The Socratic Web refers to the web as you know it, but all content that has ever been critiqued, contradicted, or argued against, provides easy access to those critical responses.
For example, whenever anyone sees this article in their Facebook feed, in their search results, or in their browser, they will be able to access this list of critical articles. The critiques are inseparable from the critiqued content.
It is simple. It is scaleable. It is trustless. Its consequences are profound.
The white paper below will outline a specific version of The Socratic Web that currently makes the most sense to me. I believe this vision is simple enough, and sufficient to get the job done satisfactorily. However, it is true that there are numerous different ways this “Socratic Web” concept could be implemented. Others will be better at certain things, but suffer in other ways. For example, sentence level annotation provides more fine grained control, but makes it significantly more difficult to cover everything and track all of the ‘claim-response’ relationships.
Regardless of how many different ways The Socratic Web could be created, what really matters is that it is about critical responses. If the response does not take a contrary critical attitude to the page it is responding to, then it is inappropriate for the system. There is no room, nor reason to cover “agrees with”, “also saids”, “more details” or any other sort of relationships of the type. These sorts of relationships are already well covered by the way content is constructed naturally. ie: Authors link to content which supports their position.
I expect that the version I am describing here will simply be the first version, and more nuanced control will evolve over time as this concept gains recognition and attracts more resources.
The white paper is broken up into the following sections:
- Overview: What is The Socratic Web?
- Benefits: Why build the Socratic Web?
- The Use-Case: What does The Socratic Web look like?
- Core Technology: What exactly is The Socratic Web?
- Organisational Structure: Who runs The Socratic Web?
- Implementation: How do we make this a reality?
- Challenges: What obstacles are known in advance, and how do we manage them?
- But, what about…? : Addressing common objections
1. Overview: What is The Socratic Web?
The Socratic Web is based around a simple database of URL pairs where one URL critiques, corrects or contradicts the other.
ie: URL 1 critiques URL 2.
Who Builds the Platform
This database should be built by a non-profit consortium in collaboration with key web organisations like W3C, Google, Facebook, Mozilla, Microsoft, Twitter, and so on. The consortium will work with these organisations to ensure the Socratic Web system meets the needs of all major web organisations. For example Facebook et al would need to know that the content provided by the system would not be age inappropriate, would not have malware, and would have some assurance of quality before they would be willing to publish those links on their websites.
The organisation will then build the database, its core functions and algorithms, and the APIs that allows third parties to interact with it, all as an open source project for everyone to use. Most users will have free access to the data, though super-heavy users like Facebook, Google, Twitter, etc will likely need to pay access fees to help cover the operational costs of the system.
What the Platform Handles
Along with storing the URL pairs the system will need to handle 3 other important functions:
- Grouping functionally identical content into single entities so that a critique of one URL is automatically applied to all duplicate content (for example, compare the content of the pages linked from the left hand column here).
- Removal of inappropriate submissions (spam, or otherwise anything which is ‘not a critique’).
- Sorting the critiques to get the best critiques to the top of the list of any given claim.
The system will not manage any input systems or user-interfaces other than the APIs that third parties will use to submit new URL pairs, vote on critiques, spam-flag critiques, and display critiques on their own websites, platforms or software.
For example, Facebook will be able to display critiques alongside all links published in newsfeeds and comments by querying the database. How they choose to display this content is entirely up to them and their user research. Similarly Mozilla, Google and Microsoft could incorporate the data into their browsers, providing a simple button in the browser to access the list of critiques to any page presently being viewed.
On the submission side, organisations like Snopes or PolitiFact, or for that matter, any other critic of any nature is free to submit their (false) fact checks, their debunkings, their criticisms, and their corrections to the system through the API, or through the various systems built by other third parties in the form of browser plugins, wordpress plugins, or whatever.
Voting up good critiques and flagging inappropriate content as spam will also be managed through an API. Exactly how this will be managed will require iterative experimentation during the building and expansion phase because this is the most mission critical part of the whole system, and most prone to manipulation.
This leaves us with a final vision of a non-profit consortium managing an open-access, open-source central database of all critiques and their connections to the content they critique. Major web platforms, browsers and websites choose to publish these relationships through their platforms as a way of addressing the significant problem we all face with online misinformation and disinformation. All authors of critiques and corrections submit their content to the database as a way of gaining traffic directly from the content they critique.
The Socratic Web is now in place and it is virtually impossible to avoid seeing criticism of any idea you encounter online, providing an easy way out of every filter bubble and a guaranteed ever-present reminder to be skeptical of what you see, read and hear.
2. Benefits: Why build the Socratic Web?
Fighting Misinformation with Editorial Control is a Mistake
Misinformation is a serious problem, but most efforts to resist misinformation are even worse.
