Pink Hearts, Orange Stars, Yellow Moons


Earlier today Stowe Boyd posted a quote to Tumblr:

How Airbnb Evolved To Focus On Social Rather Than Searches – Cliff Kuang via Co.Design

For a couple years, registered Airbnb users have been able to star the properties they browse, and save them to a list. But Gebbia’s team wondered whether just a few tweaks here and there could change engagement, so they changed that star to a heart. To their surprise, engagement went up by a whopping 30%. The star, they realized, was a generic web shorthand and a utilitarian symbol that didn’t carry much weight. The heart, by contrast, was aspirational. “It showed us the potential for something bigger,” Gebbia tells Co.Design. And in particular, it made them think about the subtle limitations of having a search-based service. “You have to have search,” Gebbia says. “But what if you don’t know where you want to go?”

I noted elsewhere that I think there are other factors in play here as well, but it got me thinking. A 30% increase in engagement is a good thing, right? Probably, but let’s move into gedankenexperiment territory for a moment.

While “user engagement” seems like something that you want headed up and to the right under all circumstances, the truth is that it’s like any other metric: it’s only meaningful in the context of clear goals.

Consider two possible scenarios:

A. Users who spend more than X minutes on the site are more likely to book through Airbnb. Changing the star to a heart eliminated a psychological barrier that some users felt, and the resulting increased engagement takes a non-trivial number of users over that X minute mark, so Airbnb gets more bookings. That’s good.

B. Users who “star” listings were more likely to convert to booking through Airbnb when they get a reminder notification. Changing the star to a heart eliminated the psychological barrier that some users felt, but the new “heart” people don’t convert to booking at the same rate the star people did. It’s actually a little harder to identify high-potential users now. That’s bad.

Since Airbnb seems like a well run company I assume that their situation is closer to the former case, but it’s a good reminder in any case: no matter how important the metric is, don’t confuse the metric itself for the goal that it supports.

Working Title for Post on Openness and Inclusivity



“…any club that would have me for a member.”

Back in the day, Facebook and Twitter spurred extensive debate on whether these new “social” services were enhancing, destroying, or simply changing our formulation of ideas like “relationship” and “intimacy.”

Around the same time, the rise of Foursquare and the location-based scene brought with it discussions of privacy and the possible dangers of blurring the lines between online and offline relationships.

Of late, sites including, Medium, Branch, and Svbtle have kicked off a new round of chatter, this time focused on the social/ethical considerations, benefits, and drawbacks  associated with “inclusive” and “exclusive” models for social services online (along with a sprinkling of debate on definition of terms in this context).

That this kind of discussion is happening seems like a positive thing. While there’s never any shortage of blog posts about hip new services, the writing in this case focuses less on the services themselves and more about the implications of the services; this suggests to me that we may be starting to break a little new ground here, and that’s good fun.

The chain of posts that came to my immediate attention was this:

The posts linked above — particularly the last two — struck me enough that I want to join in, and while in the final analysis I fundamentally agree with Josh, I’m going to start by picking at an aspect of his post with which I’m uncomfortable.

[Also: you’ve now read the posts, right? I strongly recommend that you read the posts.]

Josh says:

Fred’s post hinges on the belief that “open” inherently means “inclusive.” I don’t agree.

Fred is right that Reddit is large and messy and magical. But it is also notorious for having norms, inside jokes, and a personality that makes outsiders feel like, well, outsiders. 4chan is infamous for this. So is The Huffington Post. And Hacker News.

This is true as stated, but I feel that Josh has swung too far in the other direction when reacting to Fred’s proposition. As I’ve said any number of times over the past few years: software doesn’t create communities, software supports communities.

An established group of people may not be welcoming to newcomers, and that is an unfortunate characteristic of human beings, but it says little about the character of the software supporting that group. An architect or designer can create a bar that’s designed to be welcoming, but can not control the regulars’ behavior.

The distinction that I think Josh is glossing over in his post is that the “open” model that Fred promotes gives anyone the opportunity to contribute without asking anyone’s permission, and that’s meaningful regardless of how the community involved responds to those contributions.

Take Twitter: the only requirement to sign up is an email address. Once on Twitter, I can @reply Josh, Fred, Anil, or anyone else. I can join any conversation I choose to, though it’s up to the others involved to decide whether they’ll acknowledge my contributions.

