# Factland Scoping Sprint
*UX Notes from TW - October 15, 2023*

- [Product Demo](https://demo.factland.org/)
- [Factland Team's UX Research Miro Board](https://miro.com/app/board/uXjVNc_-pvo=/?share_link_id=26162775931)
- [Scoping Sesh Miro Board](https://miro.com/app/board/uXjVNaeFoSQ=/?share_link_id=506329340960)

## Primary Objectives for Day 1

1. Articulate as concrete of a problem statement as possible.
2. Brainstorm on a hypothesis for "what would need to be true" to succinctly solve this problem.
3. Align on the next steps, to be covered in Day Two.

## High Level Keystone Concepts

-

## Initial Impressions

### Landing Page

- First impression: Confronted with a list of content without context.
- Might benefit from some instruction, walk throughs or explainers to give the user general orientation about what they are seeing and what they are expected to do with it.
- No apparent organization of the topics: appear to be chronological order with no particular arrangement logic.
- Sorting functions are based on vote activity, but do not allow to sort based on interest. Implies a user that is agnostic to the content and primarily interested in tracking activity.
- More False topics are highlighted deeper read, attracting more attention to them. UX is primarily steering towards false information.
- Currently no clear indication of why the user should be invested in the activity.
- To discover answers, I navigate to the FAQ.

### FAQ

- The following will attempt to probe and question the underlying intentions of the product, in support of the goals outlined above.
- We will find a recursive loop on some issues, so best to cover all of this content in order to speed run towards that high level comprehension.

#### What is Factland?

> "Factland is a new way to sort fact from fiction. It pops the social and information bubbles that prevent people with different beliefs from talking with each other online and forming a consensus on important issues of our time."

- **Can fact and fiction be so absolutely separated?** Certainly there are degrees/gradients in between.
- What is meant by "social and information bubbles" and how do they prevent conversation? **Might be useful to define this terminology in relation to a clearly stated problem statement for the user.** "Factland addresses the problem of misinformation in web2 social media..."
- **Why is consensus the goal?** Implies a singular, universal, homogenous perspective is ideal in relation to an otherwise multiple, pluriversal, heterogeneous dissensus. There are strong political implications (biases) in this statement we might address. **How might we work towards rendering this bias as transparent as possible** and placing it at the beginning of the experience?

#### How does it work?

> "When you see something on social media that looks fishy or just interesting, post it on Factland. Anyone can share evidence to support or refute the claim and comment on the case. Use Factland tokens to bet on whether the claim is true or false. The bigger the bet, the bigger the prize. If a bet grows large enough, an anonymous jury is randomly selected to review the claim and issue a verdict based on the evidence gathered. As a member of the Factland community, you may be asked to join a jury from time to time"

- **What constitutes "fishy or just interesting"? How might we work towards converging on the kind of content that is welcome vs. content that is discouraged?** I as a user would be posting the entire internet lol. This qualifier could be more specific as to why the content may be deemed harmful and provide more direction for the user.
- **What kinds of evidence are valid? What kinds of refutations are encouraged? According to what inherent bias?** If it all amounts to comments, what distinguishes the conversation happening here from any other social media platform? Isn't it all sharing, "evidence," refuting various evidenced claims, and an infinite procession of cheap comments? **There's a key value proposition here waiting to be uncovered.**
- *Betting on truth is a big topic.* **As with any betting, what prevents the user from bribing others to swing the evidence in their favor?** This introduces a perverse (economic) incentive, but it remains unclear how this will amount to increased clarity or objectivity of truth. (Later it is mentioned as a key feature, but it is not made explicit.)
- Anonymous Jury is also huge. This introduces bureaucratic centralization and process to what appears to otherwise strive to be a democratically horizontal and decentralized consensus system. *Authority* again becomes a pain point: **who is this jury, how are they selected (sortition or vote?), what is the review process, how is evidence leveraged** *These are keystone points that provide the foundation for the whole app.*
- **How are members validated to participate in this process? What qualifies them, what qualities should they demonstrate, what qualifications are required?**
- Here is an important opportunity to introduce a trust network, P2P validfation, PoH via a Gitcoin Passport-style flow, etc.
- The core functionality can be found in the formation of a DAO. A Moloch-style DAO would close the loop between token voting, user identity/reputation, transparency, on-chain accountability, etc.
- This is a key inflection point: **does the Factland team prioritize privacy/anonymity and embrace all of the potential corruption/collusion that includes, or does the team align with a surveillance state bureaucratic order of centralized validation to safeguard the system from manipulation?** *Very important, as it will determine how decisions are made at all layers of the UX.*

#### How do I file a claim?
> "You can submit a claim on factland.org by clicking the 'Report' button and filling out the submission form. We recommend writing your title as a question. Once you file your claim, including a link to the source, we recommend you submit additional evidence to support it. You can also share links to Factland using the share button on any social media app or Web page and emailing [e-mail]."

