This "both sides of an argument" thing is poisoning logic. Somethings dont have 2 sides. The earth is round, for instance.
People like to hide behind "science says this" because there is usually objective science available, but science provides facts about cause and effect, not priorities, which is what leads to debate, and often blatant distortion of facts. I think a bigger challenge than presenting both sides of an argument is actually getting to the core of what is being debated.
I get frustrated by the “both sides” presented by mainstream media like the NYT, CNN, etc. it’s always democrats on one side, republicans on the other, as if there’s not an entire political spectrum that expands beyond those two.
The effect is it limits the “allowed” perspectives to within those two bounds. This is often exploited by those in power to control the bounds of possibility (that’s the manufacturing consent part).
Been thinking it would be nice to have comments / replies showing the top 3 conservative takes, the top 3 liberal takes, the top 3 science, 3 libertarian? top 3 fiscally conservative/socially liberal etc.
Ability to vote up and notate inaccuracies and removing personal attacks / insults.. finding a way to stop trolls from another side upvote frivolous things.. what the icons would look like.. all sorts of things I've been thinking about this.
Seeing the two sides thing really makes me feel the 'powers that be' really want people fighting on two sides.
I'm actually making tool for this exact same thing and the issue is most noticeable in kialo (which I argue is the same thing, with more mature UI). If you made all the research, conversations and planning to know, that answering this 'A or B' problem solves it, then Kialo (or that app) are perfectly fine tools. But most of the time people don't.
https://www.loomio.org is another attempt at solving the same problem. It does provide some improvements, but also fails in interesting ways (mostly, I think, through not considering temporality very seriously).
Ok I see, and yes I agree.
Although to me, I don't think this is necessarily as big of a problem as people seem unable to stop making it out to be, which I tried to point out in my comment:
https://news.ycombinator.com/item?id=25904058
As I see it, the fundamental problem is the infinitely dimensional (recursively self-referencing, paradoxical, etc) nature of reality. A "right way" to model it is impossible - the best we can do is try to find a "good enough" (across all variables) approach, acknowledge (and gather opinions on) where we have fallen short and could improve, and iterate as you go. Kind of like life itself, "coincidentally".
In the super short run, even adding a "middle" column would be a huge improvement.
> I'm actually making tool for this exact same thing and the issue is most noticeable in kialo (which I argue is the same thing, with more mature UI). If you made all the research, conversations and planning to know, that answering this 'A or B' problem solves it, then Kialo (or that app) are perfectly fine tools. But most of the time people don't.
> https://www.loomio.org is another attempt at solving the same problem. It does provide some improvements, but also fails in interesting ways (mostly, I think, through not considering temporality very seriously).
I think this is the most interesting (and largely unrecognized) niche out there - in no small part due to the obvious inherent complexity and no obvious business model, but I feel like those may not even be the main reason - I truly that this space is "hard to see" for most people.
As for design of your product, what do you think of these pie-in-the-sky ideas (from the link above):
> I think that approaches like this website are a not yet a perfect solution, but it feels like a step in the right direction. An improved implementation would have some clever means of allowing a crowd-sourced decomposition of the issue into it's constituent entities and behaviors, graphically displayed in some magical way to maximize the likelihood of everyone seeing the big picture free from misunderstanding, which "all" parties could agree upon (no agreement means more refinement is needed) and refer to in their arguments (so we're all talking about the same thing), and arguments/thinking could take place with precision (attached to the the specific entities and behaviors where the true disagreement resides).
> A lot of people may have a strong aversion to this approach as it would turn easy argumentation into a painstakingly laborious & pedantic process, which is a natural consequence of using slow but accurate cognitive processing, rather than fast but unreliable heuristic processing (the root problem in the first place). Arguments under such a methodology would also more closely resemble the scientific method, or systems analysis in software (where one does not simply throw together an analysis and insist that it is 100% free of material flaws).
So if you take Trophi as a working prototype, could we add a second phase onto an argument, where "someone smart and unbiased" goes through all the arguments, and decomposes them into some "proper" representation, essentially comprehensively aggregating all the distinct points made in the arguments into a master summary, ideally with some sort of a "properly representational structure" between all the points. Perhaps then, in a phase 3, there could be voting &/or further discussion on this summary, the results of which would be passed back through phase 2 to expand and refine the model, also noting where genuine disagreement remains (specifically where in the complex, multidimensional model), and this process is repeated indefinitely - and when some new event occurs in reality, then that could be run through this very same process, updating the multiple related models based on the results of the process.
Obviously this is an incredibly ambitious idea, and is fraught with difficulties (gosh, why is reality so darn complicated!!), but even a half-assed version of it might provide some value - and, perhaps along the way, the true nature of reality might slowly start to dawn on some people, which plausibly might improve the performance of the participants in the overall process, recursively over time.
After some time that's how I explain what I'm building:
https://news.ycombinator.com/item?id=25788274
if there's some way to contact you, I'd be super happy to send you an introductory video and some alpha version.
For the original problem, I think the flatter the ontology, the better. I find short videos to be great explanation tools for example - 3Blue1Brown is absolutely insane. But it can be just linked in text after someone asks specific question. I think what is being attempted here is kind of doing philosophy and philosophy is done by giving responses (often: 'you are having wrong argument' responses).
I've seen approach of making argument follow some traditional form (like enforcing using predefined logic to structure points), but I would like to have it as an emergent application. Extension from some subcommunity. It's just a very costly solution - it works perfectly fine when you need to explain management methodology in corporate setting, for example. Lots of resources, expensive mistakes. And I've seen some attempts at providing just that, for example (warning, in Polish) https://mysljaknaukowiec.pl/dyskusje
In a way, that is a methodology debate. Making a tool for checking 'what is' is one approach, making smaller tools that let people fail/iterate upon ideas faster is another.
