knowledge of how to make it do so. > That's incredible. > > And yet, my personal use cases are meager, and clear guides of how to > "choose your own adventure" for particular use cases seem missing (or I > can't find them). While the ability to capture non-linear, networked > thought appears to be the shining strength of TiddlyWiki, it would do well > to have some linear explanations, or guides, for easy setup. Linear content > has its merit, of course, and it is sometimes the best way to present > information or make a specific argument. An easy example is the work that > Anne-Laure did recently on her site providing a linear guide for building a > nonlinear notetaking tool with TiddlyWiki. Of course, she has a large > platform, but I think the ease of her linear explanation to build a > particular instance of TiddlyWiki has done much to bring new people into > the fold. It was a sizeable plank in my own bridge. > > Ideally, I wonder if it would be beneficial to come to some form of > consensus on the majority use cases (networked notetaking, static website > generator, table top gaming and world building, etc.) and then provide some > linear walkthroughs of how to build a core instance of TiddlyWiki for those > things. Of course, with some discussion of potential options to modify and > the caveat that many more things are possible the more you dig. I often > hear TiddlyWiki described as a box of Legos from which you can build > anything. This much is true; however, a box of Legos most often comes with > a set of step-by-step instructions to build the specific thing on the cover. > > *Are your notes mostly plain text with formatting (bold, italics etc) or > do you use a lot of macros and widgets in the actual note taking?* > Yes. It's mostly plain text formatting. Very little in the way of images, > diagrams, or embedded media of any kind. My current use case is classic > notetaking. I need to quickly and easily capture literature notes from > sources I am read. I want to capture them in such a way that they (1) are > easy to find based on their connections to other ideas and (2) easy to > manipulate in an outline. Then, I need a tool where it is easy to organize > an outline and fill it out into an article. So, a minimal, easy to use > interface that allows for long form writing with as few distractions from > toolbars and buttons as possible. I'll probably never have use for > mathematics or anything computational in my instances. Though, I realize > I'm just one use case for TiddlyWiki. > > > I currently do very little with widgets and macros, though I believe I > will do more as I familiarize myself with their ability. I'm still learning > to drive TiddlyWiki. I'm very impressed by the things I see people > suggesting in this group concerning macros and widgets, and I imagine once > I have a populated wiki, I may want to use these to call certain things. I > can tell I'm just scratching the surface with this. > > > Thanks, Saq, for your labor on this project. I have this sense that the > timing of several developments with TiddlyWiki of late are surfacing it as > a very promising tool for people who live and die by knowledge management. > Feels kinda like a renaissance of sorts. > > Do let me know if you have further questions. I appreciate you asking, and > am happy to help in any way I can. > > Blessings, > Keelan > > On Wednesday, August 12, 2020 at 8:35:03 PM UTC-5 Keelan Cook wrote: > >> I'm happy to do so. I completely understand the need for feedback to help >> focus limited bandwidth. >> >> Let me think on the questions and get back with you tomorrow. Again, >> thanks for the work you've already put into Streams. >> >> On Wednesday, August 12, 2020 at 9:55:36 AM UTC-5 [email protected] >> wrote: >> >>> Hi Keelan, >>> >>> Thank you for the detailed feedback and for explaining the context >>> behind your usage. It is very helpful to get a different perspective on >>> things and to know that others find it useful. >>> >>> My TiddlyWiki fiddling is something I do on my own time in addition to >>> two fairly demanding day jobs. While it is enjoyable for me, I must admit >>> that I carry over from my professional life the desire to be productive and >>> efficient in what little time I do have for it. Feedback goes a long way >>> towards knowing what to focus my energy and time on. >>> >>> The combination of the Stories plugin and Streams is what I had in mind >>> when I worked on the drag and drop support within Streams. I haven't really >>> mentioned this anywhere so I am glad to hear you realized the possiblities >>> in this. I think there is need of a bespoke customized TiddlyWiki edition >>> around Streams (and other plugins) to fully utilize its potential. For >>> example, you could if so desired, get the backlinks in