spencer's cafe

Search

SearchSearch

Working in Public

Jan 10, 2024, 16 min read

Author:: Nadia Eghbal Tags:#media/book

  • themes::
    • the question of what role people play and what powers and repsonsibilties they have > “What defines an outsider in open source, when any developer is supposed to be able to participate? If everybody is a potential contributor, who gets to make, enforce, and follow the rules?” ^2svWdC1ut
    • consensus-seeking over consensus processes ^9C0j3KlC5
    • “Information wants to be free” from Stewart Brand but also “information wants to be expensive” ^iJvm41vsV
  • Summary::
    • a deep-dive into the open source community and the various factors and incentives at play and an exploration about how it could evolve to be sustainable with the massive load places on maintainers currently
  • notes::
    • introduction + background ^XAmhyIA93
      • What even is open source? Originally designed to be open and a dogma for pushing forward openness (license required software using it to be open too)
      • Conception that it’s a bunch of communities but actually handful of very influential individuals that are more like “directing air traffic” from a lot of low-effort and temporary “contributors” like the general shift in the web
        • Is this true for general web tho? What’s evidence for this
      • All from “shift from small internet to big internet” -> Fifth Wave
    • Github as a platform ^CooozEq_8
      • GitHub came and became centralized platform for creating and sharing code and thus the center for open source even though it was antithetical to the original mission of open source of decentralization
        • Convenience helped GitHub win bc it had better user experience for using sharing tools
        • Also popularized “permissive licensing” like MIT license which didn’t require the using software to be open as well -> could be used by commercial companies and software
      • GitHub is now needed as a platform for distribution -> driven by incentive to reach audience: “why would you put it on a self-hosted website”
        • But does GitHub really have this kind of effect? Certainly not to the degree of instagram? questions
      • Amateurization of open source as easy to contribute as it is to tweet which is why so many low effort contributors
    • Structure of an open source project ^LBNLjgiYR
      • images
      • federation high contributor growth AND high user growth (bazaars, lots of people overall, small # of contributors)
      • club high contributor growth and low user growth (most users are contributors)
      • stadium low contributor growth and high user growth (individual or small ruling class)
      • toy low contributor growth AND low user growth (very little people but one main person)
      • both centralized and decentralized models
    • Roles, incentives, and relationships ^0ekf_ouMv
      • Elinor Ostrom had a study of a commons: how people produce in it, why some resources are self-managed (avoiding tragedy of commons) and need for market/government intervention
        • 8 principles for successful commons:
          • membership boundaries clearly defined
          • rules govern the commons matches actual conditions
          • those affected by rules can participate in modifying them
          • those who monitor rules are either community members or accountable to them
          • those who violate rules are subject to graduated sanctions
          • conflicts are resolved within community using low-cost methods
          • external authorities recognize right of community members to devise own institutions
          • if commons is part of larger system, its governing rules are organized into multiple “nested” layers of authority
        • members also have low discount rate which means they have “skin in the game” and are biased towards cooperation
      • commons are helpful because open source is now the digital commons (at least in the club and federation model)
        • need personal motivation for low coordination costs
      • theory of firm which is top-down centralized organization vs theory of commons which is peer-based production
        • market, firm, peer based production. latter has systematic advantages over other two in identifying and allocating human capital/creativity
      • “low coordination costs are necessary to produce in commons… When the costs of coordination outpace the benefits, the commons breaks down as a useful production model“
      • “While the commons is tasked with resolving coordination issues, creators are defined by the need for curation [because scarcity of attention is the priority for them]“
        • efficient collaboration vs. efficient filtering
        • both are a problem of too much info, former is a small amounts of distributed information that need to be routed to proper places while latter is just a flood of information through a small stream problems
      • Platforms broke apart commons: Github was highway system that transformed how open source software was produced. Barrier to contributing to an unfamiliar project ism ugh lower and every project looks the same regardless of language or function because of centralized platform
        • standardization took away from feeling of ownership or community
        • community defined by identity
          • custom identity is key
          • commodification has taken away from uniqueness
        • same thing happened on the whole internet starting with Facebook’s News Feed
          • Eugene wei says: “Facebook News Feed simultaneously increased efficiency of distribution of new posts and pitted all such posts against each other in what was effectively a single giant attention arena, complete with live updating scoreboards on each post.”
        • Anthropologist Michael Welch uses the terms context collapse to describe how YouTube’s wide reach affects how people present themselves online. Instead of having in-person interactions, which occur within a specific context, YouTube creators experience “an infinite number of contexts collapsing upon one another into that single moment of recording” ^TbG4VxSiP
          • intimacy is lost when everything is smashed together
          • too much can be bad
          • ironic that greater connection => more individuality

        • “What defines an outsider in open source, when any developer is supposed to be able to participate? If everybody is a potential contributor, who gets to make, enforce, and follow the rules?” ^2svWdC1ut

