Metasj

• #pull [[samuel klein]]
  • a [[friend]].
  • I’d like to introduce him to:
    • [[benedict lau]]
    • [[vera]]
    • [[lorand]]
    • [[guo]]
    • [[paul frazee]]

[[2022]]

• It’s amazing knowing [[sj]]!

[[2021-07-01]]

• [[cascading]] searches
  • ahs search prefix
  • https://oadoi.org
    • non API version that just works: see telegram
    • API seems good though
  • it should take any version of a DOI indicator, either in URL form or not
  • you want it to return the URL of a world readable PDF
• [[sociofesto]]
• [[space]]
  • [[weather]]
  • [[debris]]
  • [[helios]] group is its own data collective, US based
• how to assemble good collective ontologies
• [[interlay]] vs [[interlace]]
  • [[interlay]]: the universe of what’s available around a topic. how to deal with reputation, visibility, governance.
    • has the feel of a search engine. search engines were the first interlays :)
    • fast feedback loop, discoverability
    • we have stopped at the second rung of a high ladder
    • [[q]] if there is a governance layer, is it in the interlay? I believe so
      • all layers could have governance components
  • example: an underlay that is a star catalogue
    • a single collaborative dataset
    • social aspects
    • needs coordination, e.g. how to do timestamps; how to sequence observations; how to handle a variety of sensors. these all yield rules that could be interpreted as governance.
  • [[interlace]] is a multiverse; a superset. doesn’t have governance.
    • negative space for interlay
      • as you develop reputations, governance, integration approaches for an interlay you may run across things where a decision isn’t clear and it feels restrictive (once made) — you know that one way to address it is to fork
      • pun on interlays
  • have you thought about exchange formats / interop standards
    • interlay: a collection of statements with some default schema
    • a standard schema could be: naming a source and describing its properties
      • aside: catalogue like
    • unsure now because people that care about granular knowledge and compile huge datasets get easily distracted about format questions
      • better to define it in terms of a namespace perhaps
      • it has at least a human readable name, perhaps an autogen machine readable name
      • describes which formats are available
        • structured data
        • human readable text

[[2021-05-22]]

• ahoy!
• [[wikidata hackathon]] used [[big blue button]]
• [[push]] [[ahs]]
  • [[metasj]] dedicating more time to it going forward
  • new pitch: trying to simplify
  • a [[locus of focus]]
  • a [[legal]], [[summarized]] and [[annotated]] repository of [[knowledge]]
  • a [[distributed knowledge graph]]
• [[interlay]]
  • [[hypha]]
    • [[distributed press]]
• [[commons]]

[[2021-05-05]]

• an [[intent]] for [[cooperation]]
• [[codex]]
• a node for every [[doi]]
• [[gen studio]]
• a canonical entity plus a conversation
• from [[go/meet]] notes You have two webcams.

The government takes one webcam and gives it to your neighbor.

https://en.wikipedia.org/wiki/You_have_two_cows pull

https://anagora.org/metasj pull for previous notes

===

[ hub ]

[ legal ]

[ perserverant ]

[ comprehensive ]

[ summarized and annotated ]:

a space for [[synthesis]]

===

Modularity:

• Underlay of sources + pipelines;

• Interlay of aggregations and catalogs;

• Overlays for search, reading, annotation

Interlay:

1. Namespace for (articles, concepts, authors)

• To start: 1M entities, write on read?

[identify 100 disciplines; top X historical + current articles, their authors + cites]

Bootstrap:

• get the 1m entities from somewhere

• get X users to show contribution methods by example.  Good candidates:

• reading groups paper/book clubs

• grad students surveying their discipline

• informal/out of system students

• citizen scientists

Units of contribution:

1. choose a discipline, build the "list of 100 historical articles / topics" in it, find sources for it [sync w fatcat]

2. mark papers as [[state of the art]]

3. assemble papers into a dependency graph

4. run or join a [[reading club]] focusing on an area, a book or a set of papers.

===

[[synthesis]]

[[hypha]] coop

IL: A healthy subspace for prosocial entities to outperform walled gardens —> here that maximizes your utility and connection.

Deustche Bank famously noted how Elsev is benefitting from a distorted market. (The same applies today to app stores…): https://notes.knowledgefutures.org/pub/supertanker/release/3

https://www.connectedpapers.com/

===

obsd —> github —> agora

(obsidiangit)

[[2021-04-30]]

• [[importance]]
• [[focus]]
  • for the focus to be sharper, X needs to happen
• [[context]]
• [[push]] [[ahs]]
  • instead of trying to figure out how to solve tangential complexity, a platform to focus on what’s the focus of our disciplines
  • a platform for [[scholarship]]
  • what do we need?
    • [[names]] for its components
    • a [[mapping]] from a [[goal]] to all known resources that are known to contribute to the [[goal]], or the other way around
      • this feels like a network plus a [[flow]] problem
    • [[scripts]] that iterate over the data set and take actions
      • reminds me of [[intent based automation]] and [[declarative paradigm]]