The Truly Social Net Starts With BlueSky
You’ve doubtless heard the phrase “walled garden” applied to social networks. It’s true of all the big players, and what it means in plain English is that your interactions and relationships with people on that system need to take place in that system.
For example, take Facebook. It’s all sealed. They provide you with an account; they have your login, password etc. Details of who your friends are, things you upload to share, etc, these are all held within Facebook.
Then take X. It’s all sealed (some might say THANKFULLY, like a septic tank). They provide you with an account; they have your login, password etc. Details of who your friends are, things you upload to share, etc…
You get the picture.
So if you want to interact with someone on X, you have to join X. If you want to see somebody’s photo album on Facebook, you have to join Facebook, and many people join just for that reason alone. FOMO, as millennials used to say.
This “have to be in it” factor is known as the Network Effect and - up until now - has been one of the primary attractive factors in valuing / evaluating the potential impact of any particular startup / network.
In the past, some of these mega-platforms have opened up APIs to allow a form of interconnectivity between their data and some app / website of your own choosing, but it is still very much accessing the information held in those platforms.
More than that, this very access to that information from outside is only by the grace of the platform owner and if the platform owner decides this is no longer desirable, access can be shut off immediately. (As X showed so gracelessly in 2023 with the closure of their free APIs, some might say the first Dick Move by Elon in a long long seemingly endless list of Dick Moves).
Internet Paleontology
But now, remember AOL? Remember CompuServe? Well, no, of course you don’t. Nobody is that old. But let me explain why I’m bringing these up.
Messaging. Specifically, in-platform messaging. In the beginning, these platforms allowed you to message other people on the same platform, using their usernames, but the idea of swapping messages BETWEEN platforms was just unheard of.
The same was true at your place of work, if you had email at your place of work. You could email anyone in the company. But outside? Nah. Just not possible.
If this seems impossible to imagine now then bear in mind this was A VERY LONG time ago. But it’s completely true. On CompuServe everybody was a weird 2-part number (I was 100411.1741, for example) and on IBM Work Mail everyone was a letter salad like GBLDNSMS2. And you couldn’t speak from one world to another. Such was the electronic world of the late 1980s.
Then something awesome happened.
A kind of unifying “glue” got applied to the planet in terms of email. It was magical really. The messages inside the worlds did not have to conform or standardise - CompuServe could stay CompuServe; IBM Mail could stay IBM Mail - but if at the very boundaries of their spaces, if the organisations chose to “speak SMTP”, well, then, they could pass messages BETWEEN the worlds.
Revolutionary. And what is SMTP? Well, it’s a lot of things but for the purposes of this story, Simple Mail Transfer Protocol (its full name) brings one INCREDIBLY important thing to the party: its addressing style.
Modern day email addresses - that whole yourname@somemail.com thing? That’s SMTP addressing. The part @somemail.com identifies the organisation the mail is intended for. Then, only when it arrives, does that organisation’s systems try to make sense of the yourname bit at the start.
When this was first commonly implemented, it changed EVERYTHING. Now people like 100411.1741@compuserve.com and GBLDNSMS2@ibmmail.com could speak to each other without leaving their own little worlds.
It seems obvious now, sticking an at “@” then an internet domain on the end so you could own everything BEFORE the “@” yourself but really, this was completely alien forty years ago.
But leaving the idea of email behind, we need to look at what actually happened there. WHY did this work out? The reason this worked was because of the “P” in SMTP. It’s a Protocol. Just a set of rules and formats and options for what happens when moving email around the planet. SMTP isn’t a piece of software, or a company, or an inline centralised exchange. It’s just A SET OF RULES. And if your computer obeys the rules, guess what, it can join in the conversation and move its own mail around too.
Strength through working together and mass voluntary agreement? The possibilities are endless. Who knew.
So now let’s look again at today’s Social Networks. Essentially Facebook, X, and all the others, are the modern-day equivalents of the dinosaurs in this story: CompuServe, AOL, IBM Mail. All walled in, and siloed. Incompatible. And they LIKE it that way.
ATProtocol and BlueSky
Along comes ATProtocol https://atproto.com/ . Summarising very aggressively, ATProto presents a bunch of open tools and processes for doing crucial things in social space which, until now, are all 100% in-platform. For example:
- Creating User Accounts
- Linking to other Users in some form of connection (friends, follows, etc)
- Pulling together types of content from many sources / Users
- Proving that a particular item of content came from a particular person
So now social networks can exist and pass all these things between themselves, unlocking possibilities in the same way that SMTP did at the start of the 1990s. All social networks need to do is to “speak ATProtocol” in much the same way that old email just needed to agree to “speak SMTP” at their boundaries.
Imagine a Social Network of Networks which can all just interconnect? (You can be sure that Facebook, X, et al do not like this idea - it’s an incredible loss of power for them).
Well, that’s what we see with BlueSky. BlueSky is NOT a protocol; it is in fact both an organisation and a piece of software. But CRUCIALLY BlueSky is a textbook example implementation of this ATProtocol.
So, in many ways, the point of BlueSky is not BlueSky itself. It’s the fact it’s 100% ready for the next social network to come along and plugin, whatever that’s for. And the more adoption BlueSky gets, the more users there are freely consuming ATProtocol and suddenly, we will reach a critical mass where ATProtocol is not just A Good Idea; it’s a very compelling standard to which to adhere.
Personally I certainly hope this becomes the case. I’ve argued for years now about the need to have protocols which replace the big name networks and now - FINALLY - it looks like the sky’s the limit.
Author: Mark Henwood | Mark’s LinkedIn