I posted to G+ asking for feedback on the state of tank. Got a variety of responses that deserve some attention, linked and responded to below. There are have also been some new issues created and some updates.
Luke
@lukec makes an interesting comment:
Though I haven't done wiki style writing for a long time now, since working on a lot smaller team.
This suggests that wiki style writing is for teams of some girth. This conflicts a bit with @BillSeitz might say about PersonalNotebooks and what I might say about the value of stocks in small teams as the locus of synthesis.
Jon
@jayfresh had a few questions, some of which were likely rhetorical or Socratic.
What does "use global tags in autocomplete" mean?
It's an awkward turn of phrase, but the tag interface provides completion based on either the tags used in the current tank or all tanks. So, if the box is checked you get all. I could figured out a good phrase, so just put something in until some feedback led to something better.
Why is it called a tiddler? Do tiddlers float about in tanks?
Very Socratic this one. Jon is not the only person to ask some version of "why are you continuing with the term tiddler?" It has associations for people who have previous experience; associations which may not be true in Tank. For people who have never heard the term it is likely confusing.
I guess the main answers are: sentimentality and that's what they are called in the code. I do want to use a term that is not "page" because that works against the compositional nature of tiddlers and the ideas about associated metadata.
But if it is going to be used, it does need better introduction or explanation.
...is probably a pretty powerful thing, but so far it's not entirely clear to me what it's for
You may be looking for too much. The blurb on the frontpage does pretty much cover the top layer: It's a wiki with a nice API. The Features info gets much of the second layer: it has {trans,extra}clusion, robust authorization handling, notifications.
And then there is a third layer: You can use the API and the storage to build your own stuff, either via comps or Single Page Apps or (because of CORS support) stuff elsewhere.
wonder why the content from the comp isn't loaded into the comp app in some way
I'm not really able to parse this. Comps can load content but they don't have to. Both of the TiddlyWiki comps lost content after loading the app. Same for the bookmarking comp. See, for example, my bookmarks.
What am I missing?
The Euge
@eekim provides tons of great feedback, most of it linked from Tank Is Alpha. Predictably a goodly chunk of it surrounds granular addressability in Tank Granular Addressability. Within is the suggestion:
However, why not simply implement structural identifiers?
Perhaps but doesn't this work against the intentionally dynamic nature of text which is in a wiki. In Augment wasn't much of the text journalling and thus fairly write once-ish? One of the things I've always wanted to ensure is that content in wikis grows and changes in reaction to the synthesis that is going on around and with it.
Part of the reason for seeking feedback on this issue is that I don't mind the in-markup identifiers, I'm just well aware that other people do. In my perfect universe everybody would be attaching [nid fad4] all over the place.
(The complexity of this discourse is a good sign of why granularity would be handy and I think something will happen in tank at some point, I just want to be sure to go to the right place.)
In Tank Is Alpha the granularity of access control is mentioned as a cool feature. I agree, I think it is likely far more useful than is immediately obvious. TiddlySpace intentionally hid a lot of that power to remove some complexity. This worked well for some folk but was very limiting for others.
Not mentioned there is Privateer which mints unique URIs for accessing single private tiddlers (read-only) without auth. It needs a UI.
Many of other comments have been addressed either in the issues or in responses on the G+ posting. The one missing thing is:
Page revision history / comparisons
This is coming soon. It's a matter of not having built it in yet. Revisions are stored and there is a tool for TiddlyWeb which can show diffs between versions. I'm slow to do this because of lack of interest on my part, but I understand that it is important to some (hi @FND) and why it is important.