Show HN: The Agora – an experimental social network
anagora.org
anagora.org
1. There exists a set of files, or nodes, contained in GIT repositories, that sum up to a personal wiki, of sorts. These files, in the case of http://anagora.org, are copied from https://github.com/flancian/agora
2. In those files are links in a specified format to other such wikis.
3. A server (Agora Server) renders a the files in a manner far more useful for interaction. http://anagora.org is running said server. The actual data being served is (for performance reasons, I assume) a copy of the reference wiki it can them push changes back up to.
4. The server also has a daemon that is polling other such sites (presumably the source, not the other servers) looking for more content to add to its repository.
Question: Do those other sources get added back to the primary repository?
5. If we want to add our own, we clone a starter kit, and host our own server. It could be local, or on the internet. The data needs to be public, the server itself doesn't. If this is true, this makes hosting things much easier, as there is no need to maintain a server on the internet, just have a valid GITHUB or other such account in good standing, and valid email for the signup process.
Do I have that right?
1. https://github.com/flancian/agora does not contain actual user content currently (see 4. for historical context), but rather serves as a "root node": it contains a base [[CONTRACT]] (for now just a document stating a code of conduct that all users must follow; might be expanded into something fancier using a <buzzword>blockchain</buzzword>) and pointers to repositories (currently one per user) that contain the actual content. You might laugh at how simple this currently is :) https://github.com/flancian/agora/blob/master/util/reset
2. Yes; the format is currently precisely Markdown plus [[wikilinks]] as an extension. [[wikilinks]] are "dead" by default in Markdown, being non standard; the [[agora server]] makes them go live by anchoring them to a context (an [[agora]] [[node]]).
3. Yes -- although [[agora server]] is currently unable to push changes upstream to user repositories, this is planned. For now it's just a renderer. Users use tools of their choice ([[roam]], [[foam]], [[obsidian]], etc.) to create and maintain their repositories. I use [[foam]] on VS Code.
4. Right now the Agora only pulls from user repositories as per 1.; eventually it will indeed pull from other Agoras, forming a federated network. Agoras will probably interact as long as their contracts are compatible. A: as per 1., currently no content is actually stored in the primary repository. Note that initially the repository did contain the pulled content, as it was included with 'git subtree'; but on second thought I thought it was simpler and cleaner to have the root repository only contain pointers.
5. Yes, although note that the data does not need to be public either; you can run a graph from any source as long as you make files in the right format available to [[agora server]].
https://anagora.org/node/CONTRACT
https://anagora.org/node/wikilinks
https://anagora.org/node/obsidian
https://anagora.org/node/agora-server
This is my "hobby project". I work at Google as an SRE.
One of the ideas here is to integrate the Agora with existing social networks:
https://twitter.com/flancian/status/1343979536434798598
The Agora is an overlay to the internet (semantic web) controlled by users. It is part of the open source ecosystem.
Google owns the copyright, as can be seen in the LICENSE file and headers, but the totality of the project is open source (Apache).
I own https://anagora.org in particular and plan to run it as a public service.
Giant Red Flag
I considered this for a while and I didn't find any downsides. It's open source; can be forked anytime by anyone.
I do this on my free time.
What, in your spare time? That seems... restrictive
How would they view working a second job? Does it matter in what field?
I understand there is often a reaction to seeing "copyright by Google"; but I researched it and it didn't seem to have any consequences of note.
The one thing I know of is that contributors to the project must sign a (one time only) CLA: https://cla.developers.google.com/about/google-individual
I hope this isn't an issue for you; if it is, please reach out -- or consider forking! :)
This isn't the same as the killedbygoogle complaint. I know they kill a lot because they try a lot but social is Google's Moby Dick. They'll jump on it and fuck it up if it takes off.
Fair disclosure I know it's distributed and as such I've got this link up my sleeve
https://www.eff.org/deeplinks/2013/05/google-abandons-open-s...
I'm interested in discussing how precisely Google could negatively affect the project -- I currently don't think that's possible, but I might be missing something.
I don't know how they'd mess it up, gut feeling really. I do honestly wish you the best of luck outside of my cynicism!
Google kills social projects they run, but since they don't run this one, I don't see them interfering. Their ownership is just a technicality because of stone age IP law.
https://anagora.org/node/open-source lists all relevant repositories.
http://anagora.org/node/agora-plan for a development journal+plan.
http://anagora.org/node/agora-protocol for interactions with the wider landscape.
Joining in with your writing if you already have it up somewhere and would like a new display platform helps a lot!
> Agora: (in ancient Greece) a public open space used for assemblies and markets."