Stroll to reference >>> only the root node of each backlink instead of each node. Similarly, the >>> search could be adapted to better integrate with Streams tiddlers as well. >>> Unfortunately that isn't something I have the time to devote to myself. >>> >>> I do have a few questions for you if have the time: >>> >>> 1. Have you encountered any issues or pain/choke points in your >>> workflow with Streams? >>> 2. Are there any features you find yourself wishing that Streams had? >>> 3. Overall in your larger workflow in TiddlyWiki, not just related >>> to Streams, what if any are the roadblocks that slow you down or cause >>> difficulties? >>> 4. Are your notes mostly plain text with formatting (bold, italics >>> etc) or do you use a lot of macros and widgets in the actual note taking? >>> >>> Thank you. >>> >>> Cheers, >>> Saq >>> >>> >>> On Wednesday, August 12, 2020 at 2:45:23 PM UTC+2, Keelan Cook wrote: >>>> >>>> Saq, >>>> >>>> I've been lurking in this group for a few months now, and your comment >>>> wondering whether or not people are using Streams has pulled me out of the >>>> shadows! I wanted to vocalize how significant this plugin has become to my >>>> workflow. >>>> >>>> A bit of background on my usage. I'm a professor and do a lot of >>>> research in history and the humanities. My work requires a really good >>>> notetaking workflow. On the other hand, I'm not even close to a >>>> programmer. >>>> So, I gravitate toward out-of-the-box systems, and they always have pain >>>> points. >>>> >>>> I've fiddled with TiddlyWiki off and on for several years. I've always >>>> wanted it to work for me, because I really admire the project. But, the >>>> ease of editing and quick note capture (as well as design aesthetics) >>>> usually ended my attempts and sent me back to other tools. Enter Streams. >>>> >>>> Like many others in this group, over the last 2-3 months the rise of >>>> TiddlyWiki as a Roam alternative pulled me back into the conversation. >>>> Dave's Stroll was instrumental in getting me back in. I stumbled on >>>> several >>>> themes that remove distaste for the UI. And then Streams became the pièce >>>> de résistance. Quick capture in adjustable outline form that allows for >>>> capturing literature notes and then drafting up outlines for longer form >>>> articles and books. I use your Stories plugin as well. I can pull up a >>>> Stream tiddler with source notes on one side and drag and drop content >>>> into >>>> my outline for works I'm creating. This is exactly what I've needed. >>>> >>>> Bimlas echoed my sentiments well in this recent post concerning ways to >>>> use TiddlyWiki: >>>> https://groups.google.com/g/tiddlywiki/c/Re11x96t-qI/m/GYuMKHx0AQAJ >>>> >>>> I've been actively working to create my own, bespoke, note-taking >>>> environment and I finally feel like all the pieces are there in TiddlyWiki >>>> to do so. >>>> >>>> So, I for one, am very interested in your continued work on Streams. >>>> It's minimalist, unobtrusive, and powerful for a researcher. I'm happy to >>>> provide any feedback that may be helpful to you, understanding I'm only >>>> able to speak toward end use and not the coding aspect of it. >>>> >>>> Thanks so much for your work on this. It's been a game changer for me, >>>> and I've recommended it to several others in my field at this point. >>>> >>>> Keelan >>>> >>>> On Tuesday, August 11, 2020 at 1:16:39 PM UTC-5 PMario wrote: >>>> >>>>> On Tuesday, August 11, 2020 at 6:13:53 PM UTC+2, Saq Imtiaz wrote: >>>>> ... >>>>> >>>>>> It's actually really nice to get some detailed feedback from real >>>>>> world usage, so firstly, thank you! >>>>>> >>>>> >>>>> I did like the fast workflow very much! >>>>> >>>>> >>>>>> I probably haven't tried quite so hard to make time to work on >>>>>> Streams recently since there has been radio silence around it, which >>>>>> would >>>>>> suggest it is either perfect (hah!) or that no one is using it. >>>>>> >>>>> >>>>> :) >>>>> >>>>> 6) I use [!is[shadow]has[stream-type]] as a stream enable filter. >>>>>>> This may change in the future. >>>>>>> >>>>>> >>>>>> I welcome suggestions for a sensible default for this. >>>>>> >>>>> >>>>> Yea, I didn't want to have it with every tiddler. ... But my choice >>>>> needs a manual input of the "stream-type: default" field. So I think I'd >>>>> need to make a new button. .. So we need a "Streams Icon" ... @anyone? >>>>> >>>>> >>>>>> >>>>>> >>>>>>> 7) I use this line <$vars >>>>>>> stream-root-title={{{[<currentTiddler>split[/]limit[1]]}}} in the >>>>>>> nodes-list-template. So tiddler names are created as shown in 5) .. >>>>>>> NO long node names. ... The disadvantage is, that "sub streams" don't >>>>>>> get a (+) >>>>>>> new Node button ... Only the "main stream" tiddler has this button. >>>>>>> .. >>>>>>> >>>>>> >>>>>> So