i thought i understood, but now i am crippled by hamletic doubt. is it an unlimited number of votes?
is the vote count separate in different boards? so we o ly got a certain number of votes in the video game board?
First, how I believe it’s configured, and second, about the quest board…
Each tier, of which there are five, 0-4, has a different amount. I think it is like, 20, and then 10 more for each tier above that. Something arbitrary, to see how it worked. Which I didn’t, because like, we don’t have any data to look at. I don’t really know how, um, four of us use the voting. 
As far as I know all categories on a Discourse instance share the same voting pool. From the discussions and glancing at pseudocode I believe the voting mechanisms are transactional, just updating a field in user profile. So when it renders a vote button for you, it checks if you have enough votes. When you vote, or a post closes, it subtracts or adds to the counters for all the relevant users.
Okay, okay, that’s just how it works. I’m not sure I am gonna use the quest board in that way any longer. I mean, maybe, but also differently, because I’m moving into a new system, a new engine. talkgroup will remain an integral part of the discussion, but I think there are some better tools to use for what I’m going for.
Okay, what am I going for? Well, you know how we like that weighted, redundant, looping dependency model Phabricator uses to sort the order of tasks that should be done? I really love that. Almost enough to use Phab. But while Phabricator has a lot of interesting tools put together in a particular way, it doesn’t do any of it’s things as well as I want.
So I’ve built those things: talkgroup for discussion storage, mageparty for group and instant chat.
The next step is to build the data storage and processing piece. And because part of the quest board idea was for emergent interest to come forth and draw group attention, that fits the same weighted priority model. I’m looking at building a Semantic MediaWiki-based data engine. Because the data on talkgroup is CC0 and sliced and diced into JSON all over the place, I can automate capturing our conversations and assign semantics to them, which we can then individually generate our own priority lists.
If you are following along at home, this may seem super creepy, but it should demonstrate how easy it is to scrape public data, but also fun and useful ways to do so!
For instance, we have a lot of categories. The way I use #links is to capture knowledge shared abroad, but provide a small note to self. Now I’ve started just saving the Wayback URL, and I might look into creating a plugin for Discourse that does that across the site. See, I can’t stand most sites. Tracking bullshit. But it turns out most documents on the web are hardly more useful than a bullet list of points made, and of course the discussion it generates between people authentically looking for truth. Hence talkgroup becomes a layer, a type of data/knowledge cache. It is the thing I called social documentation, or how knowledge is generated via discussion.
With this new engine in place, I can feed it my links shared, directly from talkgroup. But it isn’t just a feed of topics: each topic has metadata, such as other topics linking to it, tags, mentions, replies, votes, archive/closed status, all that stuff. I can use all those values to create queries and reports that can drive my curiosity in the way I want. But more so, I can show everyone else what I see.
I’m also interested in connecting conversations by using a slightly different model for editing data. No hard rules, but it normally plays out that most edits are from fewer people, which makes having a separate database a pain for drive-by edits; under the presumption that the age of anonymous wiki editing has passed, it means I don’t expect to have a lot of direct contributors.
But I’m gonna slurp up data from where it already exists, and take a federated approach to communicating. I’ll incorporate different “Talk” links into the site, to move conversations into email and jabber and potentially other federated systems (have we figured out how to quick link to a Mastodon account?). And I can create a category here in mailing list mode, so we’ll actually have those talk conversations here in a better conversation tool.
I’m still compiling my notes for this, and I promised @tim I would unveil the next phase next week. This serves as a good preamble.
1 Like
I like the idea of autogenerating links that cross post to the forums from the wiki. So, say a page is missing information, so block of facts, we can have a “start a quest!” link, and it will make a post to the quest board, and auto-fill in some details, like what it already says, and what it needs to know.
So that category will remain, just not in the way I had been describing. Or again, maybe it will. This isn’t really something I’m too concerned about at the moment.
I updated the amount of votes each person has, by role level.
If you join the site, you get 2 votes! Wow! Just a taste, but truly a participant in this strange machine of ours!
Everyone else gets 100 votes, except for trust level 4: they get 101, because they are a little more influential than you. Join their ranks and siphon their power!
This both answers how many votes each person gets, and why my other levels were silly: if I want people to participate in a weird weighted opinion/interest system, I can’t make the currency to play scarce. Now we all essentially have 100 things we are tracking, and hopefully that creates direction. 
1 Like
I’m almost out of votes. As I will not be able to vote any longer, I will mute Quest Board, so I can’t see any more quests. Then I’ll only interact with my quests through my QuestBoard8000 (working name).
1 Like
Another question! I thought that when a Quest was marked as Solved, then the votes get returned to people. Is that true? Is it just that my “votes” board includes Solved Quests that I’d previously voted for? I guess I could count them up but I thought I would ask because lazyweb
If you are asking because you hit your limit, solved something, and expected to get a vote back: I believe some math has to be done, and that is queued; meaning it takes a while for the votes to reflect back. Caching applies.
If that is not why you are asking, and it did not answer your questions, please rephrase.
nah i was looking at my list of votes because i started getting warnings that i had 9 votes left. my list of votes included stuff that was solved.
I don’t know much about this, but the way the votes work on a database level always surprised me, when I read about. Later I’ll look up what I mean and point it out.
1 Like