Within this process, did you have a step that decomposed each comment to extract the constituent entities and behaviors, and then append them to an overall model (which subsequent comments would refer to)? This is not an easy task at all, and expecting average users to be able to do it is likely expecting too much. However, if this is what's needed to return sanity to the world &/or vastly improve our collective understanding of reality, then so be it.
> After some time that's how I explain what I'm building
That's a good idea. I was thinking about something similar to that, except rather than picking items from a personal list, users could pull an idea(s) from a common "tree of knowledge and ideas" or something like that, and then frame their comment around or in the context of those (often complex) ideas*.
> For the original problem, I think the flatter the ontology, the better. I find short videos to be great explanation tools for example - 3Blue1Brown is absolutely insane.
Wouldn't it be nice to have a new, rich multimedia-based language for communicating complex ideas in forums, rather than being stuck with this crap, "low signal/high noise and ambiguity".
I added contact info to my profile, I would love to see what you are working on - I think this space is well overdue for some innovation, and maybe I have some ideas that might fir into what you're working on.
One of my former best friends went down the deep end. It started with 9/11 conspiracies and just spiralled down from there.
I engaged with him, because I care deeply about him, but it just got worse and worse. Initially it was 9/11 and questions about wealth and power distribution and there are some valid questions that can be asked (although all the 9/11 stuff is bonkers). It soon became crazier and crazier, it was about "energy from the vacuum" (essentially a code-word for perpetual motion), how Gates is trying to sterilize the population through vaccination.
The irony was that he was seeing himself as a freedom fighter who is speaking the truth to prevent the next fascist regime, completely ignoring that many of the people he followed or supported were literal Nazis.
Engaging with him was not possible. He would bombard me with links that somehow were supposed to proof his point. If I brought up counterarguments, he would say that he has strong evidence that one can not believe mainstream media, science, logic ... And that I'm just a sheep blindly following "propaganda" and he knows better.
There was absolutely no way of getting through to him or engaging in any kind of discussion. This was just his monologues and flooding me links (many I suspect he did not even read/watch).
In the end I had to cut contact, because it was affecting me mentally, the interactions would just leave me exhausted angry and incredibly sad.
I don't have a solution, we need to somehow find a way of how people can be drawn into these crazy web of conspiracies and I don't think that creating forums were these crazy things are debated like they are valid opinions is an answer.
Which arguments are "allowed" to have a side? Given the recent disregard for the concept of freedom of speech, this site seems quaint.
These websites always seem to be like an episode of Big Talk.[1] "Alright, let's get this sorted! Does God exist, yes or no?"
That's not rational debate, "marketplace of ideas", or whatever you want to call it. It's screeching until everyone else realises there's nothing to gain from being there, and leaves.
The other side doesn't just go away. It's kept away by constant refutation.
Not exactly round, but it's an ellipsoid with an uneven surface.
- Isaac Asimov
If the Earth was shrunk down to bowling ball size and taken to a lane it would be just as spherical as any of the other balls.
Just because a human can put together a thought, doesn't make it good.
I think that approaches like this website are a not yet a perfect solution [1], but it feels like a step in the right direction. An improved implementation would have some clever means of allowing a crowd-sourced decomposition of the issue into it's constituent entities and behaviors, graphically displayed in some magical way to maximize the likelihood of everyone seeing the big picture free from misunderstanding, which "all" parties could agree upon (no agreement means more refinement is needed) and refer to in their arguments (so we're all talking about the same thing), and arguments/thinking could take place with precision (attached to the the specific entities and behaviors where the true disagreement resides).
A lot of people may have a strong aversion to this approach as it would turn easy argumentation into a painstakingly laborious & pedantic process, which is a natural consequence of using slow but accurate cognitive processing, rather than fast but unreliable heuristic processing (the root problem in the first place). Arguments under such a methodology would also more closely resemble the scientific method, or systems analysis in software (where one does not simply throw together an analysis and insist that it is 100% free of material flaws).
See also:
https://news.ycombinator.com/item?id=25903747 The First result for me (https://www.trophi.com/page/600eb2036c0b5800157eb041) is proof why this idea of "all opinions are equal and should be given equal opportunity to be heard" is bunk.
https://news.ycombinator.com/item?id=25903962 I don't see the point of this platform being to coddle inferior points of view by elevating them to faux equal footing.
This logically flawed "both sides" heuristic has become extremely broadly distributed through the system. It very commonly makes an explicit appearance in culture war arguments, and it is surely invisibly participating in cognitive processing even where it does not make a visible appearance.
I wonder how many of these bug-ridden heuristics are out there in the wild (there are several others present in this thread already, with only 36 comments), messing up execution for reasons that are not apparent from the UI perspective? If you think of reality as a system (which it technically is) that you had been hired to debug and fix, how might one go about locating all/most instances of such bugs, and then how would you fix them, considering we do not have direct access to the source code, but rather can only "write" to it via the api's that have been made available to us?
[1] Someone points out that a sub-component of a more elaborate platform already exists:
https://news.ycombinator.com/item?id=25904525
https://en.wikipedia.org/wiki/Argument_map
I expect most of the necessary components already exist as features embedded in other platforms, if they could all be mashed up together into a new "Collective Cognition" platform of some sort, I think we'd have something really useful on our hands.
The biggest missing piece that I can see is a means (a user interface for) decomposing complex problem spaces, and then (more difficultly), a way to visually(?) represent this high dimensional space that is understandable to a very broad cross-section of people (in the shorter run, even something that only "smarter" people could understand would be a good start).
Does anyone know of any existing systems out there for the "visual" representation of highly dimensional spaces, in a manner that makes it easy/easier to understand for teams? UML, sequence diagrams, etc are the only approaches I have any awareness of.