The quick and fuzzy answer is: whatever the community wants it to be :)
My current view follows. In a nutshell:
* The Agora is meant to be a goal-oriented social network specializing in problem solving.
* The first goal of the Agora is to build itself.
* The second goal of the Agora is whatever you (the community) want it to be.
* The Agora seeks to complement and enrich the internet and existing social networks instead of "taking over".
From https://flancia.org/agora:
"You can think of the Agora as a convention based social network; an optional, user-controlled annotation layer that can be applied over any internet platform which supports user-generated content.
I think one of the best possible uses for such a network would be to use it to pro-socially maintain a distributed knowledge graph tailored specifically to the goal of solving problems: those of its users and society at large.
Its users, as a cooperative group, could by default take a naive but rational approach to problem solving:
For each problem in the set P of all problems:
Describe it as thoroughly as possible.
Maintain a set of known or argued possible solutions, S(P).
For each solution in S(P):
Describe it as thoroughly as possible.
Maintain a set of resources (people, time, attention, money) needed to implement it, R(S).
Users that agree on their defined subsets can then efficiently collaborate on solutions as they become available by pooling of resources.
We apply some good old recursivity and seed the Agora with the problem of how to build itself. That is, how to build a system that allows participating users and entities to collaborate optimally in the face of adversity (such as biases, irrationality and even actual ill intent)1.
The Agora should be built on a federated protocol to limit the harmfulness of diasporas. Groups might temporarily diverge in their views enough to want to run separate Agoras, but different Agoras should be able to cooperate on problems and solutions for which there is enough ideological alignment, and eventually merge."
So should this be understood as a kind of a multi-user Memex? Annotation and linking over any site that supports user-editable fields seems similar to that, if one squints their eyes.
Where there any other similar technologies that you considered instead of Underlay and what made you choose this one specifically?
I have to clarify at this point the Agora is not based on [[underlay]]; they are just a reference I go back to often in the context of [[distributed knowledge graphs]].
I know Samuel Klein though and I'm looking forward to collaborating more with him and the Underlay project; it hasn't happened yet mostly because the Agora is still at a very early stage in its development :)
https://anagora.org/node/underlay https://anagora.org/node/distributed-knowledge-graph
The current architecture is the simplest distributed architecture I could come up with to get started: the Agora pulls from a series of git repositories that its users maintain, these repositories containing Markdown notes with [[wikilinks]].
[[IPFS]], [[solid]] are all natural directions to evolve in. I just started with something simple enough to be coded in a weekend :) And also something that most people in tech are already using (git+markdown).
Doesn't it just get messy and overwhelming after a while? I visualize it as nodes and edges, but where the edges don't really mean anything specific other than "this makes me think of that".
I feel like a system that supported a lot of different types of edges that each had clear meaning would be more useful. "this requires that" and "this supports that" and "this needs to happen after that" or "this blocks that", etc.
Imagine I have a personal collection of recipes I like to cook, and the recipes link to foods with relation ingredient. When my friend comes over for dinner, I could ask them what foods in my kitchen they like, and instantly query all recipes containing those foods. Maybe I’ve added how long each recipe takes to cook, which appears as one of the many RDF triples I can use to filter and sort the matching recipes.
Knowledge graphs enable incredibly rich, structured queries. This is the promise and power of semantic web. Otherwise all you can do is rudimentary keyword searching.
The vision of the semantic web is to enable these rich queries over the entire collection of human knowledge, not just personal notes.
I see this exact problem with bookmarking tools. I have a large collection of bookmarks with many kinds of tags (topical, meta-tags like !important, !soon, !followup, !share, ...), etc. But the reality is those tags are not sufficient to search for what I need - adding a piece of content and searching for it are far removed in time (weeks, months, sometimes years), and "future me" doesn't remember the context something he put in with. The today-me can't anticipate which of the pieces of content will be useful and to what end.
So I don't know _what to search for_ or _how to search for it_ == but just as importantly, I don't know _what use to label something for_ and _what terms/structure_ to add to it.
I think the semantic web must have assumed we'll tag/label things correctly, and the reality is we don't do it because it's on average too hard.
RDF models facts/knowledge as subject-predicate-object statements called triples. Each part of the triple is called a resource and has a URL.
You wouldn’t bookmark pages, you would state facts about them in RDF using a vocabulary of resources.
A hypothetical semantic note taking app would let you insert many facts (RDF triples) about something you find relevant into your own knowledge graph, and the facts you insert would be constructed from your chosen vocabularies (https://www.w3.org/standards/semanticweb/ontology).
Roam/bookmarks/hypertext is the Stone Age compared to RDF knowledge graphs. https://en.m.wikipedia.org/wiki/Resource_Description_Framewo...