- Bots and automation scripts: **What PoH protocol currently exists to prevent the submission process from being immediately corrupted by malicious actors wielding bots?** This is a huge issue since *there are multiple points of submission* (via the UI and via email) directly to an automated flow (apparently via Zapier). 🔴 *This is of app breaking importance.*
- Are all form fields required? Does this impose a bias for a certain kind of information that fits within this format? **We should investigate the submission form flow for various personas to understand the constraints it imposes.**
- Again, what kind of evidence is considered valid? **How is the truth of the evidence considered?** How many sources are necessary before the title of truth can be awarded?

#### What happens when a jury issues a verdict?

> "The claim under review on Factland is "labeled 'True', 'False', or 'Undecided'. If the claim involves a statement on social media, a comment will be sent to the author announcing the decision as a reply. Bets are paid out based on the size of the prize and the amounts wagered. Other activities are also paid at this time, such as serving on a jury or submitting evidence that was cited by a juror in their decision."

- **`True`, `False`, and `Undecided` are deeply unsettling and unsatisfactory labels.** Nothing in this world is so clearly differentiated without being accompanied with mountains of bias. As a user, I would anticipate labeling and re-labeling every issue as undecided. If a claim is under review, shouldn't it default to undecided?
- **What is the role of the "author" user in stewarding the lifespan of the topic through the decision making process?** Do they have additional responsibilities after the submission?
- Bets should likely *not* be paid out based on financial gamification mentality, but based on the evaluated merit of the participation, in order to avoid the above mentioned collusion. 🔴 **Token utility currently supports malicious intent, but it could be designed differently. (See below re: adversarial design.)**
- **Being paid to be a Jury user seems like an obvious conflict of interest.** There should be socio-cultural sacrifice made by this role, rather than reward. The jury members need to be validated for their biases somehow. 🟡 There seems to be missing process here that is important for understanding the curatorial governance ops.
- Same problem with being paid for evidence and an obvious collusion point between Jury users and Evidence Supplier users.

#### Why should I trust Factland?

> "Factland is adversarial by design. The bigger the disagreements, the better it works. Factland uses cutting edge anonymity technology to ensure jurors cannot be bribed, intimidated or otherwise tampered with. Factland provides incentives to ensure its fact-checking community is demographically representative of the United States as a whole and judgments are not biased to one group or another. You will have an opportunity to serve on a jury yourself and observe the process first hand. If you don't like a verdict, you can appeal it."

- **As a user, I shouldn't need to trust Factland.** Trust should be taken off the table: the whole point of making this a web3 app, yes?
- **We might explore the deeper ontological issues between trust (faith-based) and truth (empirically-based, at least in theory).** This is a rabbit hole of research and a keystone concept.
- **Why make the app adversarial by design?** How might the app provide rails to produce successful coordination by design?
- How are the sizes of the disagreements considered? **What constitutes "working" in relation to larger disagreements working "better?"**
- What is the "cutting edge anonymity tech?" As a web3 native this sounds extremely suss and jargon-y. **As a web3 user I want to know what the tech is.** This one sentence seems to address many of the points raised above, yet I do not understand *how* the tech 'ensure jurors cannot be bribed, intimidated or otherwise tampered with." Maybe these designs have already been considered, and if so should be rendered visual on the first screen before interacting with the app. 🔴 Probably the most important point for us to cover, as it relates to numerous other issues and is not immediately apparent.
- Regarding incentives: **What is the theory of providing incentives to motivate fact checking rather than (what seems quite obviously to be) the inverse motivation to manipulate the system to steal the incentives?** Related to points above and is a keystone concept.
- Why is it limited to "demographic representations" (goodbye to anonymity then) of the USA as a whole? **Internet communities are not explicitly correlated to a national jurisdiction, so why is Factland?** What does this imply about the kinds of topics that are privileged and how the jury process works? Why not a web3 native stance starting here? 🔴 Key inflection point: solar vs. lunar fork in the road.
- Seems like "not liking" a verdict should be designed out of the jury process, as this is decidedly against the validation of truth being separated from opinions, as stated in the first sentence. **How are opinions to be separated from the consensus mechanism in this build?** Or are they?

#### Does Factland pay?

> "Factland distributes cryptocurrency tokens for betting and payment for other activities, such as serving on a jury and sharing evidence that is cited by a juror in issuing a verdict. These tokens currently have no market value but may in the future. All successful bets pay a dividend based on the size of the prize and the amount staked. Losing bets forfeit the tokens staked."

- This reads to me as "Factland pays tokens for collusion and manipulation of the voting process." 🟢 Seems like an obvious point for introducing strong tokenized governance to prevent coordination failures. ie: **Let's consider governance/community design alongside the core product.**
- **"Bets" is a loaded term with strong implications that appear to be antagonistic to the core motivation of the app.** 🟡 This might be a UX copy issue, or revealing of the need to shift how we approach the underlying psychology of the UX.