Most efforts to fight online misinformation attempt to use the same tools we have always used: editorial control and authority. Unable to control the web itself, most efforts to ‘fix’ the problems of misinformation have focussed on the key portals to the web: Facebook, Google, Twitter etc. The platforms which allow us to find and discover content. If we can filter out the bad and promote the good, then we have solved the problem, the rationale goes.
I have previously explained some of the problems with this approach here, and recommend reading it if you haven’t already.
In short, these approaches push these ostensibly neutral platforms closer to editorialised publications, motivating people to pick and choose them in the same way we pick and choose our favourite news outlets (Fox News? The Guardian? The Daily Mail? Huffington Post?), resulting in a serious case of global ideological echo-chambering the likes of which the world has never seen. Most problematically, dividing us up along ideological lines online despite the fact that we all live and work together in local communities.
Secondly, these approaches promote an atmosphere of intellectual laziness. Why think critically about content you encounter online when the web giants have already screened it for you? If this approach is successful (and that is a big if, considering the difficulty of identifying “fake news” and correcting misinformation and controlling for confirmation bias, etc), then our children and grand children will grow up in a world where they will be lulled into a false sense of security around everything they read online. This is the exact opposite to what we want as a society. We want to promote critical thinking, not remove the need for it.
- Facebook promised to tackle fake news. But the evidence shows it’s not working
- Tagging fake news on Facebook doesn’t work, study says
- Shane Greenup’s talk at the Post Truth Initiative on The Socratic Web
It will Help Society Become Less Wrong
One thing which has been made abundantly clear throughout our entire philosophical history, is that we have no way of actually knowing what is and is not Objectively True.
In place of absolute truths, we have repeatedly discovered that we can approximate truth by iteratively moving our beliefs further away from error. That is, we actively work on making ourselves less wrong.
Socrates was the first to employ this concept, while Sir Karl Popper re-imagined a version of it as the ideal of the scientific method. In both cases critical examination of ideas is the highest order, and disproving false beliefs and theories is the best we can ever hope for. Neither approach ever results in anyone finding “The Truth”, but by constantly discarding false beliefs and hypotheses, both approaches help us become incrementally less wrong.
The end result is the ability to hold something as ‘functionally true’ until we can improve upon our understanding at a later date. Isaac Asimov wrote a brilliant essay about this called “The Relativity of Wrong” which perfectly encapsulates how becoming less and less wrong with time is a perfectly valid way of coming at ‘truth’.
The Socratic Web vision allows us to programmatically implement this approach to knowledge into the web as we know it.
It will Improve Critical Thinking ‘By Osmosis’
An indirect benefit of arranging the Internet according to the Socratic Web vision is that everyone will be constantly exposed to critical reflection of every idea they encounter. It won’t be forced on anyone, and it won’t be ‘taught’ to them, but it will be ever present and accessible. As such, for the first time in human history, critical thinking will be normalised in society.
Just as the improved availability of information through the web have had an incredible impact on the world, I believe this change into a ‘socratic web’ will have at least as profound an impact on the minds of the citizens of the world.
It will Challenge Dogmas and Epistemic Bubbles* (Previously Echo Chambers)
An internet which forever provides access to critical analysis of whatever content you are viewing is an internet where epistemic bubbles* can never fully control someone’s information access.
The Socratic Web vision of the internet is one where no claim is ever presented as an absolute truth which can never be challenged, and an internet where no one can ever be so thoroughly manipulated by just one world view or ideology that they fall for it out of lack of alternative options.
* I had previously used the term ‘echo chambers’ in this section, as most people would have, but now prefer the definitions given by Chris Nguyen as they seem more nuanced and useful to me. The Socratic Web would easily do away with epistemic bubbles, but echo chambers would be much more resistant because they actively resist alternative arguments by design.
It will Reward Good Journalism
Journalism is struggling to cope with the changes brought about by the hyper-personalised flow of content from around the planet. The revenue from online content doesn’t compare to the revenue from print in decades gone by, and journalists are under incredible pressure to produce more content than ever before, and get it out faster than anyone else on the planet.
The current situation too heavily rewards people who produce clickbait headlines and confronting, upsetting articles which are likely to get shared more on social media. Accuracy and nuance are barely rewarded, if at all.
Appealing misinformation is often shared more readily than factual information, and corrections rarely manage to reach even small fractions of the number of people reached by the original misinformation. Combined with the fact that misinformation and low quality journalism takes significantly less effort, skills, time and money to produce than the high quality fact checks which correct them.
The Socratic Web allows high quality critiques to be attached to the low quality misinformation and the nuance-lacking hack journalism that they respond to. Once connected in this way, they will get to ride on the popularity of the low quality content. The more viral the misinformation goes, the more people will click through to see the critiques of that content.
