garden/iap21.md

Iap21

• A workshop.
• by [underlay].
• [go] https://knowledgestructure.pubpub.org/iap21
  • [meta] https://anagora.org/go/iap21 redirects to the above URL due to the presence of this 'go' block.
• slides https://docs.google.com/presentation/d/1EFCzr6Wrur_QgRjB6ibCuJXHauYMGl_2oLrF0nEVEyM/edit#slide=id.p
  • [meta] https://anagora.org/go/iap21/slides and https://anagora.org/go/slides/iap21 both redirect to the above URL due to the presence of this 'slides' block in a note called 'iap21'.
• agenda https://knowledgestructure.pubpub.org/pub/agenda-21

[2021-01-25]

• [quote] exploration: structuring collective knowledge (developing a case study).
• [metasj] kicks off
  • A follow on to an experiment that happened previously on-campus.
  • goal by the end of this week, we should be able to describe what a satisfying collective experiment could be. Also, we'll have cross-pollinated.
• introductions
  • [metasj]
  • [agnes cameron] works with [metasj], part of knowledge collaboration innovation data. In uk.
  • samir ghosh runs a maker space. Does vr.
  • sarah s neuroscience and machine learning. crowdsourcing to probe human perception.
  • amr kayid munich research engineer at open mind. reinforcement learning.
  • thomas renkert religious studies knowledge representation.
  • kevin (pending)
  • favour kelvin
• recap of approaches to collective knowledge
  • [metasj] how much of the audience's work is about collaborative/collective effort?
    • samir ghosh used a mega google doc
    • [metasj] how fluid is the membership?
    • samir ghosh there's a core group of people, but the collective is nebulous.
    • [metasj] there's a tradeoff there.
• [metasj] presents slides (see above).
  • [knowledge futures group]
    • [pubpub]
  • goal
    • choose a problem that is bigger than what you could solve yourself :)
    • we can take inspiration from biology; organisms offer lots of patterns that are useful in aggregate
    • city construction
  • classes of models
    • you're probably within ~10km of a weather sensor part of a global network, all maintained by volunteers.
    • alignment vs standardization
    • access vs catalysis
    • private service vs common good
      • ancestry.com vs iceland (?)
    • feedback loops
    • incentives vs shared vision
  • examples from biology and earth science
    • [inaturalist] (!)
    • [geograph] (!)
      • had to come up with a local map projection not to change base assumptions they made early on :)
    • [open street map]
    • [ebird]
      • [agnes cameron] is a fan. Comments section is great :)
  • examples from the group
    • [flancian] on the [agora]
    • thomas renkert in some countries, people are assembling lists of people that died due to coronavirus
      • [nytimes did a story on the first 10k people that died.
    • sarah s on imagenet
      • collective in more than one way:
        • corpus based on human annotation
        • also building off and on several existing efforts, like wordnet
      • related problem: updating datasets, contexts.
      • uses of latent space
    • [agnes cameron] on [arena]
      • [push] [arena]
        • https://www.are.na/agnes-cameron/what-s-are-na
        • organized in channels
          • can be [open] and closed
        • [open source]
• inception practice
  • plus challenge for [wednesday]
  • think about [units of collaboration] / units of work
  • who are the [audiences]?
• other interesting references
  • protocols.io
  • shader toy
  • shader park
  • repl.it
  • [cyberfeminist index]
  • [quotebacks]
  • [how to make almost anything]
  • [free universal construction kit]
  • [movebank]
  • [wikidata subsetting]
  • corpse2corpse
  • [place subreddit]

[2021-01-27]

• [quote] inception: knitting the lacework - connecting sources, queries, and outputs
  • Working with public knowledge graphs, building new visualizations + overlays
• [agnes cameron] presented an example of underlay, interlay, overlay reasoning in the field of food sustainability.
• blake elias ai research covid response
  • is health vs economics a zero sum game?
  • analysis: you can optimize for two things
    • having the fewest covid cases
    • optimize for economics
  • it turns out that in simulation runs the two don't look that different.
    • A Poisson distribution:
    • future cases = random[(current cases) * (reproductive number) + (imported cases)]
  • aside: [epidemic forecasting]
  • aside: [worry dream]
  • aside: [covid tracker]
  • samir ghosh (feedback)
  • summary is: 'flattening' at a certain constant level (e.g. 100 cases/day) does not work as well as locking down hard and getting to zero (albeit temporarily)
  • [agnes cameron] [imperial college tracker]
  • [pandemic game]
• samir ghosh on [force directed graphs]
  • https://bl.ocks.org/vasturiano/02affe306ce445e423f992faeea13521
  • https://editor.p5js.org/generative-design/sketches/M_6_1_03o
  • http://yifanhu.net/GALLERY/GRAPHS/index.html
  • http://act.mit.edu/cavs/
  • https://www.ideaflow.io/
  • https://graphcommons.com/
    • https://www.youtube.com/watch?v=aNLmt7q-W2s
    • https://www.youtube.com/watch?v=5pdvWszMaLM
    • https://www.youtube.com/watch?v=2ZRCc-6tRZo
  • https://graphcommons.com/graphs/a936d1c7-2b52-47ba-bc21-79a2631149d0/selection/47f0218d-e5b1-9ea7-f748-dfa7e7810071
  • https://www.youtube.com/watch?v=5pdvWszMaLM
  • see [tuesday] agenda for more examples
• favour kelvin on animal tracking
  • https://ebird.org/about/resources/finding-birds-with-ebird
  • https://www.systemanaturae.org/projects/

[2021-01-29]

• [push] [underlay]
• [push] [interlay]
• [push] overlay
• [push] invite to [interlay]
  • BigQuery
  • PatCit
  • Historians
  • How things work / xkcd / tim urban
• [push] [2021-01-30]
  • [meet] [metasj]
    • [fediverse]
    • [pubpub] and [agora]
• [wikipedia]
  • [push] [flanbridge]
    • Possible [root nodes].
      • A root node in a subgraph is a good node from which with to optimally encode the surrounding space.
      • meaning of life
      • list of lists of articles
      • wikipedia articles have many rankings