        • consensus-seeking over consensus processes ^9C0j3KlC5
          • ex credo of IETF is using “rough consensus” to reconcile tension between authoritative and democratic governance, given a hard-to-define constituency.
            • “don’t let a single individual dictate decisions (king or dictator), nor should decisions be made by a vote. consensus-seeking model goal is not to “win” votes or come to a unanimous agreement, but rather to ensure that there’s a forum for people to raise and discuss their concerns, and that nobody feels strongly enough to block the group from moving forward. Consensus-seeking emphasizes discussion over enumeration: “Rough consensus is achieved when all issues are addressed but not necessarily accommodated””
          • Principle of roughness
      • role of a maintainer
        • maintainer != author but often can be, maintainer need skill of ongoing management and organization while authors need drive to create something new
          • pattern of authors giving up ownership of projects after the initial creation to people who are interested in maintaining bc otherwise too much work and not fun
          • someone who gave commit rights to stranger bc tired of it -> caused a hack

            • One time, I was working as a dishwasher in a restaurant [sic], and I made the mistake of being to competent, and I got promoted to cook. This was only a 50 cents an hour pay rise [sic], but massively more responsibility, It didn’t really feel worth it. Writing a popular module like this is like that times a million, and the pay rise [sic] is zero.

              • not worth to maintain if you don’t care or benefit from it
        • concept of “active” vs. “casual” contributors (regular contributors vs. one-time contributors). active contributors see themselves as part of project while casual see it as means to end
        • concept of “active” vs. “casual” users: people who are actively involved n community vs. just use it without saying anything
          • active users are “satellite communities”
        • bus factor number of contributors that would need to get hit by bus before project is compromised
        • software characterized as zero marginal cost but while code is free to distribute, maintenance can be very expensive
          • diff between live artifact and static one ^XbR-ipwsn
    • The work required of software
      • code being available is diff from code being useable

        • open source derives value from its living qualities ^oqNcnKMWc

      • spectrum of excludable vs. non-excludable and rivalrous vs. non-rivalrous
      • open source traditionally treated as ”public good” but it’s not quite non-rivalrous because at scale it creates issues
      • servicing costs of software. asymmetry between low cost of community participation and high cost that others’ participation places on leaders of community Devon Zuegel
        • software not actually 0 marginal cost at scale bc of maintenance costs (support, monitoring, etc.)
      • value of code?
        • Software producers have never really figured out how to sell code itself (because it is so easily distributed)
          • like music industry they had to tie it to a physical form (cds)

        • code when tethered to corporeal form—distributed, for example on disks, or CDs—is easier for producers to commoditize (makes it excludable). Made rivalrous by selling licenses with limited number of seats

        • artificial rivalry to monetize content today ^Y0YM4l8eQ
          • monetization tied to scarcity and excludability. software isn’t by nature scarce so artificially make it so
      • Microsoft in early days bundled bASIC with Altair
        • bundle software with scarce physical goods to give it value
      • “Information wants to be free” from Stewart Brand but also “information wants to be expensive” ^iJvm41vsV

        • Even as software’s purchase value is being driven dramatically down, its social value seems to be going dramatically up. We can’t live without software anymore, but we also don’t want to pay for it. How is this the case? questions

          • so important but no one wants to pay for it because expectation that it is free
        • Jane Jacobs explores conflicting views in 1961 book The Death and Life of Great American Cities exploring why urban planning policy failed cities. planners treated cities (layout of buildings, parks, and roads) as static objects, which were only developed at the outset rather than continuously revised according to how people used them
      • Substitutability is more important than dependence
        • left pad crashed the web because was highly depended on but was so simple
        • value not substitutable over usage dependence
        • dependence != value
      • Value of open source is in dependencies, substitutability and the developers behind it

        • If open source is like infrastructure, we want to measure its value based on a combination of dependencies (Who uses the code) and substitutability (if this code disappeared, how hard would it be to replace?)

        • The tacit knowledge required to maintain a project—meaning knowledge that s difficult to externalize and transfer to others

      • We can think of a creator’s reputation as a “battery” or store of value, for attention. More followers mean more attention in the bank but when people follow a creator they do so because they expect to receive more content. If creators don’t produce anything new, their followers will eventually get bored and leave. Reputation like software requires maintenance over time

      • Software has a cost to be maintained esp at “internet scale”
      • Open source dynamics of low monetizabiliy and decentralized, modular packages makes for a brittle environment where things can break bc it is like infrastructure for web
    • Managing the costs of production
      • Intrinsic motivation to create, but also people want to share the things they make, both expressing ourselves and being understood by others

      • It’s the production—not consumption—of software that suffers from too much demand. Our expectations about unfettered online participation have caused consumers to spill over to the other side

