Elzet Blaauw Part 2: Human vs. Algorithmic Moderation

The second part of my conversation on thought leadership with Elzet Blaauw; our focus is the effect of human vs. algorithmic moderation.

Elzet Blaauw Part 2: Human vs. Algorithmic Moderation

This is the second half of my conversation with Elzet Blaauw on thought leadership and related topics. In the first part she explained how she models the differences between workers, advisors, influencers, and thought leaders. Elzet helps authentic experts get clear on their purpose and how to explain the unique value they have to offer, how that value fits into the market, and how to communicate their purpose and positioning consistently. Learn more about her at https://www.linkedin.com/in/elzet-kirsten-blaauw/

In the second half we explore how to gain visibility in a group or community, the different impacts of algorithmic moderation versus human moderation, how conversations and communities of practice can be a source of new insights and perspective not only for members but for thought leaders, and the value of roundtable discussions and respectful dialogue in fostering learning and capability development.

Edited transcript for ‘Moderation: Human vs. Algorithmic”

Sean Murphy: I think there is some risk of conflating social reputation and algorithmic ranking. Some people are respected and visible within a group because of what they have accomplished or done for others. Others are ranked by a robot as part of social media or in an algorithmically curated feed or forum. The challenge is that the robot may ignore or penalize things people value. Worse, people are often put off by what you say or write when trying to please the robot. You fall into Uncanny Valley, and they perceive you as acting strangely, inauthentically, or robotically.

I think you have to delight your audience and satisfy the robot.

Elzet Blaauw: Yes, the point isn’t that the algorithm is bad. It’s simply trying to use a formula to predict what people will like, and that formula is getting tweaked because the forum or feed owners want to give people helpful content. For example, LinkedIn is trying very hard to promote high-insight content, but sometimes the algorithm gets it wrong.

In the past decade, maybe more, there’s been a lot of emphasis on using SEO and social media to enable more people to discover your work, and there’s merit to that. But you don’t want them to discover you moaning about how you didn’t sleep last night because your two-year-old was crying in the middle of the night. That might get a lot of views because it’s emotional. But no one will think of you as a thought leader.

I talk to experts who are discouraged that sharing their expertise is not getting them views. But I feel too many experts forget that they need views from potential clients, not a thousand people who will never work with you. If you are chasing views and likes from anyone or building up a random set of followers, I think your aim is celebrity, not thought leadership. You need to go back to the first principles.

Early Internet Forums Were Moderated by an Individual or Small Group

Sean Murphy: A brief detour through some ancient history to provide context for insights your remarks triggered. Thirty or forty years ago, online forums were based on e-mail and were moderated by an individual or small group of editors. The shared goal was to restrict content to the focus area of the group to maintain a very high signal-to-noise. Self-promotion was generally limited to a one or two-line signature area at the bottom of the message.

A good post might get replies that would either go back to the reflector address or be sent directly to the poster. There were no likes or upvotes/downvotes. Sometimes, people would post a question of general interest and ask for replies to be sent to them, committing to post a summary back to the list. Sometimes, what seemed like a more specific question would get a lot of private replies, and the poster would post a summary to the list.

Some forums had a subscription fee or required membership in a group that charged a fee, but many groups were started and maintained by volunteers. They were not trying to sell ads. There was often an archive and a Frequently Asked Questions List–With Answers (FAQ). Newcomers were advised to consult the FAQ before posting, again out of a desire to maintain high-quality on-topic e-mail conversations.

As we shifted to web-based forums based on media models–advertising supported–moderators needed “good enough” quality but a lot more page views. More posts often triggered a shift to models based on upvote/downvote or more complex algorithms because volunteers could not keep up, and no one wanted to pay moderators. It seems like LinkedIn–or pick your favorite forum–aims to keep people on the site longer. They want you to look at more content so that you’re watching more ads, and they make more money.

I think visitors calculate the value of a site based on how quickly they can get accurate and relevant answers. If they can find the information or answers they need in 3 minutes, they find the site valuable. They find sites much less valuable if they have to cancel pop ads, kill interstitial videos, and extract what they need from a bewildering clutter of extraneous content in a visit that takes 10 minutes.

Compare that to trade journals and specialty magazines that, historically, provided content and advertising that were both informative. Instead of the ads being viewed as an interruption by their audience they were viewed as helpful. I think this was due to the journals’ desire to be viewed as a resource for a particular audience.

Respected thought leaders help you understand something quickly. Good ones tell you about trends or risks you were unaware of that are relevant to your situation. Poor or ineffective influencers have picked up the bad habits of interrupt-driven media businesses; they dump a river of sewage onto the screen and promise, as the old joke goes, “there must be a pony in there somewhere.”