The success of the correction(s) will directly correlate to the success of the misinformation. Compared to the current arrangement which has no formal association between these two halves of the equation, this will create a significant improvement in ensuring quality journalism is rewarded.
- How Fake News Goes Viral: A Case Study
- This Analysis Shows How Viral Fake Election News Stories Outperformed Real News On Facebook
- Planet Money Episode 739: Finding The Fake-News King
Scams and Hoaxes Will be Destroyed
While most disagreements and conflicts may be complex and nuanced, there are an almost endless number of scams, hoaxes and complete fabrications being pushed onto the population which can be easily revealed by even a superficial look. Unfortunately, these scams and hoaxes work because enough people forget to look before acting. These scams are emotionally manipulative, provocative and create a sense of urgency which often overwhelms us, causing people to act without thinking.
Having easy access to debunkings and critiques of these scam operations (from wherever you encounter them!) would do incredible damage to their business model. If people don’t need to think about going to Google and searching for reviews or critiques, then the majority of the challenge is already removed. If the critiques in the Socratic Web are guaranteed to actually be critical (unlike Google search results and social media comments), then most of the remaining part of the challenge is also covered.
The only limitation is that the socratic web won’t force people to notice the critiques that are made so easily available, nor can it force people to accept the critiques.
Those Unfairly Targeted Will Have Equal Opportunity to Reply
Legitimate businesses are frequently the target of smear campaigns and slander. When this happens, there is very little that businesses can do to address these slanderous accusations other than pay a lot of money to SEO consultants and reputation management companies who try to drown out the negative posts and accusations with positive stories about the company.
The Socratic Web arrangement will allow companies and individuals who have been unjustly targeted to respond directly to the critiques wherever they are found, without needing to respond to each individually, without possibility of censorship, and without needing to register and participate at websites which are often part of the racket (e.g. requiring financial payments to ‘adjudicate’ the disputes or to publish responses).
A Better Way to Explore and Learn About New Topics
The Socratic Web will add a whole new dimension to how we navigate topics of interest online.
We currently find content based on keyword relevance or social connections/recommendations, simply hoping that they have given us the best content possible on the topic of interest. Unfortunately both of these methods involve amplification of our existing biases through social confirmation, algorithmic filters and the way we phrase the searches we use to begin with.
The Socratic Web adds a brand new dialectic dimension to the process of discovering and exploring content of interest.
You would start your investigation as usual, through search or social media, and then you have access to a series of articles which will follow the natural claim — response — counter response flow of a discussion between two (or more) people who hold contrary positions. Each response in this series will be directly relevant to what you are interested in learning about, quality controlled by an engaged community, and critical of the arguments which came before it.
I believe that this process is far more likely to yield a stronger understanding of the topic than only being handed a single perspective, and more likely to create more valuable transferable skills in the process of learning, like critical thinking.
It is worth emphasising that the benefits of the dialectic dimension, as described here, would be complementary to the existing system. That is, you would still search for articles which argue for particular positions, and are free to stop there, or find multiple such articles and gain what you wish from those types of articles. The dialectic dimension would be complementary to what we already have, providing another way to engage with and learn from the content available to us.
3. The Use-Case: What does the Socratic Web look like?
The Socratic Web is the web as we know it, with easy access to the best critiques of any content we encounter. How easy that access is depends entirely on how many companies integrate the Socratic Web concept into their technology/platform, and how they design their user interface.
The lived experience of The Socratic Web will be entirely in the hands of the companies that already control how we access content on the web.
What It Could Look Like…
These are quick mockups to demonstrate how integration could be done. Obviously the companies in question would implement significantly better design than what you see below.
As a normal internet user, you check Facebook on your computer or phone, and in your news feed you see the following post on a topic you happen to be incredibly interested in:
Noticing the “…Critiques” option after Share, you click and get access to a list of articles critical of the URL linked to in the post.
You open the top two options in new tabs and the original article and start reading. Because of a keen interest in this subject for personal reasons, after reading the rebuttals, you wonder what the response to them looks like, so you look in the address bar of your browser while viewing the top response and click on the icon for critiques of that page.
You now have access to articles written by 5 different people, all on the exact same topic, providing you with a wide ranging discussion on the issues around the original study. You are significantly more well informed than someone who didn’t have easy access to those critical responses, and in a much better position to make up your own mind on what you think about it all.
Meanwhile, you mate Bob didn’t care about that topic at all, and didn’t bother reading any of the rebuttals, or even the original article, because everyone is still free to pursue their own interests however they see fit.
Also, no one was confronted with any authoritarian assertions by Facebook, Google, or any other major platform that their beliefs are objectively wrong. The opinions and arguments of the authors of the respective articles are entirely their own and not centrally decreed by some big brother figure.
