Introducing the Encyclosphere

This is the text of a speech I gave yesterday (October 17, 2019) at TheNextWeb’s Hard Fork Summit in Amsterdam.

For now, you can go to Encyclosphere.org to sign up for news of the project. We’ll also give you opportunities to get involved within the next week or two.

Update: I’ve added a video (produced later, after giving the speech).\

Update 2: This speech is included as Chapter 10 of my 2020 book, Essays on Free Knowledge.


Here’s a video of the speech.

We are fed up.

After ten years of domination by big social media—which might finally be in decline—we are tired of giant Silicon Valley corporations using us contemptuously. We still remember an Internet in which we charted our own destiny and owned our own data.

It’s not just social media. It’s Wikipedia, too. If you want to participate in the world’s largest encyclopedia, you must collaborate with a shadowy group of anonymous amateurs and paid shills on exactly one article per topic. If you’re new, you won’t be treated very nicely. If you don’t play their strange game, you’ll be summarily dismissed. Like the social media giants, Wikipedia has become an arrogant and controlling oligarchy.

Like Facebook, Wikipedia is also controlling its readers. It feeds them biased articles, exactly one per topic, does not let users give effective, independent feedback on articles (you’re forced to become a participant if you just want to give feedback) or to rate articles. They have, in a very real way, centralized epistemic authority in the hands of an anonymous mob. This is worse than Facebook. At least with Facebook, Congress can call Mark Zuckerberg to testify. There isn’t anyone who is responsible for Wikipedia’s content—certainly not Jimmy Wales. The situation is, in some ways, more dire than with Facebook, because you can’t effectively talk back to Wikipedia.

The old proverb tells us that knowledge is powerful. More specifically, authoritative statements of what is known on various subjects are powerful. How? Such statements can be used to influence elections, justify policies, and articulate controversial points of view—in effect to gain, wield, and build and consolidate power. The power to declare what is known is nearly the power to rule the world. No small group—no person, corporation, oligarchy, or cadre of insiders—should wield such power.

We believe in democracy: we believe that political power is best spread out, not concentrated in the hands of a few, where it is apt to be abused. We should also believe, therefore, in epistemic democracy: the power to declare what is known should also be very widely distributed.

So it should not be concentrated in the hands of Wikipedia, Facebook, Twitter, The New York Times, or any such exclusive group. The history of publishing, including Internet publishing, makes all too clear that the authority to declare what is known is wielded by selfish, powerful interests to advance their own agendas, which always unsurprisingly have the effect of consolidating their own power.

We don’t have to tolerate this. We don’t have to be at the mercy of these people.

A few thousand people work regularly on Wikipedia. But what if millions more—orders of magnitude more than are working on Wikipedia—wrote encyclopedia articles and rated them, as part of a completely decentralized knowledge network, with no individual, group, corporation, or government in charge of the whole? That is surely possible. There are surely that many people who, if given the freedom to do so, would be highly motivated to volunteer their time to add to the world’s largest collection of knowledge.

We could create a knowledge commons, defined by neutral, open, technical standards and protocols: a network that decentralizes encyclopedias, exactly as the Blogosphere has done for blogs.

The Encyclosphere

Blogs give everyone an independent voice. All blogs taken together are called the “Blogosphere,” but there is no single, central blog repository and no blogging authority. It’s a good thing, too. Can you imagine what it might be like if all our blogs were ultimately controlled by a giant, powerful organization like Facebook, Twitter, or Wikipedia?

What made the Blogosphere possible were technical standards for formatting, sharing, and interlinking blog posts: the RSS and Atom specifications. The nontechnical basics about these standards are easy and important to understand. They are simply a way to format info about blog posts in a consistent, machine-readable way, and to let bloggers alert the world when their blog has changed. In general they allow for an organized type of interconnected, networked activity—blogging—without a central, controlling body.

