stack

install agora bridge api
write [[chasing moloch]]
write [[iremos juntos hasta el final de la noche]]
[[pull]] [[do]]

📓 garden/neil/stack.md by @neil ️👁 ️📝

Stack

The geek-colloquial meaning of stack, in the most relevant sense, is a set of interoperating hardware and software.

– [[Governable Stacks against Digital Colonialism]]

Benjamin H. Bratton (2016, xvii) goes further, describing the stack as “a new architecture for how we divide the world into sovereign spaces”.

– [[Governable Stacks against Digital Colonialism]]

A stack might include all that enables one to use a social media service, for instance: the server farms, the corporation that owns them, its investors, the software the servers run on, the secret algorithms that analyse one’s data, the mobile device, its accelerometer sending biometric data to the server farm, the network provider, the backdoor access for law enforcement, and so on.

– [[Governable Stacks against Digital Colonialism]]

The layers of a stack might further include the sun or coal powering it, the wars fuelled by rare-earth mining, and the mythologies and rituals that dictate what people in it will tolerate.

– [[Governable Stacks against Digital Colonialism]]

The urgency of forming [[governable stacks]] stems not only from tech giants and surveillance states. Groups dedicated to racism and authoritarianism have become particularly intentional about their network stacks, migrating to dedicated platforms such as Stormfront, Parler, and Gab as more mainstream networks remove them (Bev- ensee 2020). Will governable stacks become the speciality of reactionaries? The stack has already become a field of conflict, like the spinning wheel.

– [[Governable Stacks against Digital Colonialism]]

⥅ node [[do]] pulled by user

📓 garden/flancian/do.md by @flancian

a [[list]].
- [[irregular]].
  - An always incomplete collection of my [[intents]] :)
  - #pull [[todo]] [[now]] [[next]] [[some day]]
- [[meditate]] now.
- process inbox (pushes to this node).
- configure [[probing]]
  - I never get to this, can you tell it reminds me of work? :)
  - I need to just set a pomodoro and do this, but I feel like running the agora at low slo is OK. It's supposed to rely on distribution later on for reliability gains.
  - Doesn't really apply as an argument to [[moa]] though -- although for [[moa]] we have [[sentry]] set up so that seems reasonable.
    - set up [[probing]] for [[moa]]
  - -> As of 2024 moa is no longer supported, we shut it down when Musk closed down the API. I intend to bring it back, maybe for [[Fediverse]] <-> [[Bluesky]] -- but [[Ryan Barrett]] solved that with [[Bridgy Fed]]. So now I'm back to trying to make it work for Twitter again, even if it means paying, and running an instance for a closely knit community.
    - set up [[probing]] for [[anagora]]
  - Hmm, I should still do this :)
  - I also need to do something similar for [[social coop]] ideally, so maybe I could try to solve both at a time.
- [[push]] [[agora]]
  - implement [[auto pull]]
  - fix toggle switch for [[auto pull]]
    - this refers to /settings which has a toggle
    - hmm, maybe I should just embed the settings page in the burger. that also frees up /settings for the actual node about [[settings]] :)
    - I already have the burger menu working in a branch, which is nice
    - implement a toggle in the burger menu :)
    - implement [[auto push]]
      - probably requires [[agora protocol parsing]] to make efficient/reasonable
  - ...I wrote at some point in time. Now it's implemented, both when you say #push [[example]] and when you use a colon.
  - So I'll think about it but maybe this is done-for-now.
    - client side pulling
      - done for: mastodon, twitter, wikipedia
      - need to test general URL pulling, see how many sites allow iframes
        
        test with flancia.org
    - This is done :)
    - improve [[agora graphs]]
      - I like what logseq does: show link direction as particles. push/pull could be different particles. hmm. particle diagrams. check out [[feynman]] for inspiration?
  - this is still a good idea honestly, and I always want to work on graphs - could get inspiration from some diagrams in [[a rosetta stone]]
    - [[wikilinks everywhere]] -> [[agora ext]]
      - I really want to move wikilinking to on-demand (push button/ctrl-shift-l) or cronjob-like behaviour but right now it just wikilinks everything every time you click, and some sites break
        
        this is already better after a quick PR to linkify on click instead of mouse movement
        
        twitter still sort of breaks though
        
        [[vera]] is on this though
    - [[agora server]]
    - parse agora protocol
      - recognize (match) and style differently
- [[push]] [[agora interlace]]
  - [[agora social media integration]]
    - write [[agora bot]] for [[mastodon]]
    - write [[agora bot]] for [[twitter]]
      - fix auth
        
        restore context
        
        this could be a great workaround for the worse of the re-replying behaviour, and probably will be needed even if I want to support zero state best effort
      - fix deduping
        
        need to do this again
    - test [[vera]]'s code for pulling tweets and toots
      - only pulled some for some reason
      - figure out what's up, try again and get the context back
  - [[agora bot]]
    - fix twitter so it doesn't keep re-replying
    - the API doesn't feel great, I wonder if I'm missing something
      - I thought I'd done this but it's still failing in some cases, [[s5bug]] gave me a test case, I think I put it in [[agora twitter bug]]
    - refactor mastodon<->twitter so they share code? I don't want to implement everything twice, and I certainly don't want the mastodon version to suffer because I'm addicted to twitter (great people there as well, and there are more of them)
  - [[push]] [[agora bridge api]]
    - build/release [[agora bridge api]] for the storing posts in the agora.
      - we want to create a git repository for each calling users, which the users could then [[claim]].
      - could be flask, included in [[agora bridge]] repo, copy/paste from [[agora server]] to begin with?
        
