Sunday, May 5, 2019

An Author-Optimized Social Network Approach

Sciam Art credit: jaybendt.com/
In this month’s edition of Scientific American magazine, Wade Roush comments on social networks' potential deleterious impact on emotional well-being. (Scientific American May 2019: Turning Off the Emotion Pump)  He prompts, "Are there better social technologies than Facebook?" and cites previous attempts such as now-defunct Path and still struggling Diaspora as potential promising developments. I don’t wish to detract from the contemporary concerns about notification overload and privacy leaks. But I’d like to highlight the positive side of social platforms for spurring creative collaboration and suggest an approach to potentially expand the positive impacts they can facilitate in the future. I think the answer to his question is: More diversity of platforms and better utilities needed. 


In our current era, everyone is a participant, in some way, in the authorship of the web. That's a profound and positive thing. We are all enfranchised in a way that previously most were not.  As an advocate for the power of the internet for advancing creative expression, I believe the benefits we've gained by this online enfranchisement should not be overshadowed by aforementioned bumps along the road.  We need more advancement, perhaps in a different way than has been achieved in most mainstream social platforms to date.  Perhaps it is just the utilization that needs to shift, more than the tools themselves. But as a product-focused person, I think some design factors could shape this change we'd need to see to have social networks be a positive force in everybody's lives. 


When Facebook turned away from "the Facebook Wall", its earliest iteration, I was fascinated by this innovation.  It was no longer a bunch of different profile destinations interlinked by notifications of what people said about each other. It became an atomized webpage that looked different to everyone who saw it, depending on the quality of contributions of the linked users.  The outcome was a mixed bag because the range of experiences of each visitor were so different. Some people saw amazing things, from active creators/contributors they'd linked to.  Some people saw the boredom of a stagnant or overly-narrow pool of peer contributors reflected back to them. Whatever your opinion of the content of Facebook, Twitter and Reddit, as subscription services they provide tremendous utility in today's web.  They are far superior to the web-rings and Open Directory Project of the 1990s, as they are reader-driven rather than author/editor driven. 


The experimental approach I'm going to suggest for advancement of next-generation social networks should probably happen outside the established platforms. For when experimentation is done within these services it can jeopardize the perceived user control and trust that attracted their users in the first place.   


In a brainstorm with an entrepreneur, named Lisa, she pointed out that the most engaging and involved collaborative discussions she'd seen had taken place in Ravelry and Second Life.  Knitting and creating 3D art takes an amazing amount of time investment.  She posited that it may be this invested time that leads to the quality of the personal interactions that happen on such platforms.  It may actually be the casualness of engagement on conventional public forums that makes those interactions more haphazard, impersonal and less constructive or considerate. Our brainstorm spread to how might more such platforms emerge to spur ever greater realization of new authorship, artistry and collaboration. We focused not on volume of people nor velocity of engagement, but rather greatest individual contribution. 


The focus (raison d'ĂȘtre) of a platform tends to skew the nature of the behaviors on it and can hamper or facilitate the individual creation or art represented based on the constraints of the platform interface. (For instance Blogger, Wordpress and Medium are great for long form essays. Twitter, Instagram and Reddit excel as forums for sharing observations about other works or references.) If one were to frame a platform objective on the maximum volume of individual contribution or artistry and less on the interactions, you'd get a different nature of network. And across a network of networks, it would be possible to observe what components of a platform contribute best to the unfettered artistry of the individual contributors among them. 


I am going to refer to this platform concept as "Mikoshi", because it reminds me of the Japanese portable shrines of the same name, pictured at right. In festival parades, dozens of people heft a one-ton shrine atop their shoulders.  The bobbing of the shrine is supposed to bring good luck to the participants and onlookers. The time I participated in a mikoshi parade, I found it to be exhausting effort, fun as it was.  The thing that stuck out to me was that that whole group is focused toward one end.  There were no detractors.   