Other Examples of Integration
Social media platforms like Facebook and Twitter could easily integrate this concept into their news feeds in any number of ways.
Integrating this into browsers seems a little more confronting, but as demonstrated in the example above, browsers do already have a number of interactive features in the address bar. A Socratic Web interface wouldn’t be intrusive or problematic to add there.
Similarly Google search results, for all of the minimalism that Google has championed over the years, still includes a drop down button:
The point is, as long as the database of claim-rebuttal connections exists, then companies like Google, Facebook, Mozilla, Twitter, Microsoft, and every other platform and website out there could find ways to integrate those connections into their platform in non-intrusive, but easily accessible ways.
In every instance, the exact implementation would ultimately be up to the company integrating the feature, and the decision to click on it should always be up to the user.
When no critiques are available
A recent study of Facebook’s fake news flagging system found that many users tended to incorrectly assume that unflagged content must be true. This problem would not arise in The Socratic Web because the ‘critiques’ button should become so commonplace that its absence would actually be a reason to doubt the content.
There are 3 main reasons why a critiqueable page wouldn’t have any critiques associated with it:
- The content regularly changes URLs for some reason or obfuscates its URL so that the pairing system can’t match the critiques to the page.
- The content is new and not-yet critiqued
- The content is so unknown that no one has bothered addressing it.
Since nearly all content is critiqueable, the absence of critiques will more likely prompt people into cautionary doubt rather than blind acceptance of the content as true. They would be able to search the AskSocrat.es database and see if any criticisms do exist of other similar content, or other known URLs of the page in question.
4. Core Technology: What exactly is the Socratic Web?
The core of the Socratic Web comes down to the database, a sorting algorithm, and an API.
Everything described in this section is an estimation of what the system will consist of, based on our learnings from working on rbutr. The final product may vary considerably to what is described here based on the input and advice of tech experts, and the needs of the major web platforms.
Functionally Identical Content
The core database structure of URL1 rebuts URL2 will actually be an array of URLs where the content of each URL in each array is functionally identical to all others.
For example, these three pages would all be included in one array. If any critiques are added to any one of those 3 pages, then it would appear alongside all three of them from then onwards. This page, however, would not be included in the array despite being about the same study and making many of the same points. A single critique could however, still be used against the first array and against the second slightly different content — they would just be entered into the database as two separate entries.
Identifying ‘functionally identical content’ could be (at least) partially automated with help from Google. Community moderation and feedback mechanism would need to be available to detach non-identical content from each other, and to provide oversight on this feature.
URL Query Strings and Permalinks
Thanks to query strings, a single web page can have an infinite number of URLs associated with it. Permalinks were invented to deal with this, though not all websites use them. This issue can create a range of complications for appropriately recognising critiqued webpages which have complex and/or changing URL structures. It may be necessary to implement (and manage) some site specific interventions (regex rules), or other more clever innovations to stay on top of websites which engage in this sort of practice.
Upvotes and Spam Votes
The database must also store the community feedback on the rebuttal connections. The core variables will be upvotes: “I think this content presents a compelling critique of the content it is connected to.” and spam votes: “This content is not a critique, or it is designed to do something other than persuade people away from the content on the page it is linked to.”.
These two categories should be stored separately (along with data about the voters in order to protect against brigading and manipulations) because they work quite differently. Upvotes will be used to help rank the critiques of a specific piece of content, while spam votes will be used to trigger moderation events which block that spam content from being connected to specific pages, or potentially from all pages if the same content has been flagged across multiple relationships.
An Iterative Arms Race Against Manipulation
Like Google’s search algorithm, the ranking algorithm for rebuttals of specific content will likely need to be kept private, and incrementally improved over time in response to attempts to game and manipulate the system for profit or other benefits.
However, in an attempt to increase transparency and the ‘trustlessness’ of the system, the system’s default sorting algorithm should not be the only option available. It will be possible for third party organisations to select from a range of sort options, including raw vote, most recent, or even to apply their own sorting algorithm. These choices would ideally be passed along to the end user to choose which sort method they most prefer (eg: see reddit).
One idea for consideration is to allow third-party sorting algorithms to be uploaded to the system so that anyone could access them through the API, choosing whichever sorting algorithm they most trusted. All third party algorithms submitted would require public display of their code though, unlike the default system algorithm.