        nope, it is typescript thanks to [[vera]], looking forward to set this up :)
      - but hopefully they can just communicate through yaml? or perhaps set up and write to sqlite to begin with
      - thought of consolidating everything into [[agora server]], but it feels cleaner to have writers and readers split. should scale better; if the write path goes down, the agora keeps serving just fine without having to do anything special.
        
        probably keeps it simple to run a read only agora (mirror).
    - yes, going with [[read]] ([[agora server]]) and [[write]] ([[agora bridge]]) apis
    - request new git integrations (e.g. mount git repo X in path Y)
    - accept other inputs? like perhaps subnodes, which could go into a managed repo.
    - configure sources from social media (from agora bot)
    - in general this unlocks better [[signup]]
- [[push]] [[moa]]
  - review [[go/moa/bugs]]
- [[slay moloch]]
  - [[chase moloch]]
- [[write]] - [[flancia]]
  - [[book]]
    - choose [[top 5]] nodes to expand on
    - edit https://flancia.org
  - [[iremos juntos hasta el final de la noche]]
  - [[slaying moloch]]
    - [[building bridges]]
    - [[finding isomorphisms]]
      - measuring distance to an isomorphism? there must be a term for this
  - [[node club]]
- [[patterns]]
  - [[after the pandemic]], or [[an open letter to the agora]], or [[an open letter to an open nation]]
  - [[on cringe]] / [[agora polls]]
    - about explicit rational (mathematical) modelling of group dynamics through social media discourse
    - is [[hack the planet]] cringe? should it be?
    - why is it cringe to talk about virtue?
- [[read]]
  - what [[will walker]] told me about
    - link it here
    - this is past due
- [[push]] [[weekly]]
  - run [[weekly review]]
  - it turns out I do a run through [[do]] weekly at best as of late, so this is my weekly. I'd love to make it daily though. perhaps I could auto-pull [[do]] from the journal page? just the subnode for the 'active' user.
  - promote [[do]] to nav item?
  - we could also have 'this week', 'this month', 'this year' in the agora, have it rank top nodes in time scope, allow users to pull specific nodes from each view
an [[agora action]]
- I try to automate recurrent [[actions]] in the [[agora]]. This node is the root of my [[getting things done]] graph, which I try to use to model my [[intents]].
- [[daily]], I try to check here for pending [[work]] in all known sources of truth.
- [[weekly]], I try to run [[weekly review]] and [[garbage collect]]
- [[monthly]], I try to check the context in more specific actions/nodes and push relevant items here or to [[next action]]:
- what should I [[read]] next?
- what should I [[write]]?
  - [[flancia]]
    - [[book]]
    - [[an open letter for an open society]] (this has many titles)
  - to [[a. p.]] asking for a recommendation
    - hmm, I don't know what this means :) as in, I don't know who I meant by [[a. p.]]
- [[unblock]] people who are waiting for me
  - this is not atomic. how do I know who is waiting for me?
[[research]]
- [[federated wiki]]
- [[indieweb]]
- [[ipfs]]
  - [[ipld]]
- [[fission]]
- [[hypercore proto]]
- [[launchlet]]
- [[urbit]]
  - [[rosano]]
    - Rosano is in Urbit or did I just fat finger an indent?
- [[support]]
  - [[effective altruism]]
  - [[flancia collective]]
  - [[open source]]
  - [[friends]]
  - [[public utility]]
- [[code]]
  - [[agora server]]
    - add particles to [[agora search]]
    - [[browse as]]
    - parse [[agora protocol]] (see below)
  - [[wikilinks everywhere]] / [[agora ext]]
    - could [[d3]] replace [[jquery]]? it would make it easy to also include a graph tool directly in the library. it's 200kb though.
    - but we could just add [[force graph]], in general have the client side code from [[agora server]] be also there and potentially work in every agora/every place with wikilnks or links (pulling could be useful on every site)
  - a more flexible [[actions]] system
    - what does this mean? :)
    - write something better or remove on next collect
  - [[project snapshot]]
    - model my digital existence as a sequence of computational snapshots
    - what complexity class is this in generally?
    - what is the [[zero]]? as in, the starting state.
    - probably [[browser tabs]]
    - then more generally programs running, including optionally their context
      - could make it easier to actually dedicate one workspace for every project -- the dream of [[richard francis burton]] IIRC?
Please tell me what you think I should [[do]] next (or instead of the things I'm doing or planning to do) below in the Stoa, via social media, or via any of the channels listed in [[flancian]]. All feedback is welcome!
- https://twitter.com/flancian/status/1386048603496529925
- https://twitter.com/flancian/status/1374462566395617286
[[push]] [[done]]
- implement [[ack]]
- [[agora yaml]]
- [[agora plan]]
- [[rdf]]
  - [[graphs]]
- [[agora search]]
  - implemented a provider
  - added support for [[yubnub]] like particles
  - [[branch]] agora-server/branch
  - [[push]] [[done]]
- implemented support for [[obsidian attachments]] in [[agora server]]
  - basic but it does the job (tm)
- many items in [[agora plan]]