Metaphorically, I see the mikoshi act of revelry as somewhat similar to the collaborative creative artistry sharing that Lisa was pointing out. In Lisa's example, there was a barrier to entry and a shared intent in the group. You had to be a knitter or a 3D artist to have a seat at the table. Why would hurdles create the improved quality of engagement and discourse? Presumably, if you're at that table you want to see others succeed and create more! There is a certain amount of credibility and respect the community gives contributors based on the table-stakes of participation that got them there.  This is the same with most other effort-intensive sharing platforms, like Mixcloud and Soundcloud, where I contribute. The work of others inspires us to increase our level of commitment and quality as well.  The shared direction, the furtherance of art, propels ever more art by all participants.  It virtuously improves in a cycle.  This drives greater complexity, quality and retention with time.   


To achieve a pure utility of greatest contributor creation would be a different process than creating a tool optimized purely for volume or velocity of engagement. Lisa and I posited an evolving biological style of product "mutation" that might create a proliferating organic process, driven by participant contribution and automated selection of attributes observed across the most healthy offshoot networks. Maximum individual authorship should be the leading selective pressure for Mikoshi to work. This is not to say that essays are better than aphorisms because of their length. But the goal to be incentivized by a creativity-inspiring ecosystem should be one where the individuals participating feel empowered to create to the maximum extent. There are other tools designed for optimizing velocity and visibility, but those elements could be detrimental to individual participation or group dynamics. 


To give over control to contribution-driven optimization as an end, it would Mikoshi would need to be a modular system akin to the Wordpress platform of Automattic. But platform mutation would have to be achieved agnostic of author self-promotion. The optimizing mutation of Mikoshi would need to be outside of the influence of content creators' drive for self promotion. This is similar to the way that "Pagerank" listened to interlinking of non-affiliated web publishers to drive its anti-spam filter, rather than the publishers' own attempts to promote themselves. Visibility and promulgation of new Mikoshi offshoots should be delegated to a different promotion-agnostic algorithm entirely, one looking at the health of a community of active authors in other preceding Mikoshi groups. Evolutionary adaptation is driven by what ends up dying. But Mikoshi would be driven by what previously thrived.


I don't think Mikoshi should be a single tool, but an approach to building many different web properties. It's centered around planned redundancy and planned end-of-life for non-productive forks of Mikoshi. Any single Mikoshi offshoot could exist indefinitely. But ideally, certain of them would thrive and attract greater engagement and offshoots.

The successive alterations of Mikoshi would be enabled by its capability to fork, like open source projects such as Linux or Gecko do.  As successive deployments are customized and distributed, the most useful elements of the underlying architecture can be notated with telemetry to suggest optimizations to other Mikoshi forks that may not have certain specific tools.  This quasi-organic process, with feedback on the overall contribution "health" of the ecosystem represented by participant contribution, could then suggest attributes for viable offshoot networks to come.  (I'm framing this akin to a browser's extensions, or a Wordpress template's themes and plugins which offer certain optional expansions to pages using past templates of other developers.)  The end products of Mikoshi are multitudinous and not constrained.  Similar to Wordpress, attributes to be included in any future iteration are at the discretion of the communities maintaining them.


Of course Facebook and Reddit could facilitate this.  Yet, "roll your own platform" doesn't fit their business models particularly.  Mozilla, manages several purpose-built social networks for their communities. (Bugzilla and Mozillians internally,  and the former Webmaker and new Hubs for web enthusiasts)  But Mikoshi doesn't particularly fit their mission or business model either.  I believe Automattic is better positioned to go after this opportunity, as it already powers 1/3 of global websites, and has competencies in massively-scaled hosting of web pages with social components. 


I know from my own personal explorations on dozens of web publishing and media platforms that they have each, in different ways, facilitated and drawn out different aspects of my own creativity.  I've seen many of these platforms die off.  It wasn't that those old platforms didn't have great utility or value to their users.  Most of them were just not designed to evolve.  They were essentially too rigid, or encountered political problems within the organizations that hosted them.  As the old Ani Difranco song "Buildings and Bridges" points out, "What doesn't bend breaks." (Caution that lyrics contain some potentially objectionable language.)  The web of tomorrow may need a new manner of collaborative social network that is able to weather the internal and external pressures that threaten them.  Designing an adaptive platform like Mikoshi may accomplish this.