• [[flancian@social.coop]] https://social.coop/@flancian/109932434330981165 pull

• a [[thing]].
  • [[distributed knowledge graph]].
  • The brain of an [[Agora commons]].
  • See [[agora doc]].
• a [[project]].
  • I want to graph both the Agora as a whole and the context of a node in the agora, as per [[agora ui]].
  • local graphs: done (partial implementation).
  • global graph: running in dev but very resource intensive, may need to refactor/redo.
    • it could be cool to get a graph in [[agora doc]] though…
      • could be nice to do this through [[runnable subnodes]]:
        • [[javascript subnodes]]
        • [[python subnodes]]
    • did [[rdf]] first: exposed rdf for the agora and use something like [[rdf grapher]] to graph
  • #go https://twitter.com/flancian/status/1380210191837315073 pull

• The [[agora]] graph is an [[interlay]] between the [[underlay]] and the [[overlay]]
  • [[agora server]] is an overlay
  • [[mastodon]] [[twitter]] etc are underlays

• [[flancian]] https://social.coop/@flancian/107444853975212659 pull
• [[flancian]] https://social.coop/@flancian/107445162306288613 pull

• [[flancian@social.coop]] https://social.coop/@flancian/109919401327330702 pull
• [[flancian@social.coop]] https://social.coop/@flancian/109926254795500411 pull

• [[flancian]] https://twitter.com/flancian/status/1470704871208804354 pull

• [[agora interlay howto]]

• [[flancian]] https://social.coop/@flancian/108381503392535871 pull

• [[flancian]] https://twitter.com/flancian/status/1530650384645201920 pull

Authored By:: [[P- Rob Haisfield]]

[[C- A structural editor can make a DSL approachable to end-users]]

[[C- A DSL would let people write in a certain syntax and notation that gets transformed by functions into a data structure that can be manipulated by pre-built or custom-built functions]]

[[C- End user programming enables people to bulk process notes]] and a DSL enables that.

Some examples might include using attributes in Roam to indicate key-value pairs, or [[being able to leave a hole]] for later input (like in algebra, you figure out what x is in x+3=11 after writing out the equation).

[[C- A DSL speeds up the author]]

Authored By:: [[P- Rob Haisfield]]

Search queries only work when the searcher knows the vocabulary used in the information being searched. If I were trying to find the answer to a question about cancer biology, I might search Google Scholar or PubMed and find nothing, while an expert would easily find the answer, simply because they know the vocabulary and I don’t.

[[Q- How do we solve the problem of different people referring to the same concept with different language]]? A key barrier to interdisciplinary synthesis is simply that people are unaware of the fact that people in other fields are asking the same questions with different language. This means that it will be important for the [[Search Behavior]] affordances to support people’s ability to learn the vocabulary of the fields they are querying.

See [[R- Information Foraging Video]] discussing Mr. Taggy as a potential solution.

Authored By:: [[P- Rob Haisfield]]

[[Q- Should social knowledge management be thought of as social networks with really solid defaults, conventions, and incentives]]

A semantic knowledge graph may not be enough - as discussed in [[Q- What are the most efficient routes to useful cross-boundary knowledge]], it is difficult to imagine a more efficient search for information than asking the right people what they think. However, [[C- Synthesis is hard to do with people who don’t share context with you]], and we need mechanisms to bring people up to speed. When someone begins to research a new domain, one of the most difficult problems is [[learning what vocabulary to use to search]].

In [[R- Interactive Intent Modeling for Exploratory Search]], they describe a search process where users type in a search term, are shown the results, and then are also shown other keywords they could add to the search, grouped by semantics and frequent use by experts. This is a low-touch social interaction - the experts don’t need to do anything but publish for the system to derive keywords and groupings from their work.

⥅ [[Interactive intent modeling search circle.png]]

A high touch social interaction would involve a search for experts in your domain of interest and then direct communication. As described in [[R- Graphs in OpenCog’s AtomSpace, Webstructor and Aigents for SingularityNET]], there should be a communication layer. Given how [[C- Hypertext enables communication with high information density]], using a semantic knowledge graph in conjunction with a social graph would enable people to communicate more clearly.

⥅ [[decentralized social evidence-based data model.png]]

• [[Q- Can the blockchain be used to improve citation chains]]: this question is relevant insofar as a big reason why people are willing to share knowledge with others publicly is that they have an expectation of credit coming back to them.

Work in Progress§

#WIP

This is because of [[Q- What changes in a discourse graph as quantity of content increases]], as there is an increasing need to reduce the overall quantity of content individuals are expected to consume.

[[C- Highlighted and lowlighted search results map to how well results map to intentions]]

This is about diffusing the effort. In [[R- Information Foraging Video]], they present some research on social tagging.

This supports answering the question of [[Q- What user behavior is required to maintain a decentralized knowledge graph]]

• [[flancian]] https://social.coop/@flancian/108068854164612537 pull

Contribution Graph

Contribution Graph

Contribution Graph

Contribution Graph

• a [[site]].
  • #go https://debategraph.org
  • Looks very interesting!
  • Heard about it from [[Jack Park]], thank you Jack!

decentralized discourse graph§

• a [[project]].
  • met on [[2022-01-10]]
  • didn’t take notes as it was being recorded.

Directed Acyclic Graph§

• Love these.
• [[pull]] [[dag]]

• [[flancian]] https://social.coop/@flancian/107481472664947749 pull

• [[flancian]] https://twitter.com/flancian/status/1473047975760904193 pull
• [[cognacore]] https://twitter.com/cognacore/status/1473281073773617153 pull

Distributed Knowledge Graph§

Distributed Knowledge Graph§

• A [[graph]] where [[nodes]] are [[entities]] and [[edges]] are [[social]], [[semantic]], [[logical]] [[relationships]].
  • Such a graph, in general, is by definition [[social]].
  • [[pull]] [[social-knowledge-graph]]
• [[push]] [[what is a distributed knowledge graph]]
• [[go]] https://notes.knowledgefutures.org/pub/belji1gd/release/2
• by [[joel gustafson]] of the [[knowledge futures group]].
• The [[agora]] is built around one, or is an interface to many, depending of how you look at it.
  • It is an [[integrated repository]] of [[digital gardens]].
    • The default exchange format is [[markdown]] on [[git]].
  • I strive to integrate it as well as I can with the [[federated ecosystem]].
    • It should eventually support in particular [[activity pub]], [[web annotations]], [[rdf]], [[agora protocol]].

knowledge is key .