My current hypothesis is that [[wikilinks]] on top of the block concept popularized by tools like [[roam]] (lists in Markdown) are enough to implement a small lightweight DSL that maps to [[rdf]]. See [[agora actions]] for an early stage example representative of the direction I want to go in.
Ubuntu/Gnome's builtin search engine (Tracker) for god knows what reason uses RDF/SPARQL and anyone who wants can run "tracker sparql <query>" from the command line or their app/library/tool to query all kinds of things on their system.
Its been around for a long long time but count how many people/projects use it.
I have an idea similar to this (in many way Agoria has improved upon my idea, mostly in the fact it actually exists)
Something I spent time thinking about is adding metadata around relationships between nodes. In keeping with the KISS philosophy of markdown it's difficult to capture that without adding some messy syntax.
On the whole, I think OP is onto something. Though heavily geared towards those of the technical community.
I know what you mean w.r.t. Markdown syntax. One thing I'm trying to do with the Agora is to use a single extension ([[wikilinks]]) as much as I can; see for example [[agora actions]]. TLDR: I think wikilinks and blocks are enough to craft a simple DSL that lets you remove most of the limitations for defining/passing metadata.
See also: [[the agora is a markdown convention]].
https://anagora.org/node/open-source
https://anagora.org/node/wikilinks
So i give up, and currently i am embracing the tree structure to manage my networked notes. Now, i'm testing this new system. Maybe in the future i could find even better way to manage my notes.
https://anagora.org/ might clarify what it does: it surfaces content related to the node you're visiting using a lightweight fanout convention (lightweight in particular once you get the setup right or if you're already taking notes using something like roamresearch.com).
Essentially when you click on a wikilinks such as [[hn]] you see many pages about HN sequentially, one or more per contributor (user). It's a bit like distributed https://everything2.com.
> "It surfaces content related to the node you're visiting using a lightweight fanout convention"
Requires knowledge of what exactly a "node" constitutes in this context. Since I don't know what such a node is, the remaining part of the sentence about the fanout convention is just meaningless (it also seems irrelevant?).
Now, it is explained on Agora what a node is, namely:
> "A node is the set of all subnodes with a given title, or otherwise mapping to a single entity or wikilink. Subnodes can come from a variety of sources; currently these are mostly notes as volunteered by users via their independent digital gardens."
This suffers from the same problems: I still don't know what a node is, since I would need to know what a subnode is. I'd expect a recursive relation where a subnode is simply a node that is a child of a node, but it seems there is a distinction between both. In any case, it's still unclear.
It seems as if it's written by someone who's so familiar with these concepts that they forgot what it's like to not have this background knowledge. I'd recommend letting someone without prior knowledge explain to you what they think it is in simple terms, and build your explanation from that.
Yes, I agree; all text is a continuous work in progress. What you see in the Agora are my notes about the Agora as I develop it and think about it, essentially. Part of the reason I posted this Show HN is to get feedback from people new to the project yet presumably close to the current target audience, if there is any target audience at all :)
I did immediately get the "Everything2" vibe, which is cool, but the thing is -- I've never done anything other than a random walk through Everything2. I couldn't imagine trying a directed or semi-directed traversal of related topics in Agora.
What am I not grokking?
In a way I think of the Agora as a (hopefully) cheap open integration hub, targeting initially people that are maintaining blogs or notes repositories (a la [[roam]], [[foam]]) and would like to see them integrated within an extra ecosystem; but eventually moving on to the greater task of providing an alternative to some walled gardens.
https://anagora.org/node/activitypub
You could make the same comment about the web, RSS, WebMention, and ActivityPub (and everything else that's ever been).
Why is this always the #1 post on every Show HN post?
Oh, this is really just something else?
FaceBook was just MySpace, which was just AOL.
Google was just AltaVista.
Amazon was just Barnes & Noble with a crappy website.
Once upon a time, I spent a whole lot of time in physical libraries. To this day, I remain enamored with the 'stacks', those 'endless' rows of roughly categorized text that allowed for discovery of knowledge new to an individual in a manner that the post usenet internet seems to be unable to emulate.
Why, when we have the 'sum of all human knowledge in our pocket' does an interface similar to library stacks not exist? Search engines in their modern forms were probably unavoidable, just as librarians and card catalogues are necessary to help patrons quickly access specific information, but a library without public access to the stacks is hardly worth visiting. In fact, libraries were like this during a particularly dark period in western history and only though the work of a handful of forward thinking librarians did public access become the norm.
The agora 'nodes' listing is possibly the closest I have seen to a 'stacks' like structure for the modern internet. To me, regardless of implementation, this is a momentous achievement. Also, if anyone knows of a similar structure, please let me know.