Querying the Database
The most common API call will be to query the database for critiques. Submit a URL, get back a list of URLs which are critiques of the provided page. A bunch more data should be available with each URL returned, including, but not limited to:
- page title
- raw vote count
- system default sort order
- other sort order (as requested)
- date created
- created by
- spam vote count
- count of critiqued pages
- count of critical pages
Submitting New Connections to the Database
Submission requires at least one critiqued URL, at least one critiquing URL and between one and six(?) tags. Other desired information would be:
- created by (username)
- created by (ip address)
- submitted through (what system/platform/website/code submitted the rebuttal)
Modifying the Database
Voting would ideally be accessible to anyone on any platform, however there is a strong need to prevent abuse of the voting system. Exactly how this will be managed is still to be decided in consultation with the major platforms.
Editing entries, deleting links and de-grouping ‘functionally equivalent content’ may need to be handled on the AskSocrat.es website alone.
One of the great things about this platform is that it almost immediately works across all topics and all languages. Most critiques are written in the same language as the content being critiqued (example), so there is rarely any complication with languages.
However we will need to make sure the Socratic Web website handles international issues smoothly and that the system can deliver the correct charsets through the API for each global audience.
5. Organisational Structure: Who runs the Socratic Web?
I believe that a non-profit consortium is the best structure to use for the operation of this system. The intention of this project is to become a central hub which manages the flow of information and critical analyses to every person who uses the web. It is therefore crucial that the system and organisation are as transparent and ‘trustless’ as possible.
The organisation must be first and foremost dedicated to the impartial oversight of the functioning of the system, staying true to the principles at the heart of the socratic web: No one controls truth for the entire human race, all ideas should be critiqued, and the best counter-arguments are already being made somewhere online, we just need to organise the web to make it possible to access them from the claims they are responding to.
6. Implementation: How do we make this a reality?
A live ‘roadmap’ document comprehensively outlines the plans currently in place to continue pushing this idea into reality.
The plan is to crowdfund a small amount of money to create the foundation, its branding and a series of explainer videos and infographics. Concurrently with that, I will be building an advisory team of high profile individuals from the fake news, journalism, and technology spaces to support the project. We would then seek a second round of fundraising from philanthropic organisations.
A small team of exceptional developers dedicated to the vision would be hired to start working on the platform in close consultation with contacts from key web giants.
Beta testing would begin shortly after thanks to the incredible simplicity of the system and from the headstart provided by having already built a prototype system (rbutr).
The growth phase would be an iterative process of working with small websites, social media platforms, and browser extensions to integrate the data from the AskSocrat.es database into their platforms, then working with fact checking websites to help them automatically submit their content to the system. In this way we would be able to incrementally increase the reach of the network (the number of people seeing the critiques), and the number of critiques being added to the system, ratcheting up the overall reach and coverage of the system, while continually improving the reliability of the system as a whole.
Over time the demonstrated reliability of the platform will attract more platforms of increasingly larger size, which will motivate more people to register their critiques and rebuttals with the system, eventually leading to a threshold of quality assurance and content coverage that will meet the needs of the largest platforms and organisations.
7. Challenges: What difficulties are known in advance, and how do we manage them?
The two biggest challenges in building the Socratic Web will be raising sufficient funding to develop and implement the system, and connecting with the right people at key organisations so that we can collaborate with those organisations.
After crowdfunding a small amount, raising over $300,000 through philanthropic organisations is going to be the first major challenge of the project.
I have submitted applications for funding from several organisation in the past with no luck. To improve my chances of funding in the future I am focusing on improving the professional presentation of the project, while building a team of highly respected journalism and fake news related advisors.
I have already begun building the team of advisors with Claire Wardle of First Draft News being our first advisor. Adding people to the team will be an ongoing process and will happen gradually while I crowdfund the money to register and setup the official non-profit organisation and have a high quality brand and website designed for the project.
With a professional website and the beginnings of a team of well respected thought leaders in the fake news and digital misinformation space behind the project, I will also use some of the crowdfunded money to travel to and participate in relevant conferences and festivals. This will help me connect with more potential advisors, more potential collaborators, and more people from the philanthropic organisations which I will be attempting to raise funds from.
As of the writing of this white paper, I am already lined up to present at the International Journalism Festival in Perugia, Italy, April 11–15, 2018. I have several other conferences already on my radar which I will likely also attend or find a way to present my ideas at.
I am also working harder than ever to improve my own reputation by publishing more articles and more actively participating in the public discourse around the issues addressed by this system. This Medium blog is a big part of that, and a series of presentations is helping to get my ideas into the public sphere already.
Altogether, a more professional presentation of a revolutionary idea, backed by highly respected advisors and a recognisable director is my main strategy for improving my chances of raising the funds needed to get the project going.
Relationships with Key Organisations
A crucial part of the Socratic Web plan is that it ends with integration into all of the major social media platforms, browsers and search engines. I do not expected that the organisations behind these platforms will immediately agree to this, however it is important that the Socratic Web platform is built with that end in mind. Therefore, early and frequent consultation with those organisations is important to avoid building something that they cannot use.