We all reach a point in life mostly us who are less fortunate . its easy when your parents are wealthy and feed you all the way, one thing i can say is us who are under previlaged learn things in life in different and harder circumstances, this is an idea of change to create opportunities for those who are in that position . what can you do to put food on the table ! if jobs are decreasing by day and many others dont know how to actually bring in something and make a life from absolutely nothing. i am in this position as im writting this and yes you can change everything in a goood way BUT what does it actually take ? starting with mindset one needs to be opassionate or atleast interested in becoming something .

takes thinking and listening.

you can not be stubborn and know it all attitude .. im up 4 in the morning glued to my interests and passion an idea a dau with no pay is a push down . Today at age32 i can confidentky sat give me it all and i wi double it .. with confidence today is the last day i suffer financially theres a dosen ways to make it right in front of us

so I am editing someone else’s text here

knowledge is key .

We all reach a point in life mostly us who are less fortunate . its easy when your parents are wealthy and feed you all the way, one thing i can say is us who are under previlaged learn things in life in different and harder circumstances, this is an idea of change to create opportunities for those who are in that position . what can you do to put food on the table ! if jobs are decreasing by day and many others dont know how to actually bring in something and make a life from absolutely nothing. i am in this position as im writting this and yes you can change everything in a goood way BUT what does it actually take ? starting with mindset one needs to be opassionate or atleast interested in becoming something .

takes thinking and listening.

you can not be stubborn and know it all attitude .. im up 4 in the morning glued to my interests and passion an idea a dau with no pay is a push down . Today at age32 i can confidentky sat give me it all and i wi double it .. with confidence today is the last day i suffer financially theres a dosen ways to make it right in front of us

so I am editing someone else’s text here

knowledge is key .

We all reach a point in life mostly us who are less fortunate . its easy when your parents are wealthy and feed you all the way, one thing i can say is us who are under previlaged learn things in life in different and harder circumstances, this is an idea of change to create opportunities for those who are in that position . what can you do to put food on the table ! if jobs are decreasing by day and many others dont know how to actually bring in something and make a life from absolutely nothing. i am in this position as im writting this and yes you can change everything in a goood way BUT what does it actually take ? starting with mindset one needs to be opassionate or atleast interested in becoming something .

takes thinking and listening.

you can not be stubborn and know it all attitude .. im up 4 in the morning glued to my interests and passion an idea a dau with no pay is a push down . Today at age32 i can confidentky sat give me it all and i wi double it .. with confidence today is the last day i suffer financially theres a dosen ways to make it right in front of us

so I am editing someone else’s text here

knowledge is key .

We all reach a point in life mostly us who are less fortunate . its easy when your parents are wealthy and feed you all the way, one thing i can say is us who are under previlaged learn things in life in different and harder circumstances, this is an idea of change to create opportunities for those who are in that position . what can you do to put food on the table ! if jobs are decreasing by day and many others dont know how to actually bring in something and make a life from absolutely nothing. i am in this position as im writting this and yes you can change everything in a goood way BUT what does it actually take ? starting with mindset one needs to be opassionate or atleast interested in becoming something .

takes thinking and listening.

you can not be stubborn and know it all attitude .. im up 4 in the morning glued to my interests and passion an idea a dau with no pay is a push down . Today at age32 i can confidentky sat give me it all and i wi double it .. with confidence today is the last day i suffer financially theres a dosen ways to make it right in front of us

so I am editing someone else’s text here

• [[flancian]] https://social.coop/@flancian/108296429164624763 pull
• [[flancian]] https://social.coop/@flancian/108905734253751816 pull

• [[flaneur@merveilles.town]] https://merveilles.town/@flaneur/105723954881839758 pull

• [[flancian]] https://twitter.com/flancian/status/1524430875781484546 pull
• [[flancian]] https://twitter.com/flancian/status/1525205393697230848 pull
• [[kvistgaard]] https://twitter.com/kvistgaard/status/1525212560701771778 pull
• [[flancian]] https://twitter.com/flancian/status/1564201083202936833 pull

• [[2024-10-31T23:05:54.938Z]] @[[flancia.org]]: https://bsky.app/profile/did:plc:xu4fhdxwmrptiwfvq5ymbkxx/post/3l7tri6y6tn2r pull

Distributed Public Knowledge Graph§

• [[push]] [[distributed knowledge graph]]

• [[flancian]] https://twitter.com/flancian/status/1476957105009885185 pull
• [[flancian]] https://twitter.com/flancian/status/1500486495269969923 pull

type:: [[SPARQL Endpoint]], [[RDF Store]] url:: https://query.linkedopendata.eu triples:: 274 736 517 date:: [[Feb 14th, 2022]]

• [[Example]] queries
  • ▶ Try

    #Kohesio projects on a map
    #defaultView:Map
    SELECT ?KohesioProject ?location  {
      VALUES ?country {wd:Q23 wd:Q12} # Estonia, Denmark
      ?KohesioProject wdt:P35 wd:Q9934 ;
                      wdt:P32 ?country ; 
                      wdt:P127 ?location .
    }
    

    #Query #VALUES
  • ▶ Try

    SELECT ?countryLabel (xsd:integer(SUM(?EUcontribution)) as ?sum) {  
      VALUES ?country {wd:Q18 wd:Q14 wd:Q27 } # Portugal, Lithuania, Slovenia  
      ?KohesioProject wdt:P35  wd:Q9934 ;                  
                      wdt:P32  ?country ;                   
                      wdt:P835 ?EUcontribution .  
      SERVICE wikibase:label { bd:serviceParam wikibase:language "[AUTO_LANGUAGE],en"}
    }
    GROUP BY ?countryLabel
    

    #Query #VALUES #SUM #[[GROUP BY]]
  • ▶ Try

    SELECT * 
    { ?s wdt:P841 ?beneficiario ;
         wdt:P474 ?presupuesto ;
         wdt:P835 ?contribucion_eu ;
         wdt:P837 ?porcentaje_financiacion ;
         wdt:P270288 ?lugar ;
         wdt:P460 ?codigo_postal ;
         wdt:P1845 wd:Q2550039 ;
         wdt:P836 ?resumen ;
         wdt:P416470 ?id_spanish ;
         wdt:P20 ?fecha_inicio ;
         wdt:P33 ?fecha_fin ;
         wdt:P127 ?coordenadas .
         FILTER ( LANG(?resumen) = 'es' ).
                   }
    ORDER BY DESC(?contribucion_eu)
    LIMIT 10
    

    #Query

foam graph§

• asked in the Foam discord server about graph performance.
• [[riccardo]] "dataviz.ts is probably the best entry point to see how the graph comes together. I am also keen to provide the json output (e.g. via a command) to help with https://github.com/foambubble/foam/issues/654 and make it easier for the graph data to be used"