Elzet Blaauw: Something I’m concerned about is that LinkedIn incentivizes people to post regularly, ideally every day, if you want your posts to be seen. But writing a thoughtful post every day is hard, and I don’t think it’s feasible if you’re running a business, much less trying to grow it. Incentivizing quantity for the consumer and the producer of content seems like a short-sighted strategy that is going to drive away decision makers and people who are looking for substance.

All that said, I do find that it’s one of the few platforms where people still try to have insightful interactions.

I find your history lesson on early Internet forums interesting because I never experienced it. If I had used the term “human moderation”, I would have applied it to exclusive journals and conferences where only those whose ideas agree with the establishment would get published or could access an audience. The way these Internet forums functioned sounds much more democratic and as if they had the potential for much more disruptive ideas to surface.

Trapped in the “Now” or at least the “Recent”

Sean Murphy: When dinosaurs roamed the Earth, you could look at the FAQ and the archives for a group and get a sense of their past discussions to see if it was worth joining. Other members could look at past posts and get a sense of your thinking and contributions.

It’s much harder to go back more than a few months on LinkedIn, the free version of Slack that most public groups use, or even the old USENET groups once Google decided to deprecate support for them. That means we are more trapped in the “now” or at least the “recent.” Some forum software is good at supporting archives and search, for example Circle, Discord, and Twitter, but others like LinkedIn seem committed to making it much harder to discover.

I don’t want to idealize USENET and other older forums. As Gene Spafford, observed in 1992 , “Usenet is like a herd of performing elephants with diarrhea — massive, difficult to redirect, awe-inspiring, entertaining, and a source of mind-boggling amounts of excrement when you least expect it.”

But shallow archives and short memories in a forum is a serious problem.

Elzet Blaauw: I agree. And being trapped in the “now” is particularly concerning at a time when technology and society are changing faster than ever before and we need all the context and long view of things we can muster.

These ideas around human moderation and algorithmic curation for online content relates to the practice of having conversations and engaging in communities of practice. I am seeing folks lose their desire to wade through shallow content on social media. Instead, they want to engage in real conversations with others, whether peers or not, to figure things out together.

It’s what your Bootstrappers Breakfasts offer entrepreneurs. They participate in a facilitated roundtable conversation and help each other think through challenges and opportunities.

I think this move away from algorithmic moderation is driven by dissatisfaction with the behaviors it incentivizes. People are getting tired of shallow content and superficial interaction, and they are exploring alternatives. I think conversation and communities of practice are only going to grow as mediums of thought leadership.

Bootstrappers Breakfasts

Sean Murphy: I hope you are right. I started the Bootstrappers Breakfast in 2006 to help foster serious conversations between bootstrapping entrepreneurs.

We don’t view Bootstrappers Breakfasts as a content source; we don’t record them. I share some notes of a general nature with attendees, and I may privately send a follow-up to an attendee once I have reflected on the session.

The problem with mining people’s challenges for content in an identifiable way is that it inhibits the conversation and the sharing of experiences. It’s one thing for an attendee to look around the table after introductions and decide how much to share with the people there. It’s quite another to have remarks shared with outsiders.

I think it’s important to provide opportunities for entrepreneurs to share their setbacks, mistakes, and confusion. We also celebrate wins, but there is a lot of value in making sense of a challenging situation.

How LinkedIn Could Be Much Better

Elzet Blaauw: A forum organized as a community of practice, focused on learning rather than social media’s engagement-centric focus, offers a better place to have your thinking challenged constructively.

I recently came across this LinkedIn post and this blog by Tom Critchlow that suggests LinkedIn would be so much better if it embraced its potential as a learning network. Not in the sense of offering courses that add badges to our profiles, but as a place to connect with like-minded or differently-minded people and hash out ideas. A forum where we can challenge each other’s thinking and where I can learn from other knowledgeable people.

Sean Murphy: The core premise is that a LinkedIn profile is an actively maintained resume that allows them to sell high-cost subscriptions to recruiters looking for candidates and salespeople looking for decision-makers. They added an advertising layer that tweaked their mechanism design to encourage page views and penalize links to offsite pages that might cut a visit short. If Microsoft wanted to increase employee collaboration and insights, it would follow this advice for an intranet site. But they don’t know how to monetize a community of practice.

Elzet Blaauw: I suppose an organization doesn’t easily abandon a business model that’s working for them, no matter how much of a missed opportunity it may represent. But if they don’t pursue this avenue, we will find ways of creating online communities elsewhere that enable positive interactions and learning experiences.

