Arathorn

34 ms·

571.
A roadmap and appeal for help from The Matrix.org Foundation(matrix.org)
61 points by Arathorn 3y ago|29 comments
572.
by Arathorn 3y ago
no idea what's going on here; the post hasn't been changed since it was published: https://github.com/matrix-org/matrix.org/commits/main/conten...
573.
IndexedDB is 10x slower for storing arrays than strings(artificialworlds.net)
1 points by Arathorn 3y ago|0 comments
574.
Keep your Indexed DB keys and values small if you want good performance(artificialworlds.net)
1 points by Arathorn 3y ago|1 comments
575.
by Arathorn 3y ago
Coincidentally I set this up last night under UTM ( https://docs.getutm.app/advanced/rosetta/ ) and it worked astonishingly well. Outside of Rosetta, the native virt seemed a bit slower than whatever UTM does normal
576.
by Arathorn 3y ago
In Matrix conversations are decentralised by being replicated across the participating servers. It’s not just federated. Matrix isn’t p2p by default though - although there is a project to address that: https://arewep2pyet.com
577.
by Arathorn 3y ago
and ungodly numbers of ewoks, whether you want them or not. (although that would be The Matrix Holiday Special, whereas on more sober years it becomes The Matrix Holiday Update...) See also https://news.ycombinator.com/item?
578.
by Arathorn 3y ago
> Do you really want me to try to explain this, or is this another idle snarky comment? Yes, and thanks for doing so at https://news.ycombinator.com/item?id=38803004 . I was hoping for a comparison with Third Room particu
579.
by Arathorn 3y ago
He’s talking about https://nebuchadnezzar-megolm.github.io/ - and the fact that group membership in Matrix is currently determined by the server rather than a client. So if Alice is in a DM with Bob, Bob’s server could make
580.
by Arathorn 3y ago
None of Matrix’s typical usability problems are the protocol’s fault - they are the clients’ fault. The fact is that on the Matrix side we invested much more time on the protocol, spec process and even server implementations in the early
581.
by Arathorn 3y ago
The whole cyclical thing is interesting - I wonder if the previous generation app actually regresses while the next gen is coming together, or whether it's just perceptual that the next-gen is so much better than the previous one, the
582.
by Arathorn 3y ago
Well, the blog post is about Matrix, not Element, hence not dwelling on layoffs at Element - which are more to do with the challenges of running an commercial open source business in an economic downturn (ie persuading commercial users to p
583.
by Arathorn 3y ago
Yup, it has not been a great year at all. However, the reason I feel optimistic for Matrix is that a) focusing on fixing the fundamentals is demonstrably working (Element X is a genuine pleasure to use) b) we have dedicated fundraising infr
584.
by Arathorn 3y ago
The biggest sticking points were: a) metaverse hype collapse causing blowback oversteer against any kind of virtual world platform b) the team wasn't interested in running it standalone as a for-profit startup outside of the Matrix Fou
585.
by Arathorn 3y ago
It would be good to hear what the gap actually is, if it’s so obvious.
586.
by Arathorn 3y ago
I can speak to Third Room, although haven’t dug enough into the current incarnation of Croquet to do a side by side comparison. I’ve called out the bits which I’d guess are more unique about TR. TR is/was: * Open decentralised platform
587.
by Arathorn 3y ago
:) yup, it’s overdue, and excited to see the outcome!
588.
Matrix Holiday Update 2023(www2.matrix.org)
11 points by Arathorn 3y ago|2 comments
589.
by Arathorn 3y ago
A single user instance of Synapse in a handful of rooms should be 150MB of RAM for Synapse (and some overhead for Postgres). But if that user joined every public room on Matrix, the server would find itself dealing with a firehose of 100s o
590.
The Matrix.org Governing Board Elections(matrix.org)
6 points by Arathorn 3y ago|0 comments
591.
by Arathorn 3y ago
It’s not a server controlled secret; it’s a MAC of a shared secret negotiated via ECDH between the two clients. Diffie Hellman ftw. See https://www.uhoreg.ca/blog/20190514-1146
592.
The Synapse Matrix server is now AGPLv3(element.io)
3 points by Arathorn 3y ago|1 comments
593.
by Arathorn 3y ago
If you're talking about Element switching its contributions to Synapse from Apache to AGPL, the rationale is that folks who want to commercialise Synapse will either have to a) opensource their changes, b) pay for an AGPL exception fro
594.
by Arathorn 3y ago
No, we didn’t - as the Libera mess demonstrates. We are stuck in a nightmare situation where Matrix is successful, but most commercial projects building on it don’t contribute back either financially or otherwise. We’re now trying to fix th
595.
by Arathorn 3y ago
The linked issue and the associated PR that fixed it haven't had a reply since 2021 because they were filed and fixed in 2021. Separately, the original issue regressed in 2023 on the Libera instance, which was tracked as a security iss
596.
by Arathorn 3y ago
I think your vitriol is better directed at me (as Matrix project lead & founder) rather than Josh, whose role as MD for the Foundation is to ensure it gathers $ and distributes it as effectively as possible to support Matrix. The origin
597.
by Arathorn 3y ago
> with the state of synapse, matrix.org, and the project in general, I wouldn't consider trying that for another 12-18 months. Or more. Synapse is very stable these days, and relatively resource efficient. Matrix.org (both the serv
598.
by Arathorn 3y ago
The issue in question here is https://github.com/matrix-org/matrix-appservice-irc/issues/1... (and https://github.com/matrix-org/matrix-appservice-irc/pull/133... ) which accide
599.
by Arathorn 3y ago
> Hmm, this is where I disagree -- if you've been scrambling to help, wouldn't the issue have been fixed by now? We scrambled to investigate and fix this particular instance (and previous unrelated instances where they'd h
600.
by Arathorn 3y ago
> At least another 2/5 foundation board members have strong personal ties with New Vector Nope. Neither Ross, Jon or Jutta have 'strong personal ties' with NV. We had met Ross twice(?) before asking him to join (based on

More ›