Force Graph§

• a [[library]]
  • [[javascript]] [[d3]]
  • [[wp]] https://en.wikipedia.org/wiki/Force-directed_graph_drawing pull
  • [[go]] https://github.com/vasturiano/force-graph
    • [[graph drawing library]] in use by [[logseq]]: https://twitter.com/logseq/status/1322933174964428801 pull
  • I think it makes sense. Seems reasonable fast at the high-thousand-nodes level; the agora is currently at 10k. And most of the visualizations I want to do are actually node-scoped plus links, so really just an issue for hugely outlier nodes degree wise (like [[person]] or [[go]]).
  • Unfortunately it doesn’t seem to provide labels for edges?
    • False, it does: https://vasturiano.github.io/force-graph/example/text-links/
  • Data format, JSON:

{
    "nodes": [
        {
          "id": "id1",
          "name": "name1",
          "val": 1
        },
        {
          "id": "id2",
          "name": "name2",
          "val": 10
        },
        (...)
    ],
    "links": [
        {
            "source": "id1",
            "target": "id2"
        },
        (...)
    ]
}

• [[flancian@social.coop]] https://social.coop/@flancian/109919421801678082 pull

• [[flancian]] https://twitter.com/flancian/status/1520679524333150209 pull

Geograph§

• [[go]] geograph.co.uk

Giant Global Graph§

• [[pull]] [[ggg]]
• A [[concept]] by [[tim berners lee]].
• [[go]] https://web.archive.org/web/20160713021037/http://dig.csail.mit.edu/breadcrumbs/node/215
• [[foaf]]

Hypergraph§

• a [[thing]].
  • a [[graph]] with [[annotated edges]], a graph with edges which connect an arbitrary number of nodes.
  • #go https://en.wikipedia.org/wiki/Hypergraph pull
  • An [[Agora]] is a [[hypergraph]].
  • [[jonathan]] https://golem.ph.utexas.edu/category/2018/02/hypergraph_categories_of_cospa.html
    • "There are several notions which fall under the name “hypergraph”, and we do not follow conventions too closely here. For us, a hypergraph is simply a kind of graph which has different types of nodes and edges, with edges allowed to be possibly “open” in the sense that they are attached only on one end to some node. Composition of such graphs to build new graphs is possible by connecting open edges together (though we only allow edges of the same type to connect). The following picture illustrates such a composition, with different types of edges indicated by different types of lines."

• [[flancian@social.coop]] https://social.coop/@flancian/109932429439205525 pull

Interactive Philosophers Graph§

knowledge graph§

a knowledge base that uses a graph-structured data model or topology to integrate data.

– Knowledge graph - Wikipedia

• [[2026-02-03T17:33:05.770Z]] @[[velitchkov.eu]]: https://bsky.app/profile/did:plc:ij6qgzlhgjg2fie4lgt4inmz/post/3mdxveivquk2v pull

• [[jonny]] https://social.coop/@jonny/108499314740858740 pull
• [[tanepiper@tane.codes]] https://tane.codes/@tanepiper/109602100445737544 pull
• [[tanepiper@tane.codes]] https://tane.codes/@tanepiper/110210434112143922 pull

• [[tanepiper@tane.codes]] https://tane.codes/@tanepiper/109944362457094209 pull

• [[tanepiper]] https://tane.codes/@tanepiper/109268225048582891 pull
• [[tanepiper]] https://tane.codes/@tanepiper/109285429471392784 pull
• [[bengo]] https://mastodon.social/@bengo/109316628228767396 pull
• [[tanepiper@tane.codes]] https://tane.codes/@tanepiper/110249186811262876 pull
• [[tanepiper@tane.codes]] https://tane.codes/@tanepiper/110536351540516808 pull
• [[tanepiper@tane.codes]] https://tane.codes/@tanepiper/110537414940882260 pull
• [[tanepiper@tane.codes]] https://tane.codes/@tanepiper/110538645046136193 pull
• [[tanepiper@tane.codes]] https://tane.codes/@tanepiper/110736111024201161 pull

• [[2022-11-10 16:21:17+00:00]] @[[kausch]]: https://twitter.com/kausch/status/1590741392258895872 pull

• [[2022-11-10 16:21:17+00:00]] @[[kausch]]: https://twitter.com/kausch/status/1590741392258895872 pull

RT @heikopaulheim: There’s #KnowledgeGraph #Embeddings for #LinkPrediction (Trans* etc.) and KGE for Data Mining (RDF2vec etc.) - are they…

• [[RoamFm]] https://twitter.com/RoamFm/status/1480747717370417154 pull

Namedgraph§

• A [[person]] on [[twitter]]. Working on knowledge graphs.

open graph§

• [[pull]] [[open graph protocol]]
• [[go]] ogp.me

• [[smallcircles@social.coop]] https://social.coop/@smallcircles/110011634140768978 pull

Personal Knowledge Graph§

• [[go]] https://research.google/pubs/pub48375/
• The [[agora]] is a [[distributed knowledge graph]] assembled out of opportunistically cross-linked personal knowledge databases.
• [[roam]], [[foam]], [[obsidian]], [[athens]] and other [[roam likes]] have fostered communities around creating them.
• They are sometimes known as [[digital gardens]].
  • The [[agora]] is a repository of [[digital gardens]].
• [[pull]] [[digital garden]]
• [[pull]] [[pkm]]

[[P- Chris Granger]] claims that showing more than 15-20 nodes at a time makes a graph view unmanageable.

[[Codex OS]] has proposed an autocomplete that prioritizes the relationships for browsing:

⥅ [[CleanShot 2022-05-21 at 15.43.59.png]]

Expanding on that idea, I believe that, similar to Finder’s view of nested folders (Miller Columns), we could let people traverse node—>edge—>node—>edge relationships with simple arrow commands. So in the above example, it might look like you select the Michaelangelo option and it shows you the relationship options from there, you select the relationship, then select a node from within there, and on…

Loose mockups below:

⥅ [[Pasted image 20220521154950.png]]

⥅ [[Pasted image 20220521155159.png]]

Authored by:: [[P- Rob Haisfield]]

This is an incredibly tricky problem.

One common solution is to expect that users update their writing into a new shared structure at the point where they choose to publish. If structure is missing, [[C- It will be important to capture the potential energy of information consumption]], enabling the broader user base to update it and add in new structure. To be able to rely on readers structuring publications post-hoc, we’d likely require incentives, which are largely out of scope for our research.

