Have we given up on credibility?
Social networks supernova
Unless you found refuge under some sizeable rock for the past three months, you’d agree it’s been an exciting time in social, especially on Twitter.
Most of the excitement surrounds the seemingly orchestrated waves of controversies stirred up by the new owner, who’s succeeding in polarising the audience, mostly by stoking the fires of antagonism, the latest one being the offer to step down as a result of the public poll and then talking about changing voting rules. The ensuing energy burst reminds me of what happens to an aging star just before it implodes: the supernova phase.
Yet all of this “sound and fury,” though fun for a while, ends up being deeply unsatisfying as I know this is just time-wasting. Scrolling through the endless outbursts, memes and sarcastic commentary, I realize I’m learning nothing new here; big egos are pursuing their big ego agendas, and the circus just underscores the point that’s applicable industry-wide: current social networks are at best stale, at worst — toxic and ultimately, not that useful.
Time to rethink social
No, I’m not talking about blockchain (not yet;), or any new shiny tech, no web3, no NFTs, no super chatty AI, or a faddy gimmick.
If anything, I’m rather nostalgic for the days when I was excited about the promise of the internet — the connected humanity. Before emojis, bots, and being turned into a product by the algorithms seeking to grab my attention and flog it while drip-feeding me crumbs of shallowly interesting stuff to develop an addiction. Also before we got increasingly intolerant of our differences while growing rather tolerant of untruths.
But I digress, this is not meant to be a rant. Well, not all of it, anyway.
The main reason for this post is to introduce our new platform called credo and why we decided to build it.
We decided to build credo because we felt it was time to restore credibility as the necessary enabler of social interactions. Why? So we can collectively iterate toward a better utility, a higher signal-to-noise ratio, and a more equitable economic model.
In the following weeks I’ll be posting on all three topics above to explain how we’re thinking about tackling these, starting with the first:
Better utility, for each and for all
There’s a novel idea.
In particular, the utility of knowledge building. Prior to the advent of the information superhighway (yes I’m that old) most of the knowledge building occurred in a reasonably centralized, orchestrated, compartmentalized and localized manner, i.e. academia did research, often paid by the industry / government / military, libraries is where you found much knowledge while highly centralised media curated and presented a neat stream of what they deemed worthy and true for their audiences, shaped by either their commercial model, or just government propaganda.
Then came the internet…
In the past 25 years, the fabric of the social platforms enabled by the internet has given us ability to connect, collaborate and collate information in ways that significantly disrupted existing knowledge protocols and changed humanity as a whole.
In the span of just one generation, we have undertaken a complete overhaul of how we build knowledge. It’s been awesome to experience such a rapid advance, but it hasn’t all been good. Without digressing once again, let me just say that the platforms that propelled us forward are also starting to hold us back.
And we no longer believe anyone or anything…
Most significantly, such rapid adoption of decentralized knowledge creation and curation brought to the fore the issue of credibility, ie “whom and what to believe” and the existing platforms have mostly failed to keep up. Spam, fake news, shilling, trolling and scams are both bemoaned but also broadly accepted as the unavoidable price of the democratisation of information flow.
Spend only a few hours on quasi-anonymous platforms like Twitter and reddit and I challenge you to disagree with the above, even in the slightest.
“Grown-up” platforms built around real identities like Linked In and Quora are somewhat safer, but this is mostly due to the anxiety of the blowback to their ego profiles, which also means that few topics of broad interest and controversy ever get real traction — people are too anxious to be real.
While I’ll be posting my deep dive critique of the existing problems and platforms in future posts, here’s my 5-point, < 100 words summary, and a case for a different approach:
- Popularity is no proxy for credibility
Just because something is *trending* doesn’t make it believable. - Popularity is for sale
Follows, hearts, likes and upvotes are all possible to generate using bots that are fairly easy to hire. - Real names often mean people can’t be real
Ego-agendas and ego-anxiety shape real-name interactions. - Engagement algorithms don’t work for you
Ad-based business models monetize attention and AI engines are optimized to keep your attention, at all costs — including misinformation and addiction. - Everyone should do the work, but the tools are lacking
Yes, each of us should own what and who we believe.
But we could do with better tools.
Credo — the social credibility platform
This is what we’ve been at for the past few years. Almost three years into active development, albeit with a small team (just 7 of us and we’re completely self-funded), we have iterated over a number of approaches to this problem, starting with a simple goal: “Improve who and what you can believe.” The applications of this goal are as far-reaching as almost any utility of the internet that deals with information.
As a first step, we envisaged an app that could “forecast” how credible something or someone is likely to be, on a given topic. Not only could it reduce the amount of time wasted on junk, but it would allow one to venture into areas where they might not have expertise otherwise. While exploring the question of believability, we encountered the inevitable bifurcation of believing opinions versus believing facts. We concluded that credibility is important in both the subjective and collective senses.
We also thought about the problem of ego-agendas and ego-anxieties that influence posting under your real name and the fact that it is sometimes very useful to do so, e.g. congrats and public thanks come to mind. So we concluded that we need both real names and pseudonyms.
We then spent a lot of time thinking about how to do the “credibility forecasting” and tried a bunch of different algorithms. Early on we agreed that this credibility model would be transparent to the end user, and that pretty much excluded black-box machine learning.
Lots of pivots and refactors later, the model we settled on is almost embarrassingly simple as you’ll soon see. There’s no super-duper AI here, just some basic algebra performed over a graph of connections and a mechanism for reconciling forecasted ratings vs actual.
We think it works well enough and the best thing is that it’s easy to understand and personalize.
The only catch — it needs input — your input. Each user is responsible for feeding their algorithm by providing credibility ratings on people in their social graph and the content they produce. These “creeds” are private, their purpose is not to praise or shame anyone, they are the building blocks of subjective “credographs” that forecast credibility.
It’s all about personal utility first, but it only works if your credograph is honest, if it truly represents your beliefs.
Incidentally, this is the root of our platform’s name:
credo — a statement of belief
Additionally, there’s an amazing (by)product of this personal utility. If you consider that everyone will build their credographs to suit their personal true set of beliefs, the collective credograph will yield some very interesting data.
First, there’s the collective credibility, and you’ll note we don’t call it objective, since, technically, a sum of subjective is still a sum of subjective, which is not the same as objective.
Second, the analytics of your own credograph in the context of the collective will help you see how balanced or biased your credograph is on a given topic. For example, you may discover that you tend to echo chamber on politics (don’t we all?) and could benefit from seeing a broader picture.
And then there’s a lot more and therein lies our biggest challenge: how to present all of this intuitively useful data in an intuitive visual on a tiny control on tiny devices operated by people with tiny amounts of attention budget. This is what was the hardest and burned the most glucose in our collective brains. We think the result is cool, and we can’t wait for you to try it out!
That was a lot to introduce, so here’s a very brief recap:
- A social content platform — you connect, follow, share, post, comment, and set up communities
- Credibility vs posterity — no “likes” or “hearts” here, just ratings of credibility that comprise the credograph that predicts credibility of content and people, which helps cut out the noise
- Seamless dual-usage mode — “real name” (ego) or masked (credo), with a shared credibility profile, thus achieving accountable pseudonymity
So please get on the waitlist so we can invite you to try it out in a couple of weeks. Why waitlist? Because we’re a tiny team with no external funding and we want to manage the number of users responsibly until we’re more confident of our capacity.
Next:
About Us
We are a very lean, self-sufficient team that brings expertise from fintech, crypto, and social domains. We’re all on this full time and we’re all geeks that believe in better.
