[07:04:03] 790 [07:04:16] 9790 [15:30:41] Newsletter #119: Rockefeller Foundation supports Abstract Wikipedia with $1 Million - https://diff.wikimedia.org/2023/07/12/abstract-wikipedia-gains-new-support-from-the-rockefeller-foundation/ [16:14:00] Congratulations! Excited to learn what this new funding will unlock! Are there more ambitious plans that are now possible? Or is it still too early to tell? (re @wmtelegram_bot: Newsletter #119: Rockefeller Foundation supports Abstract Wikipedia with $1 Million - https://diff.wikimedia.org/202...) [16:15:50] It's paying for the existing plan and development, so that we use less of the general donations for that. [16:17:22] Oh, so the donation could have been to the WMF in general and the result would have been the same? (re @vrandecic: It's paying for the existing plan and development, so that we use less of the general donations for that.) [16:17:45] Thanks for the quick response. That's good news, here's hoping that the freed up funds can be directed to, say, the community tech / wishlist team 😁 [16:17:56] I thought you couldn't control where specifically your donation to the WMF went? (re @Jan_ainali: Oh, so the donation could have been to the WMF in general and the result would have been the same?) [16:21:17] For large donations it has been possible. Like the Sloan foundation grant a few years back aimed at Commons. (re @mahir256: I thought you couldn't control where specifically your donation to the WMF went?) [16:23:36] Correct. These are so-called restricted funds and can be used only for certain things. They have the disadvantage that they can be repurposed, they have the advantage that they can't be repurposed. It's always about perspective :) [16:26:46] That's also the answer to Jan's question. I do think these donations and the exact type of donation make a big difference. [16:29:17] I guess allowing for earmarking makes it possible for the donor to brag about what their money enables, even if it in reality turns out to be paying for 1/175th of everything :) [16:38:55] It's more than that, but it's difficult to capture that in writing off the cuff. [16:40:16] I was able to get more resources because of the funding, which made things go faster. Stay tuned, there's more news upcoming on that in the coming weeks :) [16:44:03] And not just faster but also to cover more scope that wouldn't be possible otherwise. [16:47:59] Wait, how does that fit together with "It's paying for the existing plan"? [17:01:45] The existing plan for Abstract Wikipedia. That's very large in scope as you know. With these funds we can cover more of that plan than what we could have covered otherwise. [17:02:55] Hmm, does that mean that some parts of the plans does not have coverage? (re @vrandecic: The existing plan for Abstract Wikipedia. That's very large in scope as you know. With these funds we can cover more of that pla...) [17:03:47] Yes, just as with the original Wikidata plan, which still has parts that were never implemented (eg Phase 3) [17:04:56] I think it’s still somewhere in the plan / pipeline ^^ [17:05:12] (but I’m the wrong person to talk about Wikidata’s longer-term plans, I only have a vague idea of them) [17:05:29] Oh, I thought WMF was in it for the long haul for Abstract Wikipedia. (re @vrandecic: Yes, just as with the original Wikidata plan, which still has parts that were never implemented (eg Phase 3)) [17:06:11] There's a task for Phase 3 on Phab, and I was just discussing it with Lydia yesterday. It's not forgotten. [17:37:09] Honestly, the support from the WMF has been amazing. I have no doubt they are in this for the long run. Nevertheless, resources are limited - and there are a few unknowns on the way, which are quite independent of resourcing. (re @Jan_ainali: Oh, I thought WMF was in it for the long haul for Abstract Wikipedia.) [20:15:28] I am curious about the schema for Abstract Content. [20:15:28] [20:15:29] At my workplace we are thinking about using a knowledge graph to capture learnings that occur in project teams. An example learning would be: [20:15:31] [20:15:32] "_Visualise current problem spaces and goals and update them regularly to gain a common understanding._" [20:15:34] [20:15:35] Each learning should be enriched with the context in which it occurred so that if a project is in a similar context in the future, potentially relevant learnings pop up. [20:15:37] [20:15:38] I am thinking about what kind of schema/ontology would be helpful for this. An upper ontology like gist maybe, that already contains some concepts. [20:15:40] [20:15:41] And then I started wondering how Abstract Wikipedia is planning to represent Abstract Content. After all, if you found a way to represent anything that can be said as Abstract Content, the representation of learnings should be a tiny subset of that? I had a look at the examples for Abstract Content + rendering you have on this (https://meta.wikimedia.org/wiki/Abstract_Wikipedia/E [20:15:43] xamples) site. But it seems to me, there are different concepts how to do this ? It also states that the look of the constructors will be up to the communities. Will there be a core concept of how to model Abstract Content that will be fix? And if so, is that sort of done already or still in development? I am wondering about terms like `Instantiation`, `Object_with_modifier_and_o [20:15:44] f `or `Relational noun` for instance. Are, or will they be flowing from an ontology/schema that governs how Abstract Content will be modelled? [20:15:46] [20:15:47] Basically I am wondering if it could make sense to harvest learnings as Abstract Content so that queries like this become possible: "_Find all learnings that occurred in moments of change where risk taking was necessary and paid out in the end_". I imagine it could yield interesting sets of results that would otherwise be impossible to find together in a result set. Querying abst [20:15:49] ract content would work almost like finding bundles of analogies in some way. [23:49:11] The current draft about the planned Abstract Representation as done by the NLG working group, which consists mostly of volunteers, is here: https://meta.wikimedia.org/wiki/Abstract_Wikipedia/Wikidata_Abstract_Representation [23:51:56] I do think that something in the direction you describe will be possible, but for now it is a bit "Zukunftsmusik" ( Q228730 / L685149 ). But I hope that something like this will become possible.