However, [[C- There is a wealth of creative exhaust generated by researchers that is going to waste]] and to leave that behind as is done in the above solution would be a shame. Currently, [[P- Joel Chan]] is running a journal club (everybody reads papers on a given topic) using the [[Discourse Graph Plugin]]. Everybody participating in it uses/used Roam Research at a high level, and therefore developed their own idiosyncratic style. Some people prefer to write information directly onto the claim, question, and evidence pages, while others prefer to do so in thoughtfully indented outlines. Through the plugin, we are able to write Datalog queries that describe each of our styles such that we end up with the same discourse relationships regardless of how they are written. This works well for a small group, but may not work well on a global scale. For that, it might prove necessary to segment the queries so certain rules apply to different authors.

The takeaway to the above solution is that there can be an underlying data structure separate from the visual presentation layer. As long as individual user behavior is consistent within documents, queries can identify that structure such they compile into the appropriate data structure.

Authored By:: [[P- Rob Haisfield]]

The design of a tool for thought’s primitives and its UX will influence the behavior of its users. For example, [[Roam]] designed its indentation and bidirectional link filtration to facilitate a Zettlekasten.

When I realized that the way that filtering/queries work with indentation can lead users to nest blocks in basically the same way as a folgezettel I was like “@Conaw you dog.” Tangential to what @vgr is describing, but just shows how Roam excels at embedding philosophy in design twitter.com/vgr/status/134… Tweeted on [Jan 4, 2021](https://twitter.com/RobertHaisfield/status/1346197182383177729 pull)

⥅ [[Pasted image 20210916180319.png]]

Backlinks are displayed by as a collapsible list. The backlinks do not just display the titles of the pages where a backlink is present: they also show breadcrumbs of context from what came before the relevant block. see [[I- Search as a primitive]]

While you are looking through backlinks, you can filter them down by including and excluding items, which are sorted by their frequency.

⥅ [[Pasted image 20210922151354.png]]

One could also argue that outliners push people towards divergence. In Roam, you can branch your thoughts, but it can be difficult to group them together. Sibling nodes are unrelated to each other, so it can be difficult to query for a block that contains a list of items nested beneath it. In its current form, it can only query one branch at a time.

Take a look at the backlinks for [[Roam Research]] and [[Obsidian]]. You can expand and collapse the list of results in each.

⥅ [[expand-collapse-roam.gif]]

When there are too many results, you can filter to narrow the results

⥅ [[Pasted image 20210914182735.png]]

⥅ [[Pasted image 20210914182805.png]]

Users will have a primitive to write [[inline data structures]]

My current thinking is that the most valuable set of primitives are basically: typed concepts, typed relationships, expressive user-generated data structures, and rules. I want to be able to write in blocks with inline data structures, arbitrary relationships between blocks, block referencing, inline querying, and inline coding.

NLP, similar to [[Jump]]‘s NLU, will enable users to implicitly instantiate complex mental models into the data structure. Expressive data structures (like triplets and nested tables) will enable people to manually specify structure when necessary. They should be expressive enough that many different visualizations can be made on top.

Authored By:: [[P- Rob Haisfield]]

As discussed elsewhere, one of the most important questions to creating a decentralized discourse graph is the following: [[Q- What workflows and behaviors facilitate synthesis]]? We also acknowledge that [[C- People are lazy]], so if we can split up the work of synthesis, we should.

If we were to divide up the work of synthesis by roles instead of jobs, below are some possible roles:

• Curator
  • Curates the fire hose of information down for people to keep up with the discourse in less effortful ways
• Connector
  • Consumes information across many domains and forms new connections, addressing the question of [[Q- How do we solve the problem of different people referring to the same concept with different language]]
• Educator
  • Understands the discourse and educates others so they can more meaningfully participate. Yannic Kilcher is a great example.
• New knowledge provider / creator
  • Contributes new information and beliefs to push the discourse forward. This may come from individuals or teams within the field doing novel research or reasoning, or from idea brokers bridging knowledge from one domain to another.
• Debater / contrarian
  • Makes sure people don’t get too caught up in a consensus that is unsubstantiated.

This is one of our open questions going into [[user interviews]]. #questions/core

Authored By:: [[P- Joel Chan]], [[P- Rob Haisfield]], [[P- Brendan Langen]]

Our research draws from a long line of information models, such as the Semantic Web Applications in Neuromedicine ([[SWAN]]) ontology, the [[micropublications]] model, the [[ScholOnto ontology]] for modeling scientific discourse, the [[nanopublication model]], and the Hypotheses, Evidence, and Relationships ([[HypER]]) model. These models share a common underlying model for representing scientific discourse: they distill traditional forms of publication down into more granular, formalized knowledge claims, linked to supporting evidence and context through a network or graph model.

We use the term discourse graph to refer to this information model, to evoke the core concepts of representing and relating knowledge claims (rather than concepts) as the central unit, and emphasizing linking and relating these claims (rather than categorizing or filing them). Standardizing the representation of scientific claims and evidence in a graph model can support machine reasoning.¹ We are particularly interested in the [[potential of discourse graphs to accelerate human synthesis work]].

So what does it mean for a discourse graph to be decentralized?

Synthesis has many component parts.[^2] Given that, [[C- The responsibilities required to produce synthesis can be split up among many people]]. [[P- Michael Karpeles]] refers to this concept as [human computation](https://twitter.com/mekarpeles/status/1440886235917164546 pull?s=20)

• Some people will contribute primary research.
• Some will formalize the research contributed by others into frameworks and theories. [[C- Incrementally processing notes is a key user behavior to promote synthesis]]
• Some will help determine what directions are meaningful to explore, ranking the utility of the outputs of others.
• Some will simply read everyone else’s outputs and annotate them, all the while meaningfully connecting the conversation between the various fields they explore. [[C- It will be important to capture the potential energy of information consumption]].
• Some will rate the contributions of those who annotate to improve the signal to noise ratio. As such, a key question as we go into the interview phase of our research will be: [[Q- What community roles are necessary in a decentralized knowledge graph]]?

Dividing the responsibilities of synthesis is one of the core strengths of decentralizing the process. When we look at prior attempts at building a semantic web, we find that the primary reasons have to do with human behavior and dishonesty.

[[C- People are lazy]] and [[C- Most people will primarily consume information]], so we can’t expect them to do all of the work necessary to index information themselves. While tools like [[Roam Research]] and [[Obsidian]] enable people to develop advanced discourse graphs for themselves, over time they may end up with a system that is so complicated that maintaining it becomes work. Why not split up the effort so the ones who share information aren’t responsible for 100% of the processing?