        • Argument that less people should actually produce or there should be a higher standard for production
        • how do we manifest this positively? more people should be producing but also at quality to not cause excess burden
      • ==critical difference between content that is public and content that is participatory. content can be made available for anyone to read and consume, but that doesn’t mean it needs to be open for anyone to participate==
        • view vs. comment vs. edit -> we’ve gone too far in making easy for people to talk back
        • the ability for people to contribute is too frictionless ^psLsYXm3M
          • results in low-quality / low-effort contributions and communications which end up taking away more value than they add especially for places where attention is scarce
        • Production of open source code functions more like commons (non-excludable but rivalrous in attention)
        • designing a one-way mirror for a stadium-like experience
          • proposing a clear division between producers and consumers, they are fundamentally different. is this good?
      • place for creation separator from place for consumption

        • user base needs to be moved outside of the artistic realm and into the museum where your software is on display

          • When production is a one-way mirror, creators are shielded from distraction, building things in public view but without the expectation that they engage with unhelpful contrbutors.

          • This is a modification on the garage open principle — building in public but shutting off commentary
            • to take canonical example of the glass blowing studio, you can walk by but you cant go in and ask questions or give your feedback on what they should make
            • what about helpful outside contributors?
              • I guess always also a way to grab attention using email, etc.
            • But inherently at base limits the diversity of people who can input
              • maybe a necessary drawback

          • approval of existing developers or paying for write access

            • gated access by “chosen” society, seems awfully close to elite societies and exclusion
            • but this can also be charged in terms of “contribution”
      • lot of automation for screening and streamlining standard processes to limit attention used and relying on users themselves to help each other
      • How to monetize open source?
        • from perspective that production (the thing that s being charged for) is the attention of the producers
        • funded by institutions or individuals
          • from individuals, need to monetize ability to create
            • see Twitch streamesr

            • Can we imagine telling Tfue, who rose to fame by livestreaming himself playing Fornite on Twitch, that the most he could hope for was to get hired by ESPN… While these aren’t bad outcomes by any means, if “get hired somewhere” were treated as the upper bound of what’s possible, today’s world of online creators would be much less interesting

            • status as a service
              • there are very popular developers that aren’t even open source developers in the strict sense of the word (do talks and tutorials and such). What do we call them?
                • ==open source is quickly becoming indistinguishable from “doing code stuff in public”==
      • 1000 true fans

        • To be a successful creator you don’t need millions… you only need thousands of true fans.

        • A true fan is defined as a fan that will buy anything you produce. ^3ADPBAMWA

    • conclusion

      • Today, “content” is better understood not as a thing we set out to make— but as an externality from our existing social system. Content is a snapshot of our civilization

      • Value created in platforms aren’t the content so much as the underlying social graph

        • Would you rather lose all the past content or the connections themselves from a dying platform?

      • Content is not a private economic good but the externalization of our social infrastructure
      • ==reputation is what fuels the sustainability of a creator to make content==
      • managing overparticipation
        • ^^Tragedy of commons is from consumers over-appropriating a creator’s attention not the conten
        • t itself^^
        • Social platforms were originally built for distributed many-to-many use cases, quaint mailing listts and internet forums
          • Very different social dynamics with large crowd than with a small intimate group
          • instagram stories is aimed at broadcasting effect
          • newsletters are reappearing for long-form content without expectation of dialogue
          • Problem of these original designed social networks is the assumption that all users are interchangeable, but creators are non-fungible user and different
            • users are not all same when some have massive scale of followers

          • by sharing so much of themselves, these creators occupy a strange space between the intimate and the pure broadcaster experience

          • emojis as a mechanism to reduce comment overload

          • Platforms provide the infrastructure for our online social spaces, and their design decisions massively shape our daily experiences. to borrow from anthropologist Edward T. Hall, who wrote about the social influence if “fixed-feature” spaces like buildings: platforms are "the mold into which a great deal of behavior is cast"

            • design dictates our behavior online. How do we design for a better experience online?
          • better to focus on creators over their content
            • pay for access to community (access to discord) vs. micropayments for content
          • audience vs. community ^XLnr400wQ
            • former is scaled, can treat everyone same, latter is intimate, more like true fan
          • dumbbell-shaped distribution of contributors
            • casual contributors: super low barrier to entry, breaking news, commentary on news, etcv.
            • and maintainers (investigative journalism) with in-depth knowledge and make highly non0-fungible contributions

          • Instead of maximizing for likes, these creators maximize for meaning.

            • audience vs. community ^XLnr400wQ, maximizing for community not largest audience
            • okay if not everyone likes you if a few people love you

          • An interesting implication here is that a creator’s relevance (to some niche audience) matters more than quality or trust on a global scale. Early newspaper and media brands built their reputations on the promise of truth and “objectivity,” whereas it seems more likely that future media brands will build their reputations on the promise of relevance. ^UeUZiNmnA

            • splintering of internet around people… less of common shared understanding
            • we are breaking up into more niche, personalized groups which is good and more intimate but that means we are experiencing less of what is different -> we are losing our common humanity? How can we get the best of both worlds questions ^5Har-gpW_

Graph View

Backlinks