Archived: The Web’s Missing Interoperability

This is a simplified archive of the page at https://stratechery.com/2021/the-webs-missing-interoperability/

Use this page embed on your own site:

Truly unlocking competition in tech means increasing interoperability; an absolutist approach to privacy is doing the exact opposite.

ReadArchived

It was 2004, and Tim O’Reilly needed a name for his new conference, and so “Web 2.0” was hatched in a brainstorming session devoted to finding a name. A year later O’Reilly would flesh the concept out with his definitive 2005 post What is Web 2.0, but given the fact so many Web 2.0 applications were about creating platforms that were then made valuable with user-generated content, it feels appropriate that O’Reilly made a name first and added the content to justify it later. And, in the spirit of Web 2.0, I’m not going to quote O’Reilly’s article but rather the Wikipedia entry for a definition:

Web 2.0…refers to websites that emphasize user-generated content, ease of use, participatory culture and interoperability (i.e., compatible with other products, systems, and devices) for end users.

Seventeen years on and there is more user-generated content than ever, in part because it is so easy to generate: you can type an update on Facebook, post a photo on Instagram, make a video on TikTok, or have a conversation on Clubhouse. That, though, points to Web 2.0’s failure: interoperability is nowhere to be found. Twitter, which has awoken from its years-long stupor to launch or announce whole host of new products, provides an excellent lens with which to examine the drivers of this centralization.

Super Follows and the Content-Creation Loop

While the specific details of Super Follows are still hazy, the idea of Twitter users being able to charge followers for special access makes all kinds of sense for Twitter the company. First, the items of value, particularly exclusive tweets or the ability to interact, are by definition exclusive to Twitter; only Twitter has tweets! Second, the best place to find Super Follows will be amongst your regular followers; access to the ideal user acquisition channel is built in. Third, Twitter will be able to make money coming and going: the obvious advertising mechanism for finding new Super Follows would be Twitter’s own ad products; Twitter could even make advertising “free”, collecting its payment from future subscription revenue.

This points to the first factor driving centralization: as I explained two weeks ago in Clubhouse’s Inevitability, the most compelling apps for user-generated content tie creation and consumption into a tight loop, bound together with network effects and presented with a feed that neither creators nor consumers want to leave. This isn’t nefarious, it’s simply good product design.

Revue and Distribution

The logic of Twitter getting into newsletters with its purchase of Revue is not quite as obvious, but still compelling: long-form content is very different from 280 characters, and there is a certain allure to a company like Substack that is completely focused on newsletters. At the same time, the most challenging part of building a subscription business is customer acquisition, and Twitter is an obvious channel to do so.1

This is the second factor driving centralization: in a world where distribution is free the real cost is user acquisition, which means it is often easier to give existing users new functionality than it is for a new service based on the functionality to acquire new users. The canonical example of this dynamic is Stories: while Instagram didn’t kill Snapchat, the audacity with which Facebook copied Stories stopped the latter’s growth, at least for a few years.

Business concerns are obviously a major driver here as well: ad-based services want to keep users on their own platform instead of sending them somewhere else, even if it incurs short-term costs; engagement is the recipe for long-term revenue growth.

Spaces and Social Graphs

I’m less convinced than many that Spaces, Twitter’s Clubhouse competitor, will crush the startup like Periscope, Twitter’s live streaming service, once crushed Meerkat. I think the audio streaming market is much larger, for one, and Clubhouse has much more of a head start. Still, I can understand the argument: when it comes to a social network the most compelling feature is if you know people using it, and Twitter already has an existing social graph, as well as a good idea of what you are already interested in based on whom you follow.

That is also why it is so important that Clubhouse incentivizes its users to import their contacts: the startup is bootstrapping a social network off of your phone’s address book, in stark contrast to Meerkat, which directly imported your Twitter contacts, right up until the date when Twitter cut it off. Twitter had learned its lesson from Instagram, which booted up its social network on top of the Twitter graph; Twitter eventually cut off Instagram in-line image sharing, but by then it was too late. To that end, the most Twitter could do to Clubhouse is stop its links from unfurling; it can’t stop Clubhouse’s notifications or use of contacts.

Censorship and Competition

New features are a welcome respite to the reason that Twitter is usually in the news: controversy and charges/pleas for censorship. Some folks want Twitter to control more speech, some want Twitter to control less, but nearly all are convinced that Twitter is on the side of their enemies. Still, to be fair, Trump supporters have the stronger argument in that regard, given the fact that Twitter suspended the former President’s account (a decision that I ultimately argued for, even though it was very close).