Plenty of websites, like WordPress.com (currently the leader according to Alexa.com), Tumblr, Medium, and Blogger.com, have tried to become the home of blogging online. But none has been able to gain exclusive dominance, because it’s just too easy to move your blog elsewhere. The existence of common blogging standards makes that possible.

We need to do for encyclopedias what blogging standards did for blogs: there needs to be an “Encyclosphere.” We should build a totally decentralized network, like the Blogosphere—or like email, IRC, blockchains, and the World Wide Web itself. The Encyclosphere would give everyone an equal voice in expressing knowledge (or claims to knowledge), and in rating those expressions of knowledge. There would be no single, central knowledge repository or authority.

So, considering that RSS and Atom enabled the development of the decentralized Blogosphere, we clearly need to develop technical standards for encyclopedias. That is the mission of a new organization I want to introduce: the Knowledge Standards Foundation. (Note, the website of the future Foundation will be Encyclosphere.org, while our Twitter account is @ks_found.)

Writers and publishers would be able to post feeds of encyclopedia articles (or metadata about articles, and ratings of articles). App developers would be able to collect the data from all of those feeds and use the data to construct massive search engines, and other neat features, for all the encyclopedia articles in the world. No one app would be privileged, but all would tap into—and help build—a “knowledge commons.” Ultimately there would be a massive knowledge competition to best express human knowledge on every topic and from every point of view.

There’s never been anything like this. But if we get together, we can build it. Nobody’s stopping us. We need only the desire to get it done. We’ll never run out of runway because it’s not a startup. It’s a distributed, collective project, an open source movement that is bigger than any of us—and certainly much bigger than the Knowledge Standards Foundation, which will serve only as the catalyst, not the owner. The Encyclosphere will have no owner just as the Blogosphere has no owner.

Epistemic power should be spread out among the public. But how? I call it the “Encyclosphere,” but how would a more democratic Encyclosphere work?

  • Writers should be able to publish their own articles wherever and whenever they want, without asking anyone.
  • Raters—the general public, including people identified as experts—should be able to rate those articles.
  • The data for both articles and ratings are published according to standards, or a single common format, in a feed, similar to an RSS feed.
  • Users should be able to sort and re-sort articles according to all ratings, or selected ratings.
  • The control over whose ratings to pay attention to should always be in the hands of the user.
  • The data is slurped up and aggregated into different databases, including distributed databases such as IPFS, and open APIs.
  • Many competing apps, all around the world, use the aggregated data to build encyclopedia readers according to their own editorial standards. The Foundation’s technical standards will be completely neutral with regard to such editorial standards.

This is not a completely new concept, but I’m sure it will sound somewhat confusing. So I want to try to clarify by listing a few things that the Encyclosphere is not, or will not be:

  • The Encyclosphere is not an encyclopedia. It’s a network of encyclopedic content. It’s no more an encyclopedia than the Blogosphere is a blog.
  • The Encyclosphere is not a platform or network for building encyclopedias. It will be basically just a series of feeds. It’s not a piece of software or a library or API you can build on. It’s an old fashioned Internet network.
  • The Encyclosphere is not a blockchain. You could put it on a blockchain, sure, but it will be built directly on the World Wide Web.

By building the Encyclosphere, we, all of us little people, can, in a decentralized and democratic system, do an end run around giants like Google and Wikipedia.

The Knowledge Standards Foundation

This is the vision I’ve had for encyclopedias since around 2014. That was when I first started talking about something I called “GreaterWiki”; I even started learning to code more seriously partly in order to execute the vision. I went to work for Everipedia, the blockchain encyclopedia, in late 2017 with the promise that I’d be able to work on this project. When I joined the startup (three years after the co-founders began work on it), one thing we discussed would be the necessity of creating a nonprofit organization holding technical standards for encyclopedias. I thought that heading up such a foundation was a job I’d like to have.

For almost two years, I’ve been developing and promoting this vision (and related ideas, like decentralizing social media) as CIO of Everipedia. I’m grateful to Everipedia for the opportunity to develop and share the plan. But now it’s time for me to get to get serious about actually executing the plan. And for that, I’ve decided to get that independent foundation started.

