Friday, March 25, 2016

A proposed tool for combating internet misinformation, v3

This is the third iteration of a collaborative web platform that would allow the informed on the internet to carefully curate and cross-reference large amounts of data while simultaneously improving access to and clarifying authority of information while making the information easier to understand for people with no background in that topic and limiting or clarifying the biases in presentation and/or consumption of that information. Sounds pretty lofty, right? Well, here's what it looks like.

Read on to have this explained.
The main features of the platform, by section (marked in red letters)

A) The user profiles are designed to make it harder for special interest groups to post misleading information, and allows people of all levels of proficiency on a topic to have input in the global community of that topic.
  • Users register with their professional names, and tie in to LinkedIn or other professional sites so that there are not anonymous accounts making comments or posting articles. This adds a benefit for contributors of potential professional exposure and chances to collaborate with peers.
  • The user lists what they have knowledge in, and uploads or references credentials to back this up. This allows for the knowledge and experience of hobbyists and enthusiasts to be on more level ground with degrees and certificates, as many enthusiasts know more than recent grads, and therefore should be seen as more authoritative.
    • Credentials can be official schooling/training/certificates, experience, etc. It can be shown with photos, videos, documents, etc. That way, if somebody professes that they've built and flown model planes for years, and uploads photos of their planes and videos of them flying or scans of their math for the internal wing structures, you know that they know about it despite not having officially studied it in a school.
  • Knowledge topics would be broken down into "core principles", and based on experience, the user would be given a credential level, such as:
    • beginner->has some idea about it, but still learning
    • intermediate->fairly well along, but still has a lot to learn
    • expert->really knows what they're talking about, and if any questions come up, they'll know how to answer it
  • Core principles, topics discussed in D.
B) Pieces. This is the work that people are showing off, developing, or sharing with the world. This is designed to be a multi-layered presentation of information that can be interacted with, discussed, and categorized.
  • The platform would have similarities to, where people could post their research for free, allowing collaboration and increasing knowledge sharing. It would not be limited to documents, however, allowing for other media to be presented, similar to Hybrid Pedagogy.
  • The author of the piece, and their credential level for the topics they tagged their piece with, would be at the top, rapidly identifying how authoritative they currently are.
  • Multiple versions of documents could be stored, so readers/viewers can go back in time to see how its evolved during it's development.
  • There would be multiple levels of presentation available, typically up to 3:
    • an expert level (lots of jargon, written for others with high proficiency in that topic)
    • an intermediate level, providing those with some understanding a strong working understanding, including technical information and basically either doing away with jargon or explaining it better
    • introductory level, reducing the work to very simple terms so that somebody totally new to the topic can still get something out of it
C) Annotations and Reviews. Reviews allow for collaboration, corrections, dissenting opinions, etc, and provides a better review platform than currently exists for academics. The annotations allow users to organize information in ways that best suit their needs.
  • The pieces can be annotated (for learners) and peer-reviewed (by experts and peers). Those peer reviews would be public, and noted on the document, linking to the profile of the reviewer. Annotations could also be shared.
  • Annotations/review notes would be version and level specific, allowing the author(s) multiple avenues to receive feedback, and allowing even new beginners to the topic to offer potentially useful peer reviewing, as a slightly educated comment can be very profound, and influence even the expert version.
  • The sharability of these comments allows for better collaboration and a deeper dialogue on the information, since parts that are hard to read would be marked by people who don't like things hard to read; unsupported claims would be noted, strengthening the quality of the piece and improving authority; etc, etc. You can see how this would produce better work overall.
D) Tags, metadata, organization. These tags have multiple purposed, so I'll just let the bullets explain.
  • Topic - general body of knowledge to which the piece belongs. For example, aerodynamics, linguistics (Spanish), psychology (addition)
  • Core principles are the basic building blocks that a new student needs to know to understand the topic. For example, aerodynamics reduces to principles of lift, thrust, drag, etc; linguistics to Spanish grammar, idioms, etc; and psychology to addiction, child development, etc.