[[Q- What user behavior is required to maintain a decentralized knowledge graph]]? Presumably, there should some balance between what is required of user behavior, what is required of the design of the system, and what AI can do to reduce the effort from user behavior.

2022-04-12 reflections:

• Today in random things brought to my attention on Twitter ( @balOS @joelchan86 you’ll both get a kick out of this), Tau is a blockchain decentralized social network that uses logic based AI (similar to [[Jump]]) and program synthesis from user discussions to make updates to its codebase, allowing all participants in the discussion forums to participate in Tau’s governance.
• Supposedly they’ve figured out a way to use logic programming also to identify points of consensus so it isn’t updating its code all over the place.
• When you listen to them talk, you can hear them make an implicit point: if web3 is in part about being able to participate in the governance of the products/platforms you use, unless you expand programming capability to everybody, it’s ultimately governed by the developers who make decisions about what to build. End-user programming is necessary to self / community governance.
• https://twitter.com/TauChainOrg/status/1513777139451142146 pull?s=20&t=gVmIRmtS-DT3Cw_pDhf7Aw
• I asked on Twitter about the conceptual fit between query languages like SQL and Datalog with personal and shared knowledge graphs and they responded.
• Makes me also think - is there more of a conceptual fit between different types of AI and discourse graphs? I.e. logical AI (Jump, Tau) or probabilistic AI (GPT-3, DALL-E 2)?
• I would bet that logical AI is a better fit because it’s fully explainable. In the context of a discourse graph, you want to know exactly how it built up its answer. Architectural transparency is key.
• Also, a logical AI might be able to do more from the relatively small amount of data produced by individual researchers in their own graphs. From our interviews, people seem to want AI that is tuned to their personal knowledge graphs, not necessarily from a global body of knowledge.
• On the other hand, a probabilistic AI might be a better fit if you consider the "zettelkasten as a conversation partner" to be most important.
• Additionally, with the discourse graph generated by thousands or millions of participants, idk which would be more effective - a probabilistic or logical AI.
• I know a guy who has built a library (can’t find it) called Frankenstein that is trained to look at black box probabilistic AI results and fabricate propositional logic explanations. Spitballing, a cool model might be:
  • Probabilistic AI to identify logical relationships from natural language
  • Logical AI built on top of that
  • Probabilistic AI to answer questions from a global discourse graph
  • Frankenstein to interpret those answers in terms of logical relationships and check to see if they are correct or where they might have logical breakdowns

Thoughts from [[P- Ryan Murphy]]:

In language processing/AI-ish stuff, I want to see:

[[P- Ryan Murphy]]:

• Find similar/related documents (as per DEVONthink)
• Identify missing links (e.g., highlight important words/tokens that seem to appear significant and relevant elsewhere in your notes, too)
• Suggested tags … I thought I had more, but I’m forgetting ‘im

[[P- Joel Chan]]: love the idea of missing links! maybe a more powerful/fuzzy variant of aliases.

suggested aliases for the current note. Would be super valuable. Prevents people from talking about the same thing using slightly different terms in different places

[[P- Ryan Murphy]]: actually, and that’s a fifth (or maybe the same one): redundancy detection. "These two notes are so similar that you might want to merge them"-type use cases you are always in control, but the system can help you with your tasks: finding things you want to merge, propagating changes, detecting missing links, etc.

https://g6.antv.vision/en/largegraph [[Q- How could a graph database structure be visualized and perceived by end users in a more helpful way than a large graph]]

Notice how it creates clusters that show the number of items within that cluster as a [[ZUI]]

⥅ [[CleanShot 2022-05-21 at 15.40.02@2x.png]]

[[2021-04-14]]

The overarching goal of the Underlay project is to create a distributed public knowledge graph

… underlying all such developments is the core idea of using graphs to represent data… The result is most often used in application scenarios that involve integrating, managing and extracting value from diverse sources of data at large scale … we view a knowledge graph as a graph of data intended to accumulate and convey knowledge of the real world, whose nodes represent entities of interest and whose edges represent relations between these entities

Graphs don’t always have a ton of structure to them, and in fact, allow you to postpone a schema’s definition. [[Q- How do people come to agree on queryable schemas]]

Graphs allow maintainers to postpone the definition of a schema, allowing the data – and its scope – to evolve in a more flexible manner than typically possible in a relational setting

[[Q- How do people come to agree on queryable schemas]]

“Schemas”, if they exist, may be ad-hoc and unenforced, which often results in data that feels “half structured”. This is not a sign of poor quality or problematic incompleteness, but rather a core distinguishing characteristic of knowledge graphs.

Knowledge graphs are often assembled from numerous sources, and as a result, can be highly diverse in terms of structure and granularity. To address this diversity, representations of schema, identity, and context often play a key role, where a schema defines a high-level structure for the knowledge graph, identity denotes which nodes in the graph (or in external sources) refer to the same real-world entity, while context may indicate a specific setting in which some unit of knowledge is held true

[[Q- What would be a protocol that would allow various protocols to communicate with each other]]

Simply describing the technical structure of knowledge graphs misses a key aspect, which is where the data comes from and how the database is maintained. ==Knowledge graphs are graph databases that come from the integration of data from many sources and in many domains.== In other words, you make a knowledge graph by pulling data about lots of different kinds of things together in one place. This is part of what might justify calling them knowledge graphs, since it involves ==parsing out latent common structure behind “mere data” as it exists already, and aligning that common structure in a single space.== A knowledge graph is a graph that has somehow unified or reconciled the entities and relationships shared between (previously) separate data sources.

[[Q- What would be a protocol that would allow various protocols to communicate with each other]] [[Q- How do people come to agree on queryable schemas]]

Obviously not all entities and relationships are shared, but intuitively we feel that many are: an employee in a company directory can be treated as the same class of thing as a city resident in a census. ==Within each original dataset, the two will have different set of properties, but an external human curator can reasonably infer that they refer to the same type of real-world entity.==

[[Q- What user behavior is required to maintain a decentralized knowledge graph]]

Lastly, there’s a continuous, living, process-oriented aspect to knowledge graphs:

==… effective methods for extraction, enrichment, quality assessment, and refinement are required for a knowledge graph to grow and improve over time==

This isn’t about schemas in the technical database sense. ==Two databases might have the exact same schema, but still represent slightly (or wildly) different things.== The real thing that integration accomplishes is more like ontology alignment, an operation on abstract structures that just happens to be expressed as a mapping from one database into another. Typically we don’t work with or even think about explicit ontologies - we just rely on natural-language property names and general context instead - but it’s useful to imagine they exist because it helps explain why integration is so notoriously painful. The complexity and similarity of the actual schemas involved isn’t even that relevant; what matters is how well the underlying concepts align.