The intent here is definitely to embrace [[RDF]] and any other applicable semantic web technologies and standards -- the [[Agora Server]] is only about one month old so really quite early stage, and I only started learning about RDF recently (see notes if interested); hopefully some degree of RDF support will happen this upcoming long weekend :)
I am in contact from the [[wiki]] and [[web annotations]] communities and also have plans to adopt standards from these groups.
https://anagora.org/node/rdf https://anagora.org/node/wiki https://anagora.org/node/web-annotations https://anagora.org/node/agora-server
You might get some good ideas from there, it was a sort of similar python wiki like project that were made for App Engine, it has backlinks, it used json schema extensively (maybe too extensively).
Will check it out this long weekend :)
The hypothesis is that "niche" might be good enough to bootstrap something more generally interesting.
EDIT: I think there's an Uncle Webring in there too, somewhere.
https://github.com/altilunium/Note : Forked from PmWiki https://github.com/altilunium/rtnF : Re-written from scratch
I'm still struggling to differentiate this from a wiki, since its main feature is still the [[WikiLink]]. The best thing i could think up for now is to modify the wiki's UI and UX for personal note taking usage. To create "networked-notetaking application". Pivoting my research goal from "organization-knowledge-management" to "personal-knowledge-management".
The idea of "knowledge-graph-based social networks" is cool, but dont you think that the "post + comment thread" pattern, the "interest group" pattern and the "one-to-one direct communication" pattern are irreplaceable in a social network ?
Even the wiki itself is using a rudimentary system to simulate those pattern on top of its knowledge graph structure. Lot of people, communicating together by editing the same single page, just like using a single blackboard together. (For example, see the "Talk" page on wikipedia)
https://www.agora.vote/
https://eprint.iacr.org/2020/865.pdf
https://www.idni.org/agoras/The Agora is currently costly to set up, but cheap to use after that.
https://anagora.org/node/agora-editor for pointers on what the environment I use looks like. If you use a code editor and/or git you're almost there :) If you don't, I'm happy to provide tailored support.
https://anagora.org/go/agora-howto for a step by step :)
And without the pyramid scheme[0] aspect of urbit.
Its architecture is very simple but it's already decentralized; users keep full control of their data. I also intend to evolve it in the direction of a fully distributed system.
See https://news.ycombinator.com/item?id=25581127 for more :)
I'd read the comment you linked and am aware of that.
My comment was addressed to the poster to whom I replied.
Agora seems interesting, although I think the requirement for external dependencies (a git repository) makes it less viable as a platform.
I get that it's necessary to have standardized formats for data -- especially in a decentralized environment.
I also get that markdown/git is a ready resource for that.
However, it seems to me that unless one is willing (and able) to set up their own git repository/server on their own (physical, VPS or cloud) hardware, you're getting decentralization of Agora on the back of centralized git platforms -- which kind of defeats the point of a decentralized platform.
Unless your only target set of users are developers and other technical people with the resources to host their own git instances, forcing the use of git repositories will limit both interest and adoption.
Removing such a requirement (or baking git into the platform) would go a long way toward making Agora more attractive, IMHO.
I am most certainly not trying to discourage anyone from using Agora, nor do I think it's a bad idea.
In fact, I think it's quite interesting. Thank you for working on this and I wish you much success with it!
Thank you for your comment! I agree with, well, all of it -- as it stands now the Agora is decentralized only for users that can control their own infrastructure; and is indeed usable at all only by users with a certain degree of technical know how.
I intend to do what I can to fix this and make the Agora [[maximally inclusive]] -- it is not arcane as a design principle, but rather due to limitations of the current implementation. Put another way, I consider the current implementation just an MVP that might allow us to test some interesting hypotheses and perhaps bootstrap a more inclusive distributed architecture.
Do they? Why?
What physical law requires this?
Having said that, logseq is very promising as it's based on precisely the same architecture that the Agora uses now: Markdown in git repos. I would love it if [[sign up]] could be just "click here to start a digital garden in logseq and have it show up in the Agora".
Unfortunately the format that logseq uses to encode blocks is not compatible out of the box, as they use Markdown headings instead of nested lists -- but I've spoken to them and IIRC they have an issue open to support nested lists as a block encoding convention, which I personally believe makes sense (there is a tradeoff here, though).
You could turn a bad comment into a good comment by helpfully explaining which parts you found confusing. Better yet would be to make some suggestions for becoming less confusing.
Additionally I set up [[goals]] to try to answer this for future readers.
https://anagora.org/node/agora https://anagora.org/node/goals