Developing relationships with the right people at each key organisation is therefore incredibly important and failure to do so could cause problems.
Fortunately, the steps already outlined above to improve our chances of fundraising will also aid in addressing this challenge. For instance, travelling to conferences and meeting people from those organisations in person will be invaluable. Similarly, personal introductions through trusted contacts is another great way to initiate new relationships, re-emphasising the value of building a strong team of advisors who are able to connect us to the right people. A professionally presented organisation to tie the entire concept too is also very important, and so also backs up the need to complete that step first.
Technical Development Challenges
Thanks to our experience developing rbutr we already have a comprehensive view and understanding of the technical requirements and challenges of this project. One of the strengths of this project compared to most other technologies designed to create significant global change is the technical simplicity of it. There are surprisingly few difficult problems, and as far as difficult problems go, they are not that difficult.
The biggest challenge to solve is how we manage voting. With no single interface there is no obvious way to provide access to voting and user-feedback without opening the system up to the risk of innumerable fake ‘bot’ accounts managed by third party platforms. One option is to make voting only available to users on platforms that have stringent user validation, and perhaps even requiring identifying data about voters in order to prevent double voting.
Along with avoiding multiple account abuse, a second method which will be used to avoid vote manipulation is constant monitoring of vote patterns and signals in order to intervene when necessary. This will need to be an iterative process, responding to events and improving the solutions with each new experience. An extra consideration with this solution is how we maintain transparency of those interventions. I expect all moderations will be made publicly viewable and auditable to ensure there is no unreasonable manipulations by the Socratic Web Consortium itself.
Other technical challenges are barely worth mentioning here because at worst they only have small impacts on specific areas of the project, and more often than not, there are numerous potential solutions which simply need to be trialed. In most cases, the ultimate solution will be teased out through ongoing iterative experimentation.
8. But what about…? : Addressing Common Objections
I have attempted to identify the core objections behind the wide variety of concerns raised about this project over the years. I have listed them below, and responded to some of the variations of how that objection may be specifically expressed.
The core objections seem to come down to:
- Not enough time
- Too intellectual
- Won’t change minds
- Too simplistic
- Critique sorting will be manipulated or poorly managed
- This is dangerous
- Alternative Suggestions
Not enough time
“No one will read all of those critiques. People want summaries, not more information.”
- The Socratic Web actually does, in effect, summarise the Internet. It organises content in an attempt to cut through the noise to find the best quality critical responses, saving people the time of reading lots of irrelevant, off topic, and poorly argued responses.
- People don’t have to read every single article in a long chain of rebuttals for it to ‘work’. Being made aware of critiques may cause someone to be cautious before accepting a claim as true at face value. Reading the headlines alone can be informative. Skim reading articles can give people key pieces of information that they previously lacked.
- There may be some truth to this for most people most of the time, but not all of the time. Everybody finds themselves motivated to dig deeper into a topic of interest from time to time — whether it be for an assignment, or because of an online argument, or just out of curiosity — and when they do, that is when the Socratic Web will serve its purpose.
“Most people just don’t care enough to use this sort of thing.”
- Even if this were true, it is still the case that some people do care enough to use it, and so there is value for those people. ‘Most’ people didn’t care about the web when it was first available either. Culture changes in response to what is available and what value it delivers.
- Someone doesn’t have to be an intellectual to find themselves interested in a topic. This works on all topics, even ‘non-intellectual’ ones.
- Normalising critique in daily life is an important part of the vision. This change will have the greatest impact on the next generation who will grow up in a world where critical reflection is always available. They will internalise the idea that it is available and can be used, just as millennials have internalised and take for granted that you can ‘Google’ any fact that you want to check.
Won’t change minds
“This won’t work because people don’t change their minds based on facts and arguments.”
- It doesn’t have to change everyone’s mind to be of value. Helping provide good quality unbiased information from the beginning of someone’s investigation into a new topic may be far more valuable to ensuring someone forms reasonable beliefs to begin with, rather than focusing on changing the minds of people who have no interest in being changed.
- The idea that facts don’t change minds is only part of the complex story of belief.
The research on how minds can be changed is complex, confusing and often contradictory. Simplistic headlines taken from single studies which find a particular thing in a very particular situation should not be extrapolated to all beliefs held by all people all of the time in all circumstance.
For example, it is one thing to acknowledge that simply showing someone facts probably won’t change their mind, but it is a completely different thing to claim that facts have no impact on changing beliefs.
In any case, ‘facts’ are just one small part of how a compelling critique may be argued, and there is a lot more to an ‘argument’ than just the fact that people disagree about. The context of the disagreement, how the parties involved perceive each other, the tone of the communication, whether it is in public or private, and many other variables influence the outcome.
