cool, a garden index! love that they included an RSS feed too (https://blogscroll.com/index.xml) - i'm sticking that in my feed reader immediately. it sounds nice to have a few random new digital garden sites to peruse once in awhile.
this story touched me, honestly. humans are so funny. sometimes we just get used to the way things are, and we carry it forward, and that’s that. no reasoning. it’s bad, i know. but that’s life
great post from the libera staff, respecting the matrix folks while dealing with all of that must have felt disheartening.
matrix has other driving forces and incentives. there’s only so much time they can spend on things like individual bridges.
meanwhile, they devote developers to writing two different homeserver implementations in parallel. or writing an experimental p2p homeserver - or the three guys working on thirdroom. ugh
i just hope they realize what’s important. people just want to chat on a distributed platform that isn’t irc - make it as simple and fun as possible. that is your entire mission.
no metaverse, no experimental backend shit, no securitygasm cryptography - nobody is buying drugs on matrix. this isn’t signal or whatsapp, this is discord for tech dorks.
just focus on making GROUP chatting good, simple, and fun. the ux just utterly sucks right now.
it’s the difference between scrolling through a menu on an ipod versus a self checkout kiosk - the kiosk just FEELS bad, simple human revulsion. the element interface offers the same experience.
I wouldn't go so far as to say the ux utterly sucks but it's certainly not something I've been able to recommend to most family and friends yet (I can't be too hard on matrix since I have to use Teams sometimes which is simply worse in pretty much every category I can think of, which is sad since the microsoft product teams replaced, skype for business, was itself light years behind teams).
It's a bit unfortunate to see matrix still not there yet, but on the other hand I'm still hopeful for the future. I think ElementX is the result of realizing that an excellent chat experience is what is needed. I also think the matrix foundation governing board elections [0] show that the community agrees that a rock solid chat experience is what is needed. At least for the individual members category, none of the candidates who were elected had a platform statement advocating for more work on the experimental matrix features. Sumner (who I think had the highest support by a good margin, although I can't find the exact numbers right now) specifically stated that experimental matrix features should not receive further investment until the baseline features are acceptable.
I realize that getting funding is a beast of its own and that is probably why some of the experimental things took focus at times so I can't fault that, but I really do want matrix to be something I can eventually recommend to all my family and friends instead of whatsapp or discord. I think it is slowly heading that way..
I installed a Matrix homeserver (private, separate and non-federating) for our family at the start of the covid lockdowns, and we've been happily using it ever since.
"Recommending" anything is pretty much worthless. Stand them in front of a done thing that is already set up, that's worth a hundred times more.
Meamwhile the Wikimedia foundation are still stuck on a fragmented mishmash of IRC for the old guard, and Discord or Telegram for the younger crowd who quite literally would rather not engage then forgo GIFs.
You'd think they could dig into their hundreds of millions of donated dollars they don't even need and help Matrix and themselves but apparently not.
In their defence WMF have been an Element customer for years, for internal usage. But rolling out Matrix across a large community requires significant community management effort (e.g. what Mozilla did with https://discourse.mozilla.org/t/synchronous-messaging-at-moz...), and that's probably the gap you're alluding to here.
To be fair they never did make any effort to provide community chat spaces. Maybe they think it will legitimise not using wiki talk pages to post memes and go off topic, but 20 years of evidence says people will find their own places to hang out. Time was, everyone kind of just happily used Frenode IRC by default (after all, what else was there? MSN?). Now it's all sorts of apps all over the place, mostly set by whoever got fed up with whichever IRC channel first starting their own channel with blackjack and hookers...uh...image attachments and emojis...in their current favourite app. The Freenode/Libera debacle dented what was left, and now it's a pretty sad closed, Balkanized state of things.
I still think they'd get more value out of slinging you guys a small-to-them wodge of cash to run a community homeserver than most other things they spend money on.
I did the same except federated and it's pretty easy if you have some experience with Ansible and Docker. This repo has a great Ansible playbook that will set up the whole thing for you on whatever VM or VPS you point it at:
I mean, I was still learning how Matrix works, so it took some trial and error - especially voice/video calls - so a few days of reading docs and doing.
Later, though, I stood up my own Matrix server, this time federated, and that only took maybe an hour or two. :)
I did not use the community Ansible playbook that someone else in this thread suggested, because I enjoyed instead figuring out how the individual pieces work together, but I tried it much later, and it seems to work very nicely, so I too can recommend it.
A single guy produced in a single year something more private and with a UI that doesn't suck: SimpleX.
This Matrix case in my opinion is either lack of talent or intentional. One should never discard the second option when considering how governments tend to interfer in such tools and Matrix is quite sus on the funding side.