two things to address here: >>>>>> i) have you tried setting this format for tiddler titles in >>>>>> $:/config/sq/streams/new-node-title, instead of changing the >>>>>> stream-root variable: >>>>>> {{{[<stream-root-title>split[/]limit[1]]}}}/<<now >>>>>> "[UTC]YYYY0MM0DD0hh0mm0ssXXX">> >>>>>> I think this will accomplish the same thing without the need for >>>>>> editing nodes-list-template, which may cause issues with the drag and >>>>>> drop >>>>>> code as well. >>>>>> >>>>> >>>>> You are right. That seems to fix the next problem for me too! >>>>> >>>>> >>>>>> ii) it was a design decision to only have the (+) add node button on >>>>>> the root stream, and not the sub nodes. I now see this isn't convenient. >>>>>> This will be changed, thank you for pointing it out. >>>>>> >>>>> >>>>> Sorry. I wasn't clear enough. ... I meant "sub nodes" in a separated >>>>> tiddler. ... But as written above. Your new-node-title construction seems >>>>> to fix both problems, that I had with my "hacks" >>>>> >>>>> >>>>>> >>>>>>> >>>>>>> 8) I do like the tiddler-title format as shown in 5), because it >>>>>>> automatically creates a visible "timeline", when the nodes have been >>>>>>> taken >>>>>>> and the core <<tree "DAT-CON/">> macro will be able to deal with it >>>>>>> after I did create a new "leaf template" for the macro. >>>>>> >>>>>> >>>>>> Could you point out the leaf template please? I'd like to add >>>>>> examples like these to the documentation. >>>>>> >>>>> >>>>> I'm not sure, what I actually want, yet. ... So I'll update the >>>>> DAT-CONF link once I've done more refactoring. ... As you know, I also >>>>> want >>>>> to make my tocP macro usable with the stream structure. But I'm not sure, >>>>> if I should show the "first line of text" or if I should use "caption" or >>>>> "subtitle". >>>>> >>>>> >>>>>> >>>>>> >>>>>>> *My personal hickups* ;) >>>>>>> >>>>>>> a) I did notice again, that I can't deal with hidden information. >>>>>>> a.1) I need the "stream-node-collapser" to be visible as soon as >>>>>>> child elements exist, similar to "toc-selective-expandable" >>>>>>> a.2) Waiting for the hover effect doesn't work for me. >>>>>>> a.3) I need the "collapser" to be in a different colour. >>>>>>> >>>>>> >>>>>> This is really just down to a design decision try and find balance >>>>>> between usability and clutter where possible, but you have a very valid >>>>>> point. I'd love to get more user feedback in order to establish a >>>>>> sensible >>>>>> default, while allowing this to be customizable. Since this is just CSS, >>>>>> it >>>>>> can be easily made into a config option that toggles some css in the >>>>>> stylesheet. >>>>>> >>>>> >>>>> At the DAT-CON I did a little bit of CSS tweaking >>>>> <http://dat-conf-2020.tiddlyspot.com/#CSS> between the video >>>>> sessions. ... So I could go on with taking notes ;) >>>>> >>>>> >>>>>> >>>>>> >>>>>>> b) I need the cheat sheet to be tagged $:/tags/SideBar so I can >>>>>>> have it open all the time in the sidebar >>>>>>> >>>>>> >>>>>> I'd actually love some help with tidying up that tiddler to improve >>>>>> readability. >>>>>> >>>>>> Regarding keyboard shortcuts, the bottom line is that they need to be >>>>>> made configurable since everyone willl have different needs and >>>>>> expectations. >>>>>> >>>>> >>>>> That's right. As I wrote, during a session it's "kind of" OK. .... But >>>>> the next day I tend to mess it up, at the beginning. >>>>> >>>>> >>>>>> >>>>>> >>>>>>> - Enter - save and create new node with text after cursor -> *For >>>>>>> me personally this functionality is frustrating without an Undo :/ >>>>>>> *Especially >>>>>>> in "refactoring mode" a day later. >>>>>>> >>>>>>> >>>>>> There is actually commented out code in Streams right now that >>>>>> handles an undo. I haven't made that active yet for two reasons: >>>>>> i) I want to add a confirmation both for splitting the text at the >>>>>> cursor, and for merging back with the previous node >>>>>> ii) I have an idea to separate out all such "word processor" features >>>>>> to a separate sub-plugin >>>>>> >>>>> >>>>> OK >>>>> >>>>> >>>>>> >>>>>> >>>>>>> - Alt+Enter - split longer text into multiple nodes. I need a >>>>>>> *confirmation >>>>>>> dialogue*, even if it is slower. .. Fixing the problem if >>>>>>> hitting this combination by accident is a 100 times more time >>>>>>> consuming >>>>>>> >>>>>>> >>>>>> A confirmation dialogue makes sense and will be added. >>>>>> >>>>>> Thank you again for the detailed feedback, I've added notes on to my >>>>>> local copy of the Roadmap and will push it to github once I've had the >>>>>> chance to tidy it up. >>>>>> >>>>> >>>>> You are welcome! >>>>> >>>>> -mario >>>>> >>>>>> -- 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/6b134fcc-e8bf-4879-bd7f-7db7c671429do%40googlegroups.com.