Therefore, I am announcing that I have left my position as CIO of Everipedia to start a new Knowledge Standards Foundation. To demonstrate that the Foundation and Everipedia are independent entities, I have given back my equity to Everipedia—without compensation, i.e., they didn’t pay me for my returned equity and I did not receive or cash in any IQ tokens.

Everipedia has already committed to being among the first or the first to use the open standards that the Foundation develops, and I will continue to work with Everipedia’s technical team—along with other reference publishers and the general public.

The Foundation’s purpose will be to publish technical standards for the Encyclosphere. We will host open source tools and other software mainly for the developer community. And we will serve as a neutral public forum for discussion of such standards. We will be mostly a volunteer organization. Already over 40 people have stepped forward to help. I expect many more volunteers in the coming months.

There are also a few things that the Foundation is not, or will not be doing.

It is misleading to call the Encyclosphere “a project” of the KSF, insofar as that implies a centralized development project. We just want to be the organization to get the ball rolling and to articulate the encyclopedia specification.

The Foundation is not itself developing an encyclopedia. There will be no KSF Encyclosphere reader. We want there to be lots of competing reader software, just as there are competing blog readers.

The KSF is not an industry consortium; it is not a project paid for and controlled by reference publishers. I will have an announcement about how we’ll raise money for our modest operations next month.

I and future Foundation staff and volunteers will confer with the leadership and technical teams of a number of different app developers, standards experts, online reference publishers, and other potential stakeholders—including, of course, anyone from the interested general public. We will develop draft standards together, while vetting them in a very public, open, civil, and moderated process. As we develop software, we will host it in a Git repository controlled by the Foundation.

If you are interested in learning more, or even getting involved at this very early stage with the Encyclosphere project, please go to Encyclosphere.org and add your name and email address to our mailing list.


by

Posted

in

Comments

Please do dive in (politely). I want your reactions!

42 responses to “Introducing the Encyclosphere”

  1. Um, https://encyclosphere.org/ is a blank white page with literally no content. When I view source there’s not even a single character. Something must be broken. I tried with a couple browsers from multiple IPs.

    1. Thanks for the report. I was in the air when it was down. It’s back up now.

  2. Nameless

    Question: What does this do that Everipedia and Citizendium couldn’t? And how cannot be both decentralized and still force everyone to adhere to a set of standards?

    Didn’t you say Wikipedia’s problem was a disrespect for authority and expertise? If anything this sounds like it would only amplify that problem.

    Also, how is handing the work of making an encyclopedia to the very mob you denounce far more completely than Wikipedia ever did going to accomplish anything?

    1. I see that an anonymous coward is following me around. Come on, you can share your identity. I did!

      Let me address the questions anyway, because they actually aren’t bad questions and they give me an opportunity to explain the imagined system some more.

      First question: This was Everipedia’s plan. But once it became clear to me that the standards could not credibly be championed by any publisher—because that could hand important authority to a competitor—I decided to do something I thought would have to be done eventually anyway, i.e., to start a neutral nonprofit. Citizendium, Wikipedia, and other encyclopedias and reference publishers could try, but they would run into the same problem.

      And “still force everyone”? I don’t get that. Nobody would be forced at all. If they want to participate in the Encyclosphere network, they’d adopt the standards.

      Second question: Yes, I did say Wikipedia has a disrespect for expertise. Would this amplify the problem? No, it will directly combat it, by supporting a decentralized global rating system that will, among other things, allow people to be identified as experts (by other people and organizations). Thus we will finally be able to have a nonprofit, open, volunteer project that enables everyone to see which articles are best, according to actual expertise. I first asked for this feature in 2001, by the way.

      Third question: This is incoherent. First establish that publishing open standards, which anyone can use to submit encyclopedia articles to the general public, has the effect of “handing the work of making an encyclopedia to the very mob [I] denounce”?