## Regarding the core underlying assumption of designing/building the product towards adversarial dynamics: **Should truth-y-ness should be considered a zero sum game, rather than a non-zero sum game where everyone stands to benefit from a more transparently democratic exchange on the core issues, analysis of the evidence, and integration into our common concept clouds?**

### Sign Up Flow

- Email: considering the sensitive nature of the topics and process of this app, we might consider allowing users to register more anonymously. *See: lunarpunk or solarpunk tactics?*
- 🟢 There are numerous web3 native and hybrid onboarding solutions we might explore, depending on what value the team wants to optimize for.
- Updates/summaries: how is this information conveyed? Who is the user that would be interested in receiving it? What will they be reviewing and how is it different that returning to or actively tracking the app? **What is the goal of capturing user attention and retention with this strategy?**
- Internet Identity flow: Need to be analyzed more closely. As a web3 native, there's no way I will sync a passkey with FaceID, TouchID, or similar. I'm unwilling to complete this flow. There are other more secure options to consider for web3 native onboarding. I reverted back to password creation.
- I received 2000 $FACT. **Where are the tokens held? What wallet is used?** Lots of wallet options to explore with their own out of the box UX/UI. There's stength in familiarity here.

#### Topic Voting


- Voted 1 token on a topic, against a pool of 200 tokens from other users. **The UI progress bar is confusing.** not sure what it is telling me. Shows my vote, but not those of others? Why can't I interact with the decisions of other users? **What is the core information to convey at this moment?**

#### Submitting Evidence
- **Only options are `Supports` and `Refutes` but why not Undecided/Inconclusive?** Or other user submitted options, such as to invalidate the topic altogether?
- Form fields are basic: title and description. 🟢 There's an opportunity to introduce design constraint in ensure quality submissions: categories, tags, rational QAs, etc.
- Doesn't seem to be any constraints on link validation. **What prevents a user from uploading misinformation in these form fields and links?** 🔴 Smells like a phishing attack waiting to happen.

##### Report Now
- I submitted an obviously fake report and then voted on it with all of my tokens. What prevents other users from also up-voting this obviously false content, either because they believe it or because they want to manipulate and misdirect? **Core concept seems to be in how the internal/private jury mechanism works, so let's review that and optimize around it.**


#### Next Steps

- Explore the jury persona.
- Discuss how curation works: by a central authority or a DAO?
- Learn about the tokenomic utility strategy.
- Learn about the must haves in the underlying infra and how composable we can be with using other governance legos.

---

## Miro Board General Notes

### Personas

- **Anarchy persona will likely be one of the most common users in web3, either as a black hat or white hat hacker. If we do not consider this user the product is almost guaranteed to fail immediately do to contract and/or social manipulation.** Curious what the assumptions are that lead to considering this user "unimportant."
- `Beliefs` are currently about political orientation. **This should be general ontological/epistemological orientations if the goal is to approach objectivity (ie: truth).** Unless the goal is to visually render the demographics of (apparently US-centric) political beliefs?
- `Goals` are political goals. **This should be statements of in-app goals, motivations for using the product.**
- `Frustrations`: same point. **This should be UX/UI frustrations, not connected to politics.**
- +1 to understanding `Technical Literacy` of the personas
- `Motivations`: This section is redundant, as this is the purpose of the whole persona exercise.
- `How likely are they to use`: Goal is to ID user personas that we want to design towards, so we can assume that this column is 100% and if not then we should remove that persona.
- **Sticky Notes**:
- Good to consider the user emotions through various stages. **We could make a UJM (user journey map) to consider psychological stages of the product.** Should avoid designing for a specific emotion though.
- "Is how they are using Factland useful to us?" Who is the "us" in that statement and what constitutes a positive utility? **Does Factland contain ulterior motives for the use of the product?** Very important.

## In general, *where do the above mentioned assumptions come from? Who was interviewed to generate this content? Our first goal is to question and validate these assumptions to ensure we have identified an important problem and to use this as a reference in evaluating design decisions.

### Tokens

- **At a high level, this seems like the core value prop of the whole product, but seemingly under utilized.** This section deserves its own UX sprint: starting with assumptions, distilled into personas, visualizing the user journey, resulting in flows generated for each token. Then we can map these flows onto the user personas to understand how to refine their designs based in utility and avoiding assumptions.

### User Flows

- This is arguably the most important research to conduct in informing decisions in the build, but also apparently underdeveloped.
- By redesigning personas away from political orientation and towards a user's relationship with the app architecture we can build more meaningful stories, flows, and sitemap.
- Not sure how the screens were generated without flows for each user.
- "Curator" persona is illustrated, but does not seem to be present in the current build.