To illustrate this, suppose that we have a general-purpose knowledge graph full of entities like people, places, and things. ==Let’s say we also scrape GitHub to build a collaboration graph of GitHub accounts, and now we’d like to integrate this into our knowledge graph so that we can query for collaborator relationships between people. We’re immediately faced with the distinction between people and user accounts, which are not quite the same thing.== People might have several accounts - maybe different ones for personal and work projects - and any account might be used by more than one person.

There are two basic strategies that we can apply to align these ontologies.

One option is to just ignore the difference and merge the two classes based on email, full name, or whatever we have, and add the new “collaborator” relationships directly between people in the knowledge graph. This keeps the graph as simple as possible - we’re just adding data and not reorganizing anything - but the sacrifice we make in doing so is that we distort the data we’re integrating. People and user accounts are simply different kinds of things, and treating them as the same will give us extra or missing links due to the mismatch between the underlying ontologies. This effect could be called dilution or compression or erosion.

Our other option is to make separate entities for each user account, and then relate them to the associated people whenever we can. Maybe we’d even introduce a new “Agent” class, with both Person and User Account as subclasses, plus a “represents” relationship between Agents. Then maybe we add some inference rule about “if Agent A represents Agent B, Agent C represents Agent D, and Agent A and Agent C are collaborators, then Agent B and Agent D are collaborators too”, so that we can query for collaboration relationships between people like we wanted. Here, we’ve been more faithful with our alignment, but only at the expense of accumulating complexity in the process.

In practice, we usually end up using a combination of the two when integrating new data, leaving us with a result that is (oddly enough) both slightly simplified and slightly more complex than the data that we started with. This trade-off is unavoidable; the real value proposition of knowledge graphs is in the balancing of these extremes. ==If we apply too many simplifying assumptions, we’ll wind up with data that bears so little resemblance to its sources that it is effectively useless. On the other hand, if we accumulate too much complexity, we’ll get a knowledge graph that is so convoluted that it becomes effectively unusable.== The best we can do is shoot for a middle ground that’s reasonably expressive and reasonably manageable.

And there is some theory that we can try to apply. [[Category theory]] in particular is rich with many of these themes, like qualified relationships and compositional mappings. And sure enough, lots of applied category theory work revolves around databases, schemas, integration, ontologies, and knowledge representation - sometimes very abstract, sometimes shockingly practical. Things are only theoretical until they’re not.

Humans are so adept at context-switching that we give ourselves the illusion of having a single big ontology. Our goal is to build a large-scale data system that is so adept at context-switching that it gives the illusion of being a knowledge graph.

• term I’m using to describe a deeply linked set of things that tells a story by the nature of their relations

• [[dougholton@mastodon.social]] https://mastodon.social/@dougholton/109461051821775680 pull

Social Distributed Knowledge Graph§

• Essentially what the [[agora]] is, or tries to be.

• My definition: a [[distributed knowledge graph]] that is produced by, and embeds, a [[social network]].
  • Sometimes just a synonym for [[distributed knowledge graph]].
  • [[pull]] [[social-knowledge-graphs-discussion]]

• [[flancian@social.coop]] https://social.coop/@flancian/111563951724737021 pull

• [[flancian]] https://twitter.com/flancian/status/1560992279074152448 pull

• #pull [[the agora is a social knowledge graph]]

H2 {The Agora is a Social Knowledge Graph} [[Flancian]]

In this chapter we describe an Agora, being a social knowledge graph provisioned and maintained by a community as a commons.

The Agora differs from other projects in the knowledge space in a few ways: whereas a personal knowledge graph usually contains resources authored or collected by a single person, and a wiki usually contains resources produced by a group, an Agora contains and integrates both personal and group resources and interlinks them liberally. Whereas links in a personal knowledge graph or wiki usually have a single target, Agora links fan out by default and can be thought of as mapping to sets of resources. Finally, the reference Agora tries to remain tool, format and platform agnostic, trying to build first on general conventions common to many tools and platforms in the knowledge space for maximal inclusivity and diversity.

Being a graph, an Agora can be defined as a set of vertices or nodes N (entities) and edges E (known links between entities, optionally annotated). An Agora node contains the set of all known resources about or otherwise relevant to the entity described by the node title or any provided metadata. Each such resource is called a subnode. Note that because links can be arbitrarily annotated (i.e. #tagged or qualified by other nearby links) and have multiplicity, the Agora is in fact a hypergraph.

The free and open source reference Agora provides a minimum viable implementation of the underlay, interlay, overlay components of a distributed knowledge graph based on off-the-shelf components. Individual Agora instances are expected to federate and organize into a greater Agora network. We describe how this network can integrate with the wider internet ecosystem and how it could be used to run experiments on distributed thought.

• an [[agora abstract]].
  • #push [[agora abstracts]]
    • #pull [[social knowledge graph]] [[distributed knowledge graph]]
    • An Agora is a [[distributed knowledge graph]] provisioned and maintained by a community as a [[commons]].
      • The Agora can be said to be a [[social knowledge graph]] as it is a [[distributed knowledge graph]] produced in a social context and containing social information.
      • Whereas a [[personal knowledge graph]] usually contains resources authored or collected by a single person, and a wiki usually contains resources produced by a group, an Agora contains and integrates both personal and group resources.
      • Whereas a [[personal knowledge graph]] is usually maintained using a single [[personal knowledge management]] tool and stored in a single format, an Agora tends to be tool and format agnostic, trying to provide and follow the most general conventions.
    • Being a [[graph]], an Agora can be defined as a set of vertexes or [[nodes]] N (entities) and [[edges]] E (known links between entities, optionally annotated).
      • Agora [[nodes]] are defined by the set of known resources about the entity described by their title or other metadata.
        • A [[node]] is a community-maintained [[collection]] of voice-preserving individual [[subnodes]] defined by the resources contributed by a certain user or group.
      • Because links between two nodes in an Agora can be [[annotated]] (i.e. #tagged or qualified by other nearby links) and have multiplicity, the Agora is in fact a [[hypergraph]].
      • Individual agoras are expected to federate and organize into greater Agora networks, which are in themselves graph-like at a higher level.
    • On a system level, there exists a [[free]] and [[open source]] [[reference Agora]] that provides a minimum viable implementation for the [[underlay]], [[interlay]], [[overlay]] components of the [[distributed knowledge graph]].
      • In the reference Agora, links can be said to [[fan out]] by default in the sense that they are evaluated in social context in individual contributions (resulting in following a link sometimes surfacing more than what the individual author envisioned.)
      • We are using said reference Agora to refine the proposed system and run experiments.
      • Some hypotheses that we are testing:
        • Individual contributions can be made maximally useful to others on average when served best-effort in a social context at the only cost of adopting a default social stance (at little extra effort over baseline), and this mechanism benefits from network effects.
        • A [[knowledge commons]] model can provide utility to participating communities efficiently, as the cost of systemic integrations with a hub design such as the Agora can scale with O(N) instead of the O(N^2) provided by a naive full mesh.
        • Best effort social [[composition]] and [[integration]] of notes might be sufficient to surface yield higher level meaning and order, or at least significantly complement both taxonomic approaches (hierarchical) and individual-scoped eventual convergence (non hierarchical).
          • A composition of personal [[hierarchies]] yields a social [[heterarchy]].
          • The social context afforded by an Agora provides a path towards faster eventual convergence (that is, K people contributing to N nodes as categorized using loose or implicit personal ontologies can converge on useful emergent categories as K increases, at low individual effort).
        • Going from a set of voice-preserving [[individual contributions]] to a shared [[group resource]] might be an efficient way to foster opportunistic collaboration at scale.