- It is possible that many of the critiques delivered by The Socratic Web will use of the best techniques available to change people’s minds. Ultimately, this is up to the creators of the critiques, and the community to identify and bring them to the top of the list.
“Won’t change the minds of people who most need to.”
- It may be true that the most dedicated believers won’t have their minds changed by even the most effective and accurate counter-arguments, but The Socratic Web doesn’t need to do that to have a positive impact. It only needs to help stop the people spreading misinformation from being able to infect other people with those false ideas.
If an idea can’t spread to new minds, then it is basically dead.
“There are more than two sides to every story.”
- Agreed. That is why this isn’t a Pro-Con debate platform. It is an argument, response, counter response system. Even the most factual positions may be validly critiqued.
- One limitation of the system is that there may be many different ways to approach a response, and we don’t discern between different approaches, so unique and interesting replies may be lost in the mass of standard responses. This is a limitation we are aware of and hope to find clever solutions to in the future when significantly better data and resources are available to do so.
“A web page level response to an entire article will likely miss key points or only respond to part of the article. This needs to work on specific claims or sentences, not on webpages alone.”
- That may be the way forward in the future, but to get started this ‘blunt instrument’ approach is the easiest to implement to get satisfactory results.
Many people have tried to create sentence level annotation plugins over the years, and yet none have gained mainstream adoption. One reason for that, in my opinion, is that it requires too much work on behalf of the users/submitters. With the web growing at an exponential rate, with many millions of webpages added every single day, expecting the web community to then also categorise and annotate each appropriate sentence on all of those pages is just too much.
Page level connections may not be perfect, but it is at least manageable.
- The work required to identify every claim, match them to their appropriate responses, and ensure a high level of quality is also a much harder problem to solve on a technical level.
Critique sorting will be manipulated or poorly managed
“But people who believe in nonsense are better organised than us and will push their nonsense aggressively into the system, crowding out the good arguments.”
- Due to the forced structure of the Socratic Web which requires that each connection must involve a critique of a claim, every agenda which is pushed, is equally exposed to critique. This is unavoidable and simply cannot be overwhelmed. Any non-critical replies will simply be removed or buried.
- Community efforts to push their beliefs into the system are encouraged. The Socratic Web works best when large groups of people put all of their best arguments forward, and work together to identify the strongest of those arguments. This is in their best interest, and also in the interest of everyone who disagrees with them. It is always better to respond to what they actually believe than to respond to a weaker version of their arguments.
- Manipulation of how the critiques are sorted is the most prominent vulnerability for the system, so we will be constantly watching for efforts to manipulate the results away from natural community feedback. Sorting algorithms will be developed and constantly improved, and human oversight will be ongoing to spot new innovations by would-be manipulators, and they will be resisted.
“What is popular is not necessarily correct / accurate /or true. Letting people vote on rebuttals won’t give you the best responses.”
“Quality will decline with widespread adoption.”
- This system isn’t designed to guarantee ‘truth’ at any individual point. The purpose is to remind people to be skeptical of anything they encounter online, and provide them access to contrary perspectives. Critical thinking skills are far more valuable (and transferable) than any individual fact accurately reported.
- It isn’t necessary to have the best critique at the top in order for the system to perform its core functions. Having any rebuttal will prompt people to hesitate before accepting the critiqued content. Being able to access an average quality critique is still better than having no critique. Having access to a poorly argued opposing perspective is better than being trapped in an echo chamber.
- It is worth noting that the system is aimed at the public, not at academics. This isn’t about pushing the forefront of knowledge, it is about communicating the best understandings that society has achieved to the greatest number of people possible. In that context, whatever response is the most popular, is in fact the ‘best’ response. Whatever response connects with the most people and makes them rethink whatever they just read, is in fact the best response. Even if it isn’t completely accurate, or isn’t as detailed and rigorous as it could be. Any factual errors or misrepresentations present should then be challenged in the counter-responses anyway, taking advantage of the system’s self-correcting nature.
“We’re going to end up with stupid memes as the top rebuttal in every case.”
- There is some risk of something like this happening. The average page may end up with simplistic, sarcastic, and/or dismissive responses rather than well thought out and argued responses. However, at this stage, this risk is pure speculation. The Internet does love memes and cat pictures etc, but they are usually found in contexts that make sense. Contexts like community boards, social networks, and image sharing sites. The Socratic Web is not the right context for them (usually), and so they will likely not become a big problem.
- Different topic areas may attract different types of responses. A contentious article about the ethics of human gene editing will more likely attract serious scientific responses, while an argument that cats make better pets than dogs might be best met with a silly meme image. The Socratic Web is just an extension of what the web already is, after all.