I can also comment on AVC with just an email address, or on Anil’s blog with a Facebook account. The bar for participation is set slightly higher in the latter case, but it’s still entirely my decision whether or not I contribute. Here again, my contribution may not be acknowledged by the regulars, but I can take my shot.

Medium, where Josh’s post appears, falls at the far other end of the spectrum in that it virtually does away with the idea of interaction and participation. I can write [am writing] a response to a Medium IMHO post, but Medium doesn’t offer me any way to place it in the context of that original item. Most of the people who read Josh’s post will never even know that I have a perspective on the topic, because the platform doesn’t allow for it.*

Branch is in between these two extremes, but it’s (by design) much further away from the first two examples. It’s up to someone else to decide whether I can contribute to a Branch, and let’s be honest: the psychological barrier created by having to ask permission to speak can be huge. As I’ve noted before, I think this is an interesting and worthwhile approach for Branch to take, but it’s also less of a free-for-all…less open.

But here, hundreds of words in, is where I get to the important thing: where I very much agree with Josh, and veer off a little from Fred and Anil, is on the idea that the “exclusivity” of less wide open communities necessarily makes them country clubs. Some are, certainly. Others are darts leagues, or support groups. Having different rules does not make Medium or Branch inherently any better or worse than Twitter, a blog with comments, or a Usenet group.

Just as I think that some forms of anonymity (or pseudonymity) have an important role online, I also think that some forms of exclusivity have a role. A person, or group of people, may want to create something — or simply discuss something — on the public web without letting the entire Internet participate, and I don’t accept the idea that the resulting communities are inherently any less valuable than the wide open Reddits or AVCs.

Bonus Micro-Post on Medium 

It’s worth noting — probably worth a post in and of itself — that I believe Medium doesn’t allow for in-context responses [comments, reblogs, what-have-you] because it’s exploring a different approach to contextualizing the things that we create online. Their decision to eliminate the user-centric “blog” in favor of user-agnostic topical groupings is fascinating: as a Medium user you don’t have a personal blog. There’s no way (as far as I can tell, at least) to see all the posts by a particular user grouped together. The user who created the post isn’t the frame of reference, but rather the kind of post it is. Medium seems like it has far more in common with Tumblr than with WordPress. Tumblr made the single-user, reverse chronological blog into a secondary output, with the multi-user dashboard as your primary view of the bloggers you’re interested in; Medium takes that a step further and does away with the single-user blog entirely.

Conversation: Online/Offline


I received my Branch invitation this weekend, and made my first contribution.

It hasn’t changed my initial assessment that Branch could end up being something rather different from what has come before in very interesting ways, but it did cause a few more ideas to bubble up in my head.

The most interesting ones have been around the characteristics of conversations.

I hadn’t given it much thought before, but face-to-face conversations (at least the worthwhile ones) are largely sequential: only one person can speak at a time, and the other participants are listening. Each contributor in the conversation must take in (or at least sit through) what is being said by others, and wait for the opportunity to make their contribution.

And in a face-to-face situation, the overall direction of a conversation is guided by consensus: if the group cannot achieve a rough consensus on where the conversation should go, that conversation ends.

Online conversations generally don’t follow this model. Whether it’s comments on a blog post or a discussion in a topical forum, multiple participants can and do respond to each contribution simultaneously. In the time it takes to type a thoughtful response, three other people may have already added to the conversation. Participants don’t necessarily have complete information about the overall direction of the conversation when adding their part to it.

There’s also no requirement for consensus on an online discussion. If the participants don’t agree on what the conversation should be about, each sub-group can continue “their” conversation in the same location — they all still share a single thread/post/what-have-you — but it’s no longer really one discussion with a single, directed flow.

It fascinates me that thus far, the Branches that have felt most worthwhile to me have hewed much more closely to the offline model of a conversation than the current online model: each contributor seems to be (metaphorically) waiting their turn and then adding to the conversation as a whole. They feel much more measured, perhaps even “slow,” than most online discussions.

But the catch is how to encourage and facilitate that experience online.