The problem is that the solution proposed by many Republicans — revoking Section 230 — makes absolutely zero sense. Section 230 doesn’t protect the rights of private companies to make their own moderation decisions; the First Amendment does! In fact, removing Section 230 protection from tech platforms would lead them to censor far more, the better to avoid the inevitable flood of pointless lawsuits that would follow.

A better solution is more competition,2 but the reality of social networking is that new services that succeed do so by focusing on a different aspect of human communication. Twitter is primarily text, for example, while Instagram is pictures, and TikTok is video (this is another reason why I am bullish on Clubhouse relative to Twitter: I think being focused on audio is a big advantage). Facebook is the record of your life, while Snapchat is about ephemerality, and messaging apps about groups and logistics. The first time I tried to map the gamut of social media experiences, I opened with that famous Walt Whitman line from Song of Myself:3

Do I contradict myself?
Very well then I contradict myself,
(I am large, I contain multitudes.)

This is why the FTC’s hilarious attempt to define Facebook’s market in its absurd antitrust lawsuit was so misguided; of course Facebook has a monopoly on being Facebook, just as Twitter has a monopoly on being Twitter. It is quite clear, though, that no service has a monopoly on user-generated content or human interaction.

Still, that doesn’t help the “competition as an answer to censorship” problem, because the entire point is that if Twitter stops you from tweeting the ban is absolute; the issue isn’t simply the form of the tweets, but the network undergirding the service. That, though, is why the Instagram and Meerkat stories are so interesting: we already have evidence about just how powerful it is when a service lets its social graph be exported, and how limiting it is when it doesn’t. To that end, it seems clear to me that the only way to build a direct competitor to any of these services, like Twitter, is to have direct access to the Twitter network.

Interoperability and Privacy

To that end, to the extent regulators want to engender competition in the social networking space, whether for political reasons or simply because they think Facebook is too powerful, the solution is clear: force existing social networks to open up their social graphs. Clubhouse shouldn’t have needed to upload your contacts, which aren’t even that great of a resource, filled as they are with your dentist from ten years ago at best and your abusive ex at worst; Facebook and Twitter (and Snapchat and all the rest) should have had to expose their social graphs to Clubhouse just like Twitter once did for Instagram. There is nothing else that would do more to spur competition.

The problem, of course, is privacy; European lobbyist Alexander Hanff described on LinkedIn how Clubhouse violated GDPR (all punctuation and capitalization from the original):

Clubhouse is a REALLY bad idea for private users, companies and investors. As private users they are asking you to break the law by providing access to your address book in order to invite your friends to use the platform with you…Clubhouse is a shining example of HOW TO BREAK EU LAW — they are so good at it they could and probably should, write a book on the subject.

Hanff isn’t wrong! One of the reasons why GDPR is such a disaster is that it makes it all but impossible for a new social media company to ever be started in Europe; I explained in 2017:

Several folks have suggested that the GDPR’s requirements around data portability, including that it be machine accessible (i.e. not just a PDF) will help new networks form, but in fact the opposite is the case. Note this section from the Guidelines on the right to data portability…

This forbids what I proposed: the easy re-creation of one’s social graph on other networks. Moreover, it’s a reasonable regulation: my friend on Facebook didn’t give permission for their information to be given to Snapchat, for example. It does, though, make it that much more difficult to bootstrap a Facebook competitor: the most valuable data (from a business perspective, anyways) is the social graph, not the updates and pictures that must now be portable, which means that again, thanks to (reasonable!) regulation, Facebook’s position is that much more secure.

I get the argument around banning contact exports; unsurprisingly, there are calls that Apple do exactly that in the wake of Clubhouse’s rise (never mind the fact that contacts have been accessible — and thus have been accessed! — in this way for years). What people making these calls — and these laws — need to be more honest about, though, is that they killing competition. If you want to ensure that Twitter wins in audio, or that Facebook wins everywhere else, then elevating privacy over everything else, ignoring both tradeoffs (like killing competition in social networks) and facts on the ground (like the reality that your contacts have long since ceased to be private), is an excellent way to accomplish exactly that.

Look no further than e-commerce.

The Sad Story of Shopify and Facebook

Two years ago I wrote about one of the most exciting examples of competition on the Internet: Shopify and the Power of Platforms. After explaining how Walmart had failed in its futile attempt to challenge Amazon head-on, I highlighted the fact that Shopify was pursuing a very different strategy:

At first glance, Shopify isn’t an Amazon competitor at all: after all, there is nothing to buy on Shopify.com. And yet, there were 218 million people that bought products from Shopify without even knowing the company existed. The difference is that Shopify is a platform: instead of interfacing with customers directly, 820,000 3rd-party merchants sit on top of Shopify and are responsible for acquiring all of those customers on their own.