This is Dangerous
Before getting into the specific concerns, I want to state upfront that yes, this idea is indeed dangerous.
Any forum which encourages un-filtered exchange of ideas is a dangerous concept. The Internet is the ultimate example of that danger. The Socratic Web is just a particular type of organisation of the freely accessible information which is already present on the web.
It is dangerous, but it is also empowering.
Free exchange of ideas is probably the most important thing we, as a species, have available to us. I don’t believe we should ever allow the risks of the free exchange of ideas to stop us from reaping the innumerable benefits of exchanging those ideas.
“This will increase misinformation. People will be shown manipulative deceitful replies to valid science/journalism/facts, and they will fall for the tricks of the people with no obligation to keep to the facts.”
- The system is self-correcting. Though it may take people from factually accurate articles to confused, misinformed or propagandistic articles, it also takes people from those articles to critiques of those articles. If someone is unsure about the topic, but interested enough to take the first step, then they are likely to be interested enough to at least take the second step too.
- The amount of misinformation online is already immense. However, most of it is in ‘claim’ form. For example, ponzi scheme scammers will not waste their time trying to argue against condemnation of ponzi schemes, nor will they write critiques of valid investment options. Scammers focus their time and energy on writing convincing arguments that you should trust them and invest.
They make claims, not rebuttals.
Meanwhile, scams and misinformation attract enormous amounts of scrutiny and criticism.
This asymmetry means that the Socratic Web will disproportionately critique misinformation and scams over anything else, and statistically speaking, help move people from low quality claims to higher quality critiques.
- The foundational principle behind the Socratic Web (based on modern Western Philosophy / Science) is that we don’t know what is and isn’t true with complete confidence. Discourse and debate are the best mechanisms we have for approximating the truth. As such, we are all, to some extent or another arguing from a position of belief, rather than objective knowledge. The only thing which separates those who are closer to the truth and those who are further away from it, is that there is more evidence and more rational arguments available to the person closer to the truth. Otherwise, all else is equal. People closer to the truth are just as free and able to lie, manipulate, use clever rhetorical devices, and emotional arguments as people who are less correct.
“There is no quality control over who writes or submits the rebuttals.”
- This criticism is fundamentally aimed at the Internet. It is a major problem of the Internet that any claim and belief can be published and then shared virally through social media. But this is also one of the most important features of the Internet too.
- The community itself is the quality control. Inappropriate, rubbish or nonsense critiques will either be spam binned or ignored until they are completely out of view due to the rise of high quality critiques.
Google has no quality control over what content it indexes either — you just expect all the low quality content to be in the 100 million pages that come after the first page of results that you look at.
“If everything is critiqued, people won’t know what to believe, and will become apathetic about everything”
- I am firmly of the belief that we are far better off having a world full of people uncertain about many beliefs, than having a world full of people absolutely convinced of things which are untrue. Confidently acting on false beliefs tends to cause great harm, while uncertainty tends to lead to further investigation and education at best, or inactivity at worst.
- While there may be some basis to this concern, it seems unlikely that it will become the norm. Humans are believing animals. Providing easy access to more critiques, more evidence and more arguments around every issue seems like an unlikely way to suddenly make all people stop having beliefs.
“This will be used to sow uncertainty where there is none.”
- This already happens in society and online. This system will do more to resist that tactic than enhance it because it will make it possible for that tactic to be called out wherever and whenever it occurs. The tactic will be revealed, and thus lose its power.
“Most people don’t make real arguments, they just overwhelm with bullshit arguments”
- The Socratic Web community will help surface higher quality arguments, pushing bullshit arguments down out of sight.
- When encountering a bullshit argument through other means, it will be confronted with critiques which reveal its failings, opening way for a legitimate argument on the issue.
“What we really need is better education/better regulations/better journalism/an end to internet monopolies…”
- Go ahead!
- The approach I advocate here won’t impede execution of any of those approaches, and won’t prevent anyone from making them happen. It should in fact improve our chances of making them happen, and improve their effectiveness!
- The advantage this approach has is that it is global in impact and ever-present. It also works on nearly everyone, rather than only those with access to quality education, or those living in countries which enforce sensible regulations, or those who are lucky enough to have access to quality journalism, etc.
That’s all for now. I plan on maintaining a constantly evolving form of this white paper on the Socratic Web website when it is launched, though this version will remain as it is.
I look forward to your comments and criticisms, and highly encourage detailed critiques written offsite to be submitted to rbutr.
Want to see the critiques of this page now but don’t have the rbutr browser extension? Click here to reload this page in the rbutr frame: http://rbutr.com/https://medium.com/@Aegist/what-is-the-socratic-web-c6095c452c6