#dev 2021-08-08
2021-08-08 UTC
Seirdy and alex11 joined the channel
# 
[jeremycherfas] I knew you’ld know.
hendursa1 joined the channel
# 
capjamesg[d] !tell sknebel That was it!
# capjamesg[d] Thank you!
# capjamesg[d] !tell vikanezrimaya The indexing journey continues. I am trying my crawler on a site other than my own. It's going well so far but suffice to say you have to do more work when you are not crawling a site you built and know inside out 🙂
# 
vikanezrimaya capjamesg[d]: Yeah, one's own mf2 markup might be too pure for the crawler tests. Good for ensuring the basic functionality works, but the edge cases can only be tested on different websites
# vikanezrimaya Loqi, I know
# vikanezrimaya Oh, the joys of using matrix as a bouncer and having a persistent backlog and push notifications
jacky joined the channel
# capjamesg[d] Indeed.
# capjamesg[d] How's your crawler coming along?
# vikanezrimaya Im currently resting
# vikanezrimaya Gathering strength to rewrite my whole database connector because a certain library has a fatal bug in it that makes my website into a pumpkin after a certain number of requests
# vikanezrimaya I am 100% sure now this is not my code or anything that could be caused by my coxe
neceve joined the channel
# capjamesg[d] n=2 but my search engine has just successfully crawled another site.
nertzy_ joined the channel
# capjamesg[d] n=4 and things are going better.
# capjamesg[d] The search engine can now fill in for missing meta descriptions.
# capjamesg[d] This is fun 🙂
jeremycherfas, [Rose]1, nertzy__, shoesNsocks1, bneil, eco, unrelentingtech, angelo, neceve, jamietanna1 and jamietanna joined the channel
# 
jamietanna1 jacky by "an authorization attempt fails " do you mean a hit to `/authorize`?
# jamietanna1 I can't remember off the top of my head, but I'd expect that to redirect back to a caller with an error URI, that the caller should then display something readable for?
tetov-irc, rockorager, Kai`, Zegnat[d], chenghiz_, Sam2 and [Sam_Butler] joined the channel
# [Sam_Butler] I've started using LinkedIn as a way to communicate above the walled garden chat communities.
# [Sam_Butler] Hey all, wanted to share something I've been thinking about — which comes from similar motivations as Indieweb, and indeed has some parallels to POSSE and web identity. Would love to hear your thoughts.
# [Sam_Butler] For this use case, and many others, it seems uncontroversial that we'll move away from centralized social networks — to networks that are more free, user-centric, and dynamic.
# [Sam_Butler] I'm doing this to enable discoverability + accessibility + knowledge-sharing with my work. Much of the professional world — at least in subject areas I'm focused on — is on LinkedIn. By posting content and having discussions there, it enables anybody who may be interested in what I'm working on it to discover it — and likewise, me to discover them.
# [Sam_Butler] To start this transition, it seems that the simplest and surest first step is having access to our *graph data* — so that we can *use it* and *build/experiment on it*.
# [Sam_Butler] Specifically, I'm talking about:
# [Sam_Butler] 1. Export my contacts from LinkedIn, Facebook, and other social networks, using their APIs
# [Sam_Butler] possible as well. )
# [Sam_Butler] 2. Import those contacts into a new substrate/protocol, with some simple structure to help with identification. (For example, to tell the difference between two contacts named "Amy Bee." Or, if each of us have a "Amy Bee" in our contacts, an intelligent way to recognize if they are or are not the same person. Email or phone number or personal domain could be a useful "anchor" for this — and better approaches could certain
# [Sam_Butler] 3. Once I have all of my contacts accessible — I just want some simple functionality to *send messages*, and the ability to *tinker/experiment with new ways of interacting* on top of this new substrate.
# [Sam_Butler] Part 1 is feasible today. Part 2 seems like it requires some dedicated thought and design. Part 3 only comes after Part 1, and along with Part 2.
# [Sam_Butler] If anybody is interested in this type of thing, I'd love to hear thoughts and collaborate. Note that this approach isn't about onboarding to a new network, it's about migrating our existing networks to a new and dynamic environment. Also this is envisioned more of a commons / protocol initiative, not as a new network itself.
# [KevinMarks]1 is the assumption a shared graph, or each holding our own (in effect what Signal does, where the contacts are on your own phone)
jacky joined the channel
# [Sam_Butler] The assumption is each person holding our own graphs (when I export my contacts from LinkedIn or Facebook, I only have access to my first-degree connections anyway)
# [Sam_Butler] Perhaps individuals' own graphs could be stitched together in some privacy-respecting, capacity-augmenting way — if there was consent for it
# [Sam_Butler] A relevant tool/reference for this could be: https://idx.xyz/
# [Sam_Butler] From a person in their community:
# [Sam_Butler] > You could issue a verifiable claim to prove that a DID owns a Web2 social account (Twitter, Github, etc.) Once issued you could store the claims in an user's IDX to link it to their DID which helps to build reputation over time/confidence that "Amy Bee" is your Amy Bee bc you can see it linked to her multiple social accounts that you recognize.
# [Sam_Butler] In this direction, the protocol is a decentralized public identity table. A person's unique identifier is a DID, linked to that person's individual IDX, which is comprised of that person's public accounts across different platforms (e.g. Twitter, LinkedIn, Facebook, Instagram, email or phone?)
# [Sam_Butler] So instead of a source of truth based on a single address (e.g. email, phone, domain), the protocol's source of truth is a DID linked to a multitude of addresses.
# [Sam_Butler] That seems intuitive and directionally correct in some ways. I just wonder about IDX's business interests in this (whether they are a good building block for public commons infrastructure), and whether a different kind of architecture could be sensible
hendursaga, mgdm and KartikPrabhu joined the channel
# 
@GirelliGabriele Webmentions are my way of finally letting Disqus go!
I implemented a small Javascript-based project to easily show… https://twitter.com/i/web/status/1424461659624648724 (twitter.com/_/status/1424461659624648724)
KartikPrabhu, hendursa1, hendursaga, tetov-irc and Rattroupe joined the channel
[Emma_Humphries] joined the channel