profile profile

Decentralising the Social Web in Tokyo

This is what I talked about at the W3C TAG developer summit in Tokyo tonight.

東京で2016年11月3日にて開催されたW3C TAGデベロッパー・サミットにて話した内容となります。

I am one of the staff contacts for the W3C Social Web WG. I also co-edit a couple of specs in the group, and procrastinate from finishing my PhD thesis by implementing the group's specs as part of my own website.

まずは自己紹介から、 W3C Social Webワーキンググループのコンタクトを務めています。また、そのワーキンググループにていくつかの仕様についても共同編集者をしてたりもします。そのグループの仕様を自分のサイトへの実装がPhDの論文がなかなか終わらない理由だったりします。

I'm a strong believer in owning your online presence and social data on the Web. What this means (to me) is: making data accessible through a URL that you have authority over - either you own it and host it yourself, or someone you trust does. It also means having control and choice over how and when you express yourself; working against censorship and being free to share as much - or as little - about yourself as you want. This is particularly important when we think about a how social networking works today. Big systems like twitter and facebook - silos - are hoarding the data, and creative, expressive content, and livelihoods, of millions of people. And they're mining it and using it in ways that most people are completely unaware of. If twitter goes down tomorrow - and it could, they just shut down Vine - what happens to your history and interactions? We have this huge hoard of collective digital history that we, as societies, have aggregated over the last decade, and we just gave it right up. The culture of a generation is completely out of the hands of the people who created it, and who it's important to. This could disappear at the whim of an investor, or a glitch in a data center.

私は自分自身のオンラインにおけるプレゼンスとソーシャル・データを所有することについて、強い信念を持っています。(自分にとって)どういう意味を持つか、というと、自分がオーソリティを持っている(たとえば自分自身が所有しているか、あるいは自分が信頼する誰かが所有しているような)URLからデータをアクセス可能にすることを意味します。また同じく自分自身がどう表現させるかについて制御したり、選択したりできることも含んでいます。検閲から逃れたり、自分自身のことをどれだけでも多く、もちろん少なくでも自由に共有できることも大切です。 現在ソーシャルネットワークサービスを鑑みると、非常に重要なことだと考えています。TwitterやFacebookのような巨大でサイロなシステムは何百万人ものデータや創造物、くらしそのものをそのシステムに飲み込もうとしています。その上、ほとんどの人が気づかない内に、データは使われているのです。もしTwitterが明日無くなったら(可能性はもちろんあります、何しろVineは終了するのですから)、あなたの履歴や他者とのインタラクションはどうなるのでしょうか? 我々が社会全体として、この10年間ほどで集約してきた巨大なデジタルの歴史達をただ諦めることしかできないのでしょうか。この世代の文化そのものがそれを生み出した人たち、そしてそれらを貴重だと思う人たちの手から離れて、インベスターの気まぐれやデータセンターのちょっとした不具合によって、消え去ってしまうかも知れないわけです。

This is part of what motivates me personally to work on decentralising the social web. In the SocialWG we are creating open standards for:


  • representing social content
  • creating, updating and deleting content
  • federating updates between different servers through decentralised subscription and notification mechanisms
  • ソーシャルコンテンツの代理
  • コンテンツの作成、更新、削除
  • 分散化させたサブスクリプションと通知のメカニズムを通じて、異なるサーバ間における情報更新を連合させること

The idea is that people implementing our standards using completely different technology stacks, and without any discussion between themselves, can build systems which talk to each other for some of these kinds of interactions.


Now we know not everybody can have their own website to replace social media. We also know that the big social networking sites are not going to make it easy for people who want to move their data around, or for developers to create decentralised competitors. There have been people working in this space for a looong time, and it's been an uphill struggle the whole way.


Our specs are building blocks for different pieces of the social puzzle. It looks like we have a lot, but they're small and modular to help developers pick and choose the parts they need. Rather than having to implement a whole "social network", you can decide to integrate say decentralised comments into a site you're working on, without worrying about subscription or content creation or even account signup and data storage.


We're not aiming for adoption by major social networking players (none are involved in the group) but by individual developers and smaller business for whom collecting social data is not their business model, but rather can be an enhancement for their customers of some other product or service.


Our specifications are JSON based, and use JSON-LD for extensibility. If you're not familiar with Linked Data, this basically just means using a URL as a globally unique identifier for everything, including relationships between things (which would for example normally just be plain text keys in a JSON object). Sharing URLs this way helps us know when we're talking about the same thing as someone else on the Web, so that we can integrate data across diverse sources, without having to know anything about the other data sources beforehand.


(This is vastly oversimplifying things, see to learn more)


We still have a way to go, and we'd love your feedback as developers. This was a pretty high level overview, and if you have any specific technical questions about the specs up here, I'm happy to answer them.


Finally, the WG is finishing at the end of this year, but we're keeping the momentum going with a Community Group which anyone is welcome to join, whether you're a W3C member or not.



  • (currently redirects to Social Web Protocols, more useful stuff coming soon..)
  • WG Wiki:
  • Mailing list:
  • Twitter (sorry): @SocialWebWG
  • IRC: #social on


  • Amy Guy, W3C & University of Edinburgh
  • rhiaro on IRC, and everywhere else.

🏷 decentralisation social web socialwg Tokyo, Japan w3c

Post created with