      I introduced the idea of crowdsourcing an encyclopedia to the world. I defended the Wikipedia model in its seminal first months and years. The idea that the general public can and should participate in the development of an encyclopedia is one I have never abandoned. I have always loved that idea.

  3. Tim Davenport

    So which “years” were they when you “defended the Wikipedia model?” I get 2001, when you were on the Bomis payroll. How about 2003? Can you point me towards a public quote you made in 2003 in which you “defended the Wikipedia model?” How about 2005?

    1. Aha, so the Anonymous Coward is “veteran Wikipedian” Tim Davenport, a.k.a. Username:Carrite.

      I was giving entirely positive interviews to the press (so, look at the press history, if you’re interested) from 2001 through 2004, which is approximately when I started saying a few slightly negative things about aspects of Wikipedia. I came out with my first mostly-negative commentary about Wikipedia at the end of 2004, reproduced here. I continued to tout most aspects of the Wikipedia model (which I am more responsible than anyone for shaping, you know, Tim), through the Citizendium years. I gave plenty of speeches in 2006-8, text to be found on this website, in which I explained in detail how and why “radical collaboration” works as well as it does. I still support many aspects of the model.

      Tim, if you actually look into the early history of Wikipedia, which you obviously haven’t done, then you’ll discover that I was the first and most vocal proponent of the model I did more than anyone to invent. This wasn’t just because I was employed by Bomis, it was because, well, I invented it and I liked it. But the model I invented was allowed (by Jimmy Wales) to be twisted into a platform for mob rule and, eventually, the chaotic, contemptible oligarchy it is today. I’ve explained what went wrong in many places, but probably the one you most need to read is this (two parts).

  4. A fundamental difference between blogs and encyclopedia pages is that the latter requires a quality rating in order to be useful. One blog feed has the same value as another blog feed since blogs appeal to different audiences. But an encyclopedia page is aiming to be a source of truth and thus it’s important to know when page X is a better source of truth than page Y when it comes to topic Z.

    Since you’re aiming for decentralization the rating must ideally be crowdsourced in some way. More importantly, it needs to be stored somewhere that’s also decentralized and where it cannot be modified without consensus. I see a blockchain appearing as a solution to this.

    1. That’s probably because you associate the basic concepts supported by the Internet as a whole with blockchain. But everything you want can be achieved, without the involvement and centralizing push of filthy lucre, using basic Internet tools and concepts.

      Still, this isn’t to say we can’t put the Encyclosphere on the blockchain. The Everipedia guys have told me they plan to be among the first.

      1. > But everything you want can be achieved, without the involvement and centralizing push of filthy lucre, using basic Internet tools and concepts.

        Technologically yes as long as we don’t care about objective consensus of time ordering of publishing.

        Yet holistically I disagree. At least two challenges:

        1. Spam.

        2. Funding applications *and* archival repositories. There’s too much decentralization planned for donations, subscription models or advertising to work effectively. How will applications funnel funding to repositories which are independent from applications?

        1. I will write down now an idea for a solution to the problem which I have posited.

          Note I’m still waiting for Sanger to pull out of the moderation queue some other detailed comments I left on his more recent blog, especially the detailed comment relating the failure of Knol to the generative essence of the fundamental economic problem I have also initially posited here. I want to believe that perhaps he hasn’t seen those comments because they contained links (and not believe that he censored my more critical comments). I know for example that WordPress’ Bayesian filter typically sends any comment with two or more links to the spam folder of his moderation engine.

          I think Sanger correctly believes that allowing each reader to choose their own curators (i.e. essentially my “like-minded” idea[1]) is a complete solution to the spam (i.e. relevance) problem.

          The problem is curators I choose may not have an incentive to wade through the spam to issue likes for new content — especially from new authors who don’t have a reputation and thus indistinguishable from an unbounded quantity of spam (due to publishing being entirely free).