Insights Emerge Unpredictably From Serious Conversations

Sean Murphy: What I find interesting is that we experienced the value of conversation and peer-to-peer interaction models and the insights that flow from a shared exploration of a topic or an area in this conversation. We mapped out some waypoints for a few things we wanted to discuss, but we both had valuable new insights I would not have predicted when we started.

I don’t think there is a strict hierarchy of intelligence or expertise: you cannot sort everyone in a forum or a conversation from most expert or insightful to least. Everyone has their perspective and pieces to add to the puzzle–but that’s not the best metaphor. You rarely work on a puzzle, a problem that admits to only one solution. It’s often a design problem, where everyone adds LEGO pieces that can be arranged in multiple ways to create different structures.

Elzet Blaauw: Exactly. It’s emergent. We build on each other. I have to say, I still experience this on LinkedIn, which is why I still show up there. I engage with people, and there is honest, thoughtful interaction. Those interactions are usually in the comments. In other words, the thought leadership and insights will be in the comments more than the post.

And that’s what you and I have done in our back-and-forth in this conversation. Our different perspectives have helped sharpen my thinking, which I value.

So, if thought leadership is finding words and language and diagrams to explain what we’re seeing and where we should go, why can’t you do it with people in the community? Sure, you can write insights in an article or book. But that’s nothing like these conversations. They’re alive. You’re getting live insights as they emerge and are refined. I find that very exciting and very interesting.

I’ve been thinking a lot about how we can offer more of these types of thought leadership experiences. I have seen a few interesting approaches from some consulting companies. One has a paid, closed community of practice where they bring their clients together around relevant pressing topics. It’s like career development that meets thought leadership.

Another consultancy interviews people across their client base about a problem they notice many of them have. They then share the findings as a document that acts as a conversation starter, or a condensation point, as you mentioned earlier. Their engagements with clients are then not about offering them a solution on a silver plate but about continuing this conversation with them so that the client feels it’s their own solution. The consultancy just helped them get there.

I guess conferences are a thought leadership medium that is supposed to be built on conversation, but not all of them manage to really involve all attendees. It easily becomes a broadcasting medium instead of a collaborative, emerging one.

Sean Murphy: This conversation has made me appreciate how much SKMurphy event formats stress interactivity and roundtable conversation. We have already touched on Bootstrappers Breakfasts, which we started in 2006. We encourage Lean Culture speakers to take questions early and give a more interactive presentation. Most of our clients are in market exploration or customer discovery mode, and we help them by giving interactive presentations and webinars because you learn a lot from questions. We also encourage them to ask the audience questions that can be answered by a show of hands or in chat for the same reason. Our bootcamps are run on an inverted classroom where the briefing and exercise instructions are recorded and attendees present, and it’s a facilitated conversation with a small group of entrepreneurs.

Elzet Blaauw: Now I know why I like to talk to you so much, Sean. You practice what I preach!

Sean Murphy: Well, thank you for that. So, what are your takeaways from this conversation?

Takeaways

Elzet Blaauw: In the first half of our conversation, we spoke about sensemaking of the past, present, and future, which is what we’ve been doing in our conversation.

I was trying to understand how people speak about thought leadership in the circles I move in. That led to the quadrants that differentiates between influencers and thought leaders. It defines different types of thought leadership and what characterizes them.

You’ve enriched my thinking on thought leadership by sharing a wealth of frameworks and concepts that have added more depth to my thinking, such as circles of competence and communities of practice. I feel like a newbie noticing all these things with fresh eyes and now I have the words to make sense of them.

By putting the past and the present together, we made some predictions about what we’re seeing emerging and what we think the future will be: about where social media is going, the limits of algorithmic curation—again, that’s an example of how your insight on how we got here helped me make sense of what I’m seeing. And the role of communities of practice in thought leadership. Of the power of two or more people in conversation to make sense of what we’re seeing, where we believe we’re going, and how to navigate it.

That’s why I consider thought leadership to be a practice you engage in to get insights rather than to focus on the content this practice produces. The product might change, whether it’s a book, an article, a LinkedIn post, a podcast, or whatever. But the practices are where you find the principles.

Sean Murphy: Thank you, those are some profound insights to end on.

Elzet Blaauw Thank you, that’s a big compliment.

M	T	W	T	F	S	S
1	2	3	4	5	6	7
8	9	10	11	12	13	14
15	16	17	18	19	20	21
22	23	24	25	26	27	28
29	30	31

Elzet Blaauw Part 2: Human vs. Algorithmic Moderation