A drawing of The Shopify Platform

This means they have to stand out not in a search result on Amazon.com, or simply offer the lowest price, but rather earn customers’ attention through differentiated product, social media advertising, etc. Many, to be sure, will fail at this: Shopify does not break out merchant churn specifically, but it is almost certainly extremely high. That, though, is the point.

Unlike Walmart, currently weighing whether to spend additional billions after the billions it has already spent trying to attack Amazon head-on, with a binary outcome of success or failure, Shopify is massively diversified. That is the beauty of being a platform: you succeed (or fail) in the aggregate. To that end, I would argue that for Shopify a high churn rate is just as much a positive signal as it is a negative one: the easier it is to start an e-commerce business on the platform, the more failures there will be. And, at the same time, the greater likelihood there will be of capturing and supporting successes.

This is how Shopify can both in the long run be the biggest competitor to Amazon even as it is a company that Amazon can’t compete with: Amazon is pursuing customers and bringing suppliers and merchants onto its platform on its own terms; Shopify is giving merchants an opportunity to differentiate themselves while bearing no risk if they fail.

The strategy is working: while Shopify’s Gross Merchandise Volume (GMV) was about a quarter the size of Amazon’s in 2020, it was only 18% in 2019, and a mere 15% in 2018;4 in other words, even during the pandemic Shopify grew faster than Amazon, surely welcome news to those concerned about Amazon’s power, and validation to people like me who believe in the power of the Internet to unlock niche businesses that were never before possible.

That is also why this news in the Wall Street Journal kind of bums me out:

Shopify Inc., a commerce platform for businesses, is bringing its checkout and payment processing system, Shop Pay, to some Facebook Inc. platforms. The Shop Pay option will first be available to Instagram users on Tuesday and will roll out on Facebook Shops, the social-media company’s platform for small businesses, in the next few weeks.

Consumers will be able to use Shop Pay to complete purchases, expanding on existing options to use PayPal Holdings Inc.’s PayPal or manually enter credit or debit card information. All these methods are offered via the Facebook Pay payment system. Shop Pay, which stores credit card and shipping information to speed online checkout, hasn’t previously been available outside the e-commerce stores of Shopify clients.

Narrowly speaking, this is a huge win for Shopify; I fretted last year in Platforms in an Aggregator World that Facebook Shops would be bad for Shopify, in large part because it would limit usage of Shop Pay, itself a part of “Merchant Solutions”, the company’s biggest growth driver. I’m very happy to have gotten this wrong!

At the same time, from a big picture perspective this is clearly a case of Shopify, one of the most exciting companies in tech and the seeming leader of The Anti-Amazon Alliance, effectively moving into Facebook’s garden, because the web is increasingly a barren wasteland for small businesses. The cause is Apple: its approach to cookies makes platform-based web storefronts increasingly difficult to monetize effectively (Shop Pay performed magic in this regard), and its attack on “tracking” — which goes far beyond the IDFA — makes it increasingly impossible to acquire users in one place and convert them in another. The solution is to do user acquisition and user conversion all in one app — i.e. on Facebook — which is why Shopify is helping merchants move off the web and onto Facebook.

Again, it’s a good solution for Shopify, and Facebook deserves credit for recognizing that Shopify is a complement to their service, not a competitor, but I find it disappointing that once again elevating privacy above every other tradeoff is entrenching Facebook, the biggest incumbent of all.


The most frustrating aspect of the entire privacy debate is that the most ardent advocates of an absolutist position tend to describe anyone who disagrees with them as a Facebook defender. My motivation, though, is not to defend Facebook; quite the opposite, in fact: I want to see the social networking giant have more competition, not less, and I despair that the outcome of privacy laws like GDPR, or App Store-enforced policies from Apple, will be to damage Facebook on one hand, and destroy all of its long-term competitors on the other.

I worry even more about small businesses uniquely enabled by the Internet; forcing every company to act like a silo undoes the power of platforms to unlock collective competition (a la Shopify versus Amazon), whether that be in terms of advertising, payments, or understanding their users. Regulators that truly wish to limit tech power and unlock the economic potential of the Internet would do well to prioritize competition and interoperability via social graph sharing, alongside a more nuanced view of privacy that reflects reality, not misleading ads; I would settle for at least admitting there are tradeoffs being made.

I wrote a follow-up to this article in this Daily Update.


  1. In my experience word-of-mouth is more powerful than Twitter; both benefit from and build on Stratechery’s freemium approach where Articles like this are free-to-read on the web 

  2. More on questions of infrastructure, which apply to the Parler situation, soon 

  3. Be kind; the map in that article is from 2013! 

  4. Shopify reports GMV; Amazon’s are based on estimates