          So the solution is quite simple actually. As an (especially if I’m a new) author I should offer to pay for the ‘likes’ from curators which I choose as good curators (or which are automatically clustered to my preferences as expressed via my likes and dislikes[1]). I believe the offered bounty should be paid even if the curator dislikes the content.

          Users then have a monetary incentive to read new content and express their opinion of its quality and relevance (with likes and dislikes). Thus there’s an economic model that rewards relevance over spam or one-sided capture. Because any well funded cabal which attempts to incentivize curators to like one-sided content (e.g. capture by the leftist religion as is the case of Wikipedia) will find that curators do not want cut off their revenue stream by liking such content because the readers would then stop choosing such corrupt curators as their preferred curators (even if this is done automatically with a clustering algorithm[1]).

          [1] But perhaps Sanger has not thought of automatic clustering as I had proposed in 2016:

          (link withheld to avoid censorship)

          P.S. apologies for the typos of your last name in one of the other comments I made. I’m suffering stage 3 or 4 dementia/delirium due to “liver disease”.

  5. If the big “sale” point of your project is allowing human beings to blog once and publish everywhere, then you have the whole “industry” in your side of the fight for freedom!
    Because this world do really needs an open standard for blogging. Plain HTML, Markdown and so on were useful in the past but things in technology are getting more and more complicated at an insane phase.

    Say for example, a Wikipedia page should follow some structure, and we can’t just copy&paste a Wikipedia page into a WordPress blog without spending some time re-formatting the source code. If we have the right standards we can break technical barriers, so that bloggers can focus on what they love and truly own their content (in terms of being able to distribute, modify and protect their hard work!)

    I wish that many CMS Open Source projects like Drupal and WordPress join this cause! Along existing blockchain based platforms like Everypedia, Steem and their fast followers. Time will tell!

    Best of lucks Larry!

    1. Thanks, Fernando. Encyclopedia writing via blogs is one of the bigger draws, I think, but also the increased attractiveness of contributing via alternative collaborative projects—all the alternative wikis—and letting groups of experts get together to develop encyclopedias together and have a real say without answering to the publishing “leaders” in their fields.

  6. What a brilliant idea.

    Community-maintained versions of all human knowledge, on open standards so everyone can decide which versions of the truth they want.

    One great way to help bootstrap it would to get a copy of a big existing content source (e.g Quora) to be continuously published to the Encyclosphere, where lots of different versions of the content could fork off from. Quora would still maintain a great app for viewing this content, but it wouldn’t be the only way anymore.

    I think it would be worth reaching out to Adam d’Angelo. Having made his billions at Facebook, he may feel some remorse and be interested in helping out. This vision is very close to Quora’s:

    “Ultimately there would be a massive knowledge competition to best express human knowledge on every topic and from every point of view.”

    1. Ernest, thank you for your kind words and interesting ideas. I don’t think Adam d’Angelo would be interested, for various reasons (among which is that I absolutely excoriated Quora here on this blog).

      You’re absolutely right that “bootstrapping” the project would involve scooping content from various sources, but since Quora doesn’t actually have many of what can be described as “encyclopedia articles,” maybe it wouldn’t be the best.

      The idea I’m working with, actually, involves indexing (at least) and mirroring (the open content stuff) encyclopedic content already online. But as this is a decentralized project, the idea is actually organizing people to do this themselves and then distributing their stuff in feeds. Aggregators, or perhaps a torrent network or the like, will make it all easily available for apps to be written on.

  7. […] to best express human knowledge on every topic and from every point of view,” Sanger described in a speech last […]

  8. […] competition to best express human knowledge on every topic and from every point of view,” Sanger described in a speech last […]

  9. […] competition to best express human knowledge on every topic and from every point of view,” Sanger described in a speech last […]

  10. […] to best express human knowledge on every topic and from every point of view,” Sanger described in a speech last […]

Leave a Reply

Your email address will not be published. Required fields are marked *