In 5 years we've seen little to no remarkable progress. I'm not recommending this platform any further to anyone, other options have delivered what this one had been promissing for years. Time to move on.
Absolutely intentional. Various members of the Element community had been fighting tooth and nail for better emote support, and many implementations have been developed and tested, and even for the ones that followed the Matrix-backed MSCs and passed the necessary tests every single one was outright denied and more recently told that Element makes major money off of government interests so why listen to the desires of the people, even if they do all the work necessary for free anyway?
we’ve had to prioritise keeping the lights on by building stuff we can sell rather than merging custom emoji PRs. it’s a “please put on your own mask before helping others” kind-of thing.
There is zero logical reason you cannot do both, especially as a lab (which are all minimal value adds and have been broken for years). This has been open for 8 years. And there have been many people willing to implement it for free, who have gone out of their way to work around the painful busywork involved, where they would selflessly rework it as needed to fit within the MSC once the SCT fully approved one.
Are you telling me that with the multi-millions of dollars you get from law enforcement, you cannot delegate a manager to, as a minor part time duty, oversee the free work given to you? Especially when the various Element apps have been in an effective maintenance mode for at minimum the last year? This sounds like either Element is run very poorly and despite the massive funds rolling in can barely maintain themselves, or there is a distinct distain for external work that does not generate immediate revenue for the company.
Trust me, I understand full well what you’re saying. You’re maximizing profits with no care for the “freeloaders”. That’s fine, you can do that, but don’t act like I have to care.
> Are you telling me that with the multi-millions of dollars you get from law enforcement, you cannot delegate a manager to, as a minor part time duty, oversee the free work given to you?
Correct. We are not profitable, and frantically trying to get sustainable by focusing on things which generate money so we can keep paying our salaries, rather than taking on maintenance of new “free” contributed features when we’re still focused on the core. The reason we don’t have more $ is because we gave away everything as Apache licensed FOSS, and turned out most people were happy to use that gratis rather than paying for anything. The most likely manager to oversee custom emoji was laid off last year due to lack of $.
Fwiw, we make almost no money from law enforcement.
> there is a distinct distain for external work that does not generate immediate revenue for the company.
No. We really appreciate the various custom emoji PRs, and I personally feel awful they are stuck. At this point the most likely way to get them merged is if i do so myself.
> You’re maximizing profits with no care for the “freeloaders”.
No. We are not profitable. We are trying to stop losing money, so we might actually be sustainable.
If you don't mind my asking, what are the features that you're working on to become profitable?
I know custom emojis sound dumb, but they are basically the differentiation factor for how Discord gets their billions. People almost literally pay hundreds of dollars for Nitro just to get their emojis/reactions. Other than bigger uploads, that's all the lower Nitro tier does.
Some people just laugh in my digital face when I suggest Matrix, because they will never, ever, countenance losing their emojis. I don't know why they care that much, but they absolutely do.
It feels a bit like trying to sell a car without Bluetooth. You might think it's a trivial frill and a distraction from the best engine ever, but the interest will be lukewarm at the very best. I bought a car without Bluetooth but then I fitted my own adapter. Even that isn't possible here.
So we do care about custom emoji, and they are obviously critical for folks looking for an encrypted Discord alternative.
However, right now we're not focusing on building an encrypted Discord alternative but instead a self-hosted WhatsApp (or Teams) replacement for governments: https://element.io/sectors, on the basis that they are the ones paying. Sure, we could have gone after Discord ~5 years ago and launched our own Nitro equivalent, and perhaps we should have. But instead we observed that governments REALLY want to run their own encrypted interoperable comms systems, and we thought that having Matrix used as the backbone for public sector communication would be a good way to prove and fund it sustainably. Ideally we could then use the public sector as a launchpad into other areas - a bit like how email & the web and even the internet spread from DARPA / NSFNet / academic etc to the rest of the world. After all, what better endorsement for Matrix than someone like NATO using it for comms?
As a result, custom emoji are stuck in the middle of the todo list still.
Some of the other features that we've worked on instead have been:
* Make encryption secure - i.e. migrating from the old C/C++ libolm implementation to the rust vodozemac implementation.
* Instant login/sync/launch - i.e. an entirely new sync mechanism.
* Rather than developing 4 different client stacks (js-sdk, ios-sdk, android-sdk and rust-sdk), converge on a single one: matrix-rust-sdk, and make it excellent.
* Rework the core UX to make encryption invisible (rather than full of confusing unactionable warnings and verification nags etc).
* Migrating to OpenID Connect for auth, so getting 2FA/MFA etc
* Public sector enterprise features: antivirus, regulatory compliance, secure border gateways, cross-domain gateways, active directory sync, SCIM sync, kubernetes operators, etc. etc.