  • One purpose of these tags is so that a new student can look at a piece on aerodynamics, and quickly know that they need to understand lift, drag, and thrust to really understand the piece, and they know what to study (more on this in F).
  • The piece would list the topics (which link to the core principles for that topic) under the viewing window both to determine who's expertise is pertinent, and to help categorize the information (add metadata). Discussions, annotations, reviews, etc are organized by current expertise at the time of commenting, and knowing how much they know about the topic helps you make judgements about the validity, authority, bias, etc of their comments.
  • Finally, the tags would be another piece of metadata that can be used to notify users (depending on their preferences) about new pieces with those tags.
E) Discussion section. This is where the piece is applied to other pieces or the world in general, other information on the internet is connected to pieces, etc. These discussions can be collapsed to focus more on the piece itself, and some of the discussion groups would be limited to certain expertise levels. See the explanation of tabs below.
  • Each work would include a discussion forum similar to the ones for EdX classes, that allow explanations and evolution of the ideas, as well as a vetting ground for cross-references and other works.
  • Links is the tab to cross-reference other parts of the internet, like Wikipedia or journal articles, or blog posts, or videos, to the piece. Other users curate this information, and if some link has received enough downvotes, it is no longer shown (unless the user requests it to be)
  • Biases: This probably won't be a tab, but serves to model that problems with the piece can be discussed in a separate discussion tab.
  • Q's is a Q&A for anybody, so if you are reading a piece, and it just doesn't make sense, you can annotate what you don't get, and ask "What exactly is the author saying here <share annotation>?" and other users can answer, adding to an overall better understanding of the piece.
  • Experts would be a discussion group where only other experts on the topic could post, allowing a place for very high-level, serious discussion on the piece or topics to take place. while only experts can post, non-experts can still view it.
  • Gen disc is basically a general discussion, like any other forum, but tied specifically to the piece above.
F) You'll note I didn't put an F on the page. That's because this isn't drawn. A very critical component of the platform is an explanation section, so that beginners can quickly come up to speed, and learning about a new topic won't require hours of combing the web or reading books to begin to understand. Think of a very simplified Wikipedia, just giving you the intro section, but in even easier terms.
  • The platform would include a section that explains, simply, core principles that touch topics, and authors would be expected to tag their work with the topics (that link to core principles) that feed the work. This work would also be crowd-sourced, so other users can link it to the core principles involved.
  • The core principles are very simply explained, so that anybody uncomfortable with the idea of black-body radiation can read a few sentences and feel comfortable enough reading a discussion on climate change that uses an energy balance argument (requiring an understanding of black body radiation).
  • Again, authors can tag their pieces both with topics, which would be automatically associated with core principles, and with the principles themselves.
Other things of note
  • The platform would also offer paid and crowd-sourced services, where the author or a user could pay (or make an offer/bid) for added accessibility on a piece, which could include:
    • Translation between levels of communication (expert, intermediate, beginner), so that scientific papers can be simplified for non-experts, and a backyard hobbyist can get help "academicizing" her work
    • Translation to another language, making sharing of complex ideas easier globally
    • A definition and cross-references for the technical jargon or industry-specific key words
    • To be tagged with the appropriate core principles, though this could also easily be a part of the peer-review
    • Addition of other cross-references
    • Creation of videos, similar to the ones on one of these channels or these web sites
    • Addition of other media, video, images, etc, though this will also be crowd-sourced by people reading it
    • Mediator monitoring. Mediation will be both paid and unpaid, but the monitoring will be a paid service to somewhat regularly check in on works to make sure that the links are still valid, and that people aren't spamming by linking to many things that are very loosely connected. The date of the last mediation, and the mediator profile information, would be documented.
The point of the platform would be to increase the usability of known knowledge, and facilitate the spread and creation of new knowledge, as well as provide a powerful resource (with time) against misinformation. I picture the platform facilitating something like what Pardis Sabeti describes.
I would love to hear your thoughts on this.
This is the evolved version of this post, which fleshed out this post.

No comments:

Post a Comment