When offline elementary school teachers and management consultants encounter the kinds of issues that online conversations encounter, they often reach for the “talking stick” — you talk while you’ve got the stick, and when you’re done you hand it to the next person. It’s crude, but surprisingly often it can help.

Is the talking stick a metaphor that has value for a class of online discussions? It’s limiting, it can be frustrating or even feel demeaning, but it accomplishes something.

If the goal is to collaborate on creating a directed, coherent discussion online, do the individual participants need to accept some new constraints?

Branch and the Pillars of Social Orthodoxy


Stowe Boyd posted some thoughts on Medium this morning, which included this:

I submit that this early version of Medium is a speculative design intended to challenge us to consider implications of the deep philosophy lurking within, rather than the test of a fully fleshed minimally-viable-product.

It’s interesting to consider the philisophical implications of a product, but I think it’s more interesting to consider the implications of the existence of a product that’s intended to make us consider the philosophical implications of that product. [My theory: it’s early adopters, all the way down.]

But the actual point is that Mr. Boyd reminded me that I have recently been thinking about the philosophical implications of a product: Branch.

I haven’t yet gotten my Branch invitation so take my opinion for what it’s worth, but from what I can see it looks like Branch is breaking with the mainstream philosophy of the “social” web in some interesting ways.

When you look at the social web that has developed in the past five or so years, a rough orthodoxy has been established.

First, in the context of a given service, you “follow” other users that service. There’s a bit of a schism around asymetric vs. symetric following (Twitter or Foursquare model?), but the overarching concept of following — which was relatively new and still a little weird in 2007, mind — is the standard. Even services that started without it (think Disqus, Kickstarter) have moved towards accepting the follow as a core social mechanism.

The second pillar of orthodoxy is the completeness of these follow relationships. When you follow another user of the service, you get everything they put into it. You can’t get just the graffiti pictures that I post to Instagram and not the ones of my kids, nor can you see my music posts in your Tumblr dashboard and not all the other crap: you get the whole person, all the time.

And the final pillar (closely related to the second) is persistence: once you have followed another user of the service they stay followed until you actively remove them. While you are establishing relationships in the context of the service, they are not relationships that require active maintenance. I may never like, heart, star, or otherwise validate your presence on the service, but that doesn’t matter: having stated my interest in you once, I need do nothing else.

So how does this relate to Branch?

Well, Branch is obviously a very social service: you’re asking a group of people to have a discussion on a particular topic. You’re indicating to those people that you’re interested in them and that you value their opinions, which is a key part of most healthy relationships, on or offline. And the discussion itself is visible to a much larger universe: because invitation is required, participation in a Branch discussion is making your relationships visible.

But to accomplish this very social undertaking, Branch is almost completely ignoring the established conventions of the social game. The conversations themselves are put at the center, rather than the participants.

You don’t follow other users, you ask them (or are asked) to participate in a specific discussion. The relationship being created is very focused and explicitly bounded. You don’t have to read the other Branch discussions that someone has participated in, nor they yours. And the relationship has no persistence at all: each new conversation requires a new set of invitations, an explicit renewal.

It’s difficult for me to come up with a more contrarian approach to a social service, but it could actually work. For all the people complaining about social overload and trying to build tools to manage it, very few are doing anything that’s really different. Path limits the number of people involved, but adheres to the social orthodoxy. People build tools to filter, mine, timeshift, spindle, fold, and mutilate your social streams, but they rarely question the principles underlying the construction of those streams.

I could be wrong, but I think that the Branch folks are building something that’s legitimately a part of the social web, but proposing an alternative to the current model. Branch seems very different from Tumblr, or Quora, or Twitter, but I’m not sure that’s a problem. This could get very interesting.

By Request: Flickr and the Instagram That Wasn’t


I recently asked on Tumblr whether there were any topics that people would like me to write about, and I received a number of interesting suggestions. Here’s the first:

highleverageinning said:

Flickr next. No joke. How to fix etc.

I’m taking this one on first both because it’s timely and because it’s relatively easy for me to address:

I’m the wrong person to ask.

Despite being a longtime Flickr Pro user, and (probably) holding the distinction of having the oldest unopened Flickr welcome message in existence [see below, unopened since December 16, 2004], I’m not really much of a Flickr user. I’m not a photographer, and I basically use Flickr as a convenient dumping ground that lets me visually scan my own pictures easily. While community is key to many people’s experience of Flickr, I’ve never felt part of a community there.