H2 {The Agora is a Social Knowledge Graph} [[Flancian]]

In this chapter we describe and analyze a device we call an Agora, being a distributed knowledge graph provisioned and maintained by a community, ideally as a commons. This Agora can also be said to be a social knowledge graph as the information it contains is produced in a social context and is contributed with a social intent.

The Agora stands out from other projects in the knowledge space in a few ways: whereas a personal knowledge graph usually contains resources authored or collected by a single person, and a wiki usually contains resources produced by a group, an Agora contains and integrates both personal and group resources and interlinks them liberally. Whereas links in a personal knowledge graph or wiki usually have a single target, Agora links fan out by default and can be thought of as mapping to sets of resources. Finally, whereas knowledge graphs are usually maintained using a single tool in a particular format and hosted in a centralized platform, the reference Agora tries to remain tool, format and platform agnostic, trying to build first on the most general conventions common to tools and platforms in the knowledge space for maximal inclusivity and diversity.

Being a graph, an Agora can be defined as a set of vertices or nodes N (entities) and edges E (known links between entities, optionally annotated). An Agora node contains the set of all known resources about or otherwise relevant to the entity described by the node title or any provided metadata. Each such resource is called a subnode.

Note that because links can be arbitrarily annotated (i.e. #tagged or qualified by other nearby links) and have multiplicity, the Agora is in fact a hypergraph.

On a system level, the free and open source reference Agora provides a minimum viable implementation of the underlay, interlay, overlay components of a distributed knowledge graph.

Individual Agora instances are expected to federate and organize into a greater Agora network.

In this chapter we might also describe said reference Agora further and go into how we are using it to iterate on the system design and run experiments.

Some hypotheses that we are exploring follow:

• A knowledge commons can provide utility to participating communities efficiently, as the cost of systemic integrations within a pro-social hub design such as that of the reference Agora can scale with O(N) instead of the O(N^2) provided by a naive full mesh (N being the number of integrations, e.g. tools or platforms connected).
• Going from a set of voice-preserving individual contributions to a shared group resource might be an efficient way to foster opportunistic collaboration at scale.
• Social loose linking and opportunistic heterarchical categorization might complement traditional taxonomic approaches better than the same tried at the individual level by virtue of network effects. This might enable convergence on meaning and other emergent behaviour.

H2 {The Agora is a Social Knowledge Graph} [[Flancian]]

In this chapter we describe and analyze a device we call an Agora, being a distributed knowledge graph provisioned and maintained by a community, ideally as a commons. This Agora can also be said to be a social knowledge graph as the information it contains is produced in a social context and is contributed with a social intent.

The Agora stands out from other projects in the knowledge space in a few ways: whereas a personal knowledge graph usually contains resources authored or collected by a single person, and a wiki usually contains resources produced by a group, an Agora contains and integrates both personal and group resources and interlinks them liberally. Whereas links in a personal knowledge graph or wiki usually have a single target, Agora links fan out by default and can be thought of as mapping to sets of resources. Finally, whereas knowledge graphs are usually maintained using a single tool in a particular format and hosted in a centralized platform, the reference Agora tries to remain tool, format and platform agnostic, trying to build first on the most general conventions common to tools and platforms in the knowledge space for maximal inclusivity and diversity.

Being a graph, an Agora can be defined as a set of vertices or nodes N (entities) and edges E (known links between entities, optionally annotated). An Agora node contains the set of all known resources about or otherwise relevant to the entity described by the node title or any provided metadata. Each such resource is called a subnode.

Note that because links can be arbitrarily annotated (i.e. #tagged or qualified by other nearby links) and have multiplicity, the Agora is in fact a hypergraph.

On a system level, the free and open source reference Agora provides a minimum viable implementation of the underlay, interlay, overlay components of a distributed knowledge graph.

Individual Agora instances are expected to federate and organize into a greater Agora network.

In this chapter we might also describe said reference Agora further and go into how we are using it to iterate on the system design and run experiments.

Some hypotheses that we are exploring follow:

• A knowledge commons can provide utility to participating communities efficiently, as the cost of systemic integrations within a pro-social hub design such as that of the reference Agora can scale with O(N) instead of the O(N^2) provided by a naive full mesh (N being the number of integrations, e.g. tools or platforms connected).
• Going from a set of voice-preserving individual contributions to a shared group resource might be an efficient way to foster opportunistic collaboration at scale.
• Social loose linking and opportunistic heterarchical categorization might complement traditional taxonomic approaches better than the same tried at the individual level by virtue of network effects. This might enable convergence on meaning and other emergent behaviour.

H2 {The Agora is a Social Knowledge Graph} [[Flancian]]

In this chapter we describe and analyze a device we call an Agora, being a distributed knowledge graph provisioned and maintained by a community, ideally as a commons. This Agora can also be said to be a social knowledge graph as the information it contains is produced in a social context and is contributed with a social intent.

The Agora stands out from other projects in the knowledge space in a few ways: whereas a personal knowledge graph usually contains resources authored or collected by a single person, and a wiki usually contains resources produced by a group, an Agora contains and integrates both personal and group resources and interlinks them liberally. Whereas links in a personal knowledge graph or wiki usually have a single target, Agora links fan out by default and can be thought of as mapping to sets of resources. Finally, whereas knowledge graphs are usually maintained using a single tool in a particular format and hosted in a centralized platform, the reference Agora tries to remain tool, format and platform agnostic, trying to build first on the most general conventions common to tools and platforms in the knowledge space for maximal inclusivity and diversity.