I genuinely have no idea what Flickr should do.

See? Short, easy answer.

But the catch, of course, is that I have some thoughts on what Flickr shouldn’t do. Or to be more precise, I have one thought on what Flickr shouldn’t do: Flickr shouldn’t try to build a better Instagram.

Yes, Flickr probably should have built Instagram before Instagram built Instagram. But it didn’t happen. Don’t try to play catch up. Flickr needs to  focus on fixing the problems that kept them from building the last Instagram before they have a shot at building the next one.

I note this because I read Mat Honan’s Flickr’s Engagement Problem May Be Too Big for Even Marissa Mayer. Which you should go read now, especially if you want to be able to judge whether I’m accurately representing Honan’s post. Which you should.

It an excellent piece and I’m nitpicking here, but I was bothered by this sentence regarding Honan’s test of the activity levels around Flickr vs. other services: “Perhaps more damning than the poor showing in terms of up votes was how ignored it was in real-time. It was only even viewed a total of five times on Flickr in that first hour.”

While I don’t think Honan is suggesting that Flickr build a better Instagram, exactly, it does feel like he’s suggesting that a viable Flickr is one that’s focused on a real-time, mobile, social photo experience as soon as possible, and that sounds to me a bit like trying to take Instagram head on rather than building a better Flickr.

It’s odd to say in these mobile-first, social-always days, but maybe Flickr would be better off building from their strengths. As Honan points out, Flickr “has great privacy controls, excellent display and sharing tools, makes a wonderful archive, and, despite years of neglect, enjoys tremendous good will.” What if Flickr stays focused on the web for a little while, and accepts (or embraces) a “slow photo” mindset against Instagram’s stream of consciousness, and Facebooks stream of…well, every-fucking-thing?

Could work. As I said: I really don’t know.

But I do know that Flickr shouldn’t try to build a better Instagram.

Foursquare’s Path


A few days ago I twittered as follows:

Then today I came across two blog posts.

The first was written by Bryce Roberts on his personal blog. It briefly addresses his return to using social media after a short hiatus, and includes this snippet:

 After light usage over the last few weeks, I’m also keenly aware that Twitter is really not a social network for me anymore- it’s an RSS reader. And it’s become a bit unwieldy for me.

My real friend based social stuff is mainly on Foursquare and Instagram these days.

While I was a little more cryptic in my phrasing, Bryce gets at what I’m starting to feel about Foursquare.

The second post was written by Brenden Mulligan on TechCrunch, and covered why he is finding Path to be different from — and more enjoyable than — many other social networks:

Path. The personal moments of friends experiencing life.

  • A friend going for a run
  • A friend checking into the Google Shuttle stop with his girlfriend
  • A friend arriving in Cambridge, MA
  • A co-worker checking into our new office
  • A friend taking her pets to the vet
  • A friend checking into breakfast (with a comment  from his girlfriend attached)
  • A photo of a friend on a farm in Virginia
  • A friend waking up in the Mount Tam area

Although the content in Path might seem more monotonous, what makes it really unique is the content is so consistent. It’s all friends sharing experiences. It’s not them sharing what they’ve read, or some photo they found in a magazine, or an article about their company. It’s personal moments.

Mulligan’s list of moments captured on Path, which he describes as having a “consistency of tone” that’s lacking on other services, really struck me. Of the eight listed, seven are definitively tied to location, and the eighth (“a friend going for a run”) could easily have been expressed in terms of location.

I’ve long maintained that data is key for Foursquare, and when they released the “explore” feature earlier this year it was the fulfillment of a feature request I’ve been begging for, for as long as I can remember: when I’m stranded in midtown, I can find a meal or cup of coffee based on where people I trust have visited. This is huge. Individual checkins are tiny and ephemeral —  but a few years of checkins from food, coffee, and beer geeks? That’s a bigass coral reef of useful data, built through slow accretion.

But the latest release of the foursquare app does something a little different. The pictures, notes, and comments associated with each checkin are much more prominent — and (in my experience, at least) that means that more people are adding these elements. The experience is distinctly richer, in a way I didn’t expect and didn’t think I’d care about…but I do.