Being a graph, an Agora can be defined as a set of vertices or nodes N (entities) and edges E (known links between entities, optionally annotated). An Agora node contains the set of all known resources about or otherwise relevant to the entity described by the node title or any provided metadata. Each such resource is called a subnode.

Note that because links can be arbitrarily annotated (i.e. #tagged or qualified by other nearby links) and have multiplicity, the Agora is in fact a hypergraph.

On a system level, the free and open source reference Agora provides a minimum viable implementation of the underlay, interlay, overlay components of a distributed knowledge graph.

Individual Agora instances are expected to federate and organize into a greater Agora network.

In this chapter we might also describe said reference Agora further and go into how we are using it to iterate on the system design and run experiments.

Some hypotheses that we are exploring follow:

• A knowledge commons can provide utility to participating communities efficiently, as the cost of systemic integrations within a pro-social hub design such as that of the reference Agora can scale with O(N) instead of the O(N^2) provided by a naive full mesh (N being the number of integrations, e.g. tools or platforms connected).
• Going from a set of voice-preserving individual contributions to a shared group resource might be an efficient way to foster opportunistic collaboration at scale.
• Social loose linking and opportunistic heterarchical categorization might complement traditional taxonomic approaches better than the same tried at the individual level by virtue of network effects. This might enable convergence on meaning and other emergent behaviour.

H2 {The Agora is a Social Knowledge Graph} [[Flancian]]

In this chapter we describe and analyze a device we call an Agora, being a distributed knowledge graph provisioned and maintained by a community, ideally as a commons. This Agora can also be said to be a social knowledge graph as the information it contains is produced in a social context and is contributed with a social intent.

The Agora stands out from other projects in the knowledge space in a few ways: whereas a personal knowledge graph usually contains resources authored or collected by a single person, and a wiki usually contains resources produced by a group, an Agora contains and integrates both personal and group resources and interlinks them liberally. Whereas links in a personal knowledge graph or wiki usually have a single target, Agora links fan out by default and can be thought of as mapping to sets of resources. Finally, whereas knowledge graphs are usually maintained using a single tool in a particular format and hosted in a centralized platform, the reference Agora tries to remain tool, format and platform agnostic, trying to build first on the most general conventions common to tools and platforms in the knowledge space for maximal inclusivity and diversity.

Being a graph, an Agora can be defined as a set of vertices or nodes N (entities) and edges E (known links between entities, optionally annotated). An Agora node contains the set of all known resources about or otherwise relevant to the entity described by the node title or any provided metadata. Each such resource is called a subnode.

Note that because links can be arbitrarily annotated (i.e. #tagged or qualified by other nearby links) and have multiplicity, the Agora is in fact a hypergraph.

On a system level, the free and open source reference Agora provides a minimum viable implementation of the underlay, interlay, overlay components of a distributed knowledge graph.

Individual Agora instances are expected to federate and organize into a greater Agora network.

In this chapter we might also describe said reference Agora further and go into how we are using it to iterate on the system design and run experiments.

Some hypotheses that we are exploring follow:

• A knowledge commons can provide utility to participating communities efficiently, as the cost of systemic integrations within a pro-social hub design such as that of the reference Agora can scale with O(N) instead of the O(N^2) provided by a naive full mesh (N being the number of integrations, e.g. tools or platforms connected).
• Going from a set of voice-preserving individual contributions to a shared group resource might be an efficient way to foster opportunistic collaboration at scale.
• Social loose linking and opportunistic heterarchical categorization might complement traditional taxonomic approaches better than the same tried at the individual level by virtue of network effects. This might enable convergence on meaning and other emergent behaviour.

• [[flancian]] https://social.coop/@flancian/108229311879252738 pull

• [[flancian]] https://twitter.com/flancian/status/1520909882790227968 pull

The Next Social Graph:§

Metadata§

• Author: [[tals.substack.com]]
• Full Title: The Next Social Graph:
• Category: #articles
• URL: https://tals.substack.com/p/the-next-social-graph

Highlights§

• In a network of attribution, we aim to add to and acknowledge the work of others, intentionally driving economic and reputational value to instrumental work, precisely in order so that others will consider doing the same for us. In this system, we want others to build on what we have built. The result is an ever-climbing tower of creativity, where ideas flow up and gratitude flows down.
• The practice and culture of sharing ownership and credit can change the incentives of creation.
• Through an architecture of composable ideas, we can add to and amplify each other’s efforts that build a tower of innovative work, solidly established on the recognition and reward of its foundations.
  • Tags: [[social media]] [[composable]]

Trust graph§

• a [[graph]] expressing [[trust]] relationships
  • [[nodes]] are of three types
    • a [[person]] making a trust statement
    • the [[target]] of a trust statement, which can be an [[artifact]], a [[person]], or an [[institution]]
    • the [[reason]] for trusting the target
      • [[expert judgment]] ("I understand this article and I consider it sound.")
      • trust in the [[creator]] of an artifact
      • the [[reputation]] of the target
      • …
  • a trust relationship is a [[semantic triple]] of the form "A trusts B for reason C"

• [[2022-04-22 20:48:09]] [[@vera:fairydust.space]] (link):
  • Saw this [[trustgraph]] project from [[bluesky]] discord https://trustgraph.net/ looks super cool

• [[flancian]] https://social.coop/@flancian/108339528965907155 pull

• [[flancian]] https://twitter.com/flancian/status/1527963935823089664 pull

• a [[tool]]
  • [[graph database]] [[knowledge engine]]
  • by [[sophia xu]]
  • [[git]] https://github.com/unigraph-dev/unigraph-dev
    • [[go]] [[git]]
    • [[typescript]]

[[go]] https://github.com/unigraph-dev/unigraph-dev

• [[flancian]] https://social.coop/@flancian/107480611281406096 pull
• [[flancian]] https://social.coop/@flancian/107542780772573387 pull
• [[flancian]] https://social.coop/@flancian/107837503894384293 pull

• [[flancian]] https://twitter.com/flancian/status/1467433403339489281 pull
• [[flancian]] https://twitter.com/flancian/status/1472993223870136337 pull
• [[flancian]] https://twitter.com/flancian/status/1473406435480514564 pull
• [[flancian]] https://twitter.com/flancian/status/1476971592915861517 pull
• [[flancian]] https://twitter.com/flancian/status/1495833751019999235 pull

What Is a Distributed Knowledge Graph§

Wikigraph§

• [[go]] giantgra.ph
• [[nopush]] [[agora plan]] add as search target
  • Cancelled as https://giantgra.ph no longer works as of [[2024-08-01]] :(