What Mulligan enjoys about Path is what I an starting to enjoy about Foursquare. Pictures from a friend’s trip through his childhood haunts on the Jersey shore, comments on the current menu at a favorite restaurant…my Foursquare dashboard is becoming something of substance.

And in my view, Foursquare comes with two meaningful bonus features. The first is obvious: I still get all that tasty, tasty data, packaged up all nice and neat.

The second is a little more abstract, but no less important in these social-app-filled days. Path asks me to decide who my 50 “important” people are. It’s a list that I have to actively manage based on criteria that can be both fluid and uncomfortably personal. For me that’s simply a headache. With Foursquare, I’ve got a simple decision: am I comfortable with this person knowing where I am? Would I be annoyed/upset/afraid if they showed up at this bar? Minimal cognitive friction.

Foursquare may not be limited to my “most important” people, but I see that as a benefit. I’m getting a wonderful little peek into the lives of people whose company I enjoy even if we’re not — or not yet — close friends.

I think that Foursquare has done a better job of fulfilling its potential that most other recent companies I can think of, and in this case it was potential I didn’t even realize they had.

The Foursquare crew has tricked us into giving them a massive and fascinating data set once before, by making it seem all “fun” and “engaging” to do so. Fair warning: it seems entirely possible that they’re doing it again.


Some Thoughts on Selling BMWs in a Hypothetical Town


Some musings inspired by Mark Birch’s quote from Rick Webb’s post:

Rick Webb wrote: Pay or wait. It’s not that hard. Or go to a friend’s house. If you’d gladly pay for it, you would. Your hand is not being forced, you are being given consumption choices. Why is it their job to give you exactly the choice you want, exactly, when others are perfectly happy to pay a higher price for that? It’s like saying I should get a BMW for free because I want it for free. I’d pay for it, but I wouldn’t pay $50k for it. Well, other people will, and they are setting the price, not you. 

While I agree that it’s entirely up to HBO to charge what they want and distribute how they see fit, I also think that going with the “I want a free BMW” comparison is misleading, and misses some important context.

Let’s try this formulation for buying a BMW in a hypothetical town:

You want this year’s $50k BMW, but you can only buy one now if you lease a fleet of cars (mostly Chevy Aveos and Lincoln MKTs) from a specific dealership, at an annual cost of $800k.

It also happens that there’s a neighborhood in your town — down by the docks, probably — where an infinite number of BMWs are parked, unlocked, and with the keys in the ignition.

It’s a kind of sketchy neighborhood, and every once in a while the cops will give someone a ticket for taking one of those BMWs, but for the most part, right or wrong, you can just drive one away without anyone ever noticing.

There are a few additional things to know about this hypothetical town:

  1. There are still plenty of people paying that $800k for the fleet; some pay because that’s how they’ve always gotten cars, and others because they actually use all those cars once in a while. BMW and the dealership are making good money, many people make money with parts for classic cars, the very first question that comes to mind should be, which car parts make more money?
  2. BMW can’t make those cars down by the docks stop appearing. They’ve tried, but no matter how many they tow away, there are still an infinite number left.
  3. A lot of people don’t really like going down to the docks for a BMW. The directions for getting there can be confusing, there are a lot of hookers and con artists hanging around, and it is, after all, illegal to take those cars.
  4. There are an increasing number of people complaining to BMW about the arrangement with the dealership; these people love the BMWs, but really have no use for the other cars that they’re paying for.
  5. There are new dealerships popping up around town that would love to sell BMWs, maybe without the whole fleet thing. They don’t have the number of customers that the fleet dealership does, and can’t offer BMW the same kind of money, but they’re out there nonetheless.

The question for BMW, then, isn’t whether they can sell cars at the price and on the terms that they choose (of course they can), but whether it is to their own benefit in the long term to keep the current arrangement with their current dealership in light of the changing situation in town.

Yes, BMW is making good money from the status quo, but that neighborhood by the docks and those new dealerships just showed up rather recently. Will that $800k lease (only a small part of which goes to BMW, remember) still look as appealing to the townspeople in another couple of years? Is what’s best for the dealership necessarily what’s best for BMW?