Now, the plan is to use the govtech business to get back to funding mainstream Matrix uptake. But first we need to be able to fund ourselves to work on it.
I see your point, and it sounds like an invidious and existential position to be in and you have my sympathy to be fighting uphill against legacy like that.
Honestly though, I would contest the logic here:
> bit like how email & the web and even the internet spread from DARPA / NSFNet / academic etc to the rest of the world.
It spread because it was useful and revolutionary, not because DARPA specifically used it. Yes there were network effects, but the network effect overlap today between "NATO" and "a friend group hangout" is pretty minimal.
> what better endorsement for Matrix than someone like NATO using it for comms?
Honestly, if all I knew was NATO funded and used it, I'd assume it was enterprise-grade cost-plus shitware 15 years behind the curve, and with zero organisational interest, if not outright hostility, to it being used by anyone without a governmental chequebook and service contract in hand, and steer well clear.
It's OK to write software for NATO and governments for the foreseeable future, but the website still says "communicate with friends, family, communities and co-workers", so you can see why people might think it's still targeted at them and then wonder why it doesn't even provide the baseline of human-interest features people have been getting from other platforms for about a decade.
Not that it helps pays the bills to court Brian Everyman and his Warhammer group chat, I do realise that.
yeah. i'm not saying we've made the optimal choice here: for instance, we could have focused on growing mainstream usage at all cost without distractions of enterprise or govtech (similar to Bluesky), and assume that once we have enough users we'd be able to figure out monetisation somehow via a Discord Nitro style scheme.
However, we're at where we are now, and we're committed as to the current path as a way to get sustainable and then go back to supporting mainstream users (including Brian and his Warhammer clan). Perhaps a profitable govtech business can subsidise mainstream FOSS Matrix without having to do Nitro style value gates (which are tough to enforce anyway in a decentralised world).
> but the website still says "communicate with friends, family, communities and co-workers"
That was the messaging prior to early 2023, when we shifted gear to "get sustainable by focusing on govtech". Which is why the frontpage of the website currently doesn't say anything about friends/family/communities, sadly.
I just had to fish around to find out where you saw the old wording - i'm guessing it's in the footer for element.io/blog? ...which is a snafu due to the blog being Ghost and the website being Webflow.
All the best of luck. Genuinely, I hope it works out because I despise all the closed chat platforms and view the lack of a compelling FOSS alternative to be strange, frustrating and even embarrassing when someone uses it as an example of why FOSS just doesn't work for consumer software.
As for the wording, it's on the front page of https://matrix.org, just below the fold. Scroll down once and it's centre of the page.
Oh! Well, yes, Matrix itself would love to be a cosy safe supercharged place for your community (even if i’m not sure a protocol should describe itself like that tbh). Some clients like Cinny or FluffyChat are close to that. Element will also get there eventually.
In terms of why consumer FOSS software often sucks:
* Good UX requires top down product discipline: cathedral model, not bazaar.
* Product folks are less likely to contribute to FOSS, especially in their spare time
From the conversation it seems Matrix just grew bloated, when I look at what NOSTR is developing and the speed at which crutial features are delivered then it is quite a world of difference.
Those governments might be paying your bills for now, but you've lost the support from the open source community on the way to get there and soon be forgotten by everyone. I just hope one day we won't have the sad news of discovering that Matrix injected backdoors from the Israeli government to receive funding.
Thanks for the tipoff to look at 0xchat - it does remind me of where Matrix was about 8 years ago (in terms of maturity, speed of growth, features, etc). Good luck to them.
In terms of Matrix having lost the support from the open source community because Element hasn't implemented custom emoji yet... time will tell. In the end I suspect the open source community would be even more upset if Element had run out of money and exploded (although you never know; haters gonna hate :)
yup. the problem on the matrix.org website is a tradeoff between a very dry “this is a protocol”, like w3c.org - versus trying to give some more accessible use cases for people who expect to find an instant messaging network. the current wording oversteers.
As someone who's (sadly) moving away from matrix due to the state of clients and running out of patience for Element X (it's been... What? Two years?) I have to say SimpleX and/or XMPP are where I am looking.
SimpleX is OK as direct replacement to Telegram, very similar except you are not able to write the comfy bots (yet) over there. On the plus side, no need to sign up with a telephone number either.
If you are willing to try out new things other than XMPP, I'd recommend looking at the NOSTR protocol and more specifically at the 0xchat client. They provide private message with E2EE but the biggest advantage IMHO is that you get connected to a huge variety of other services and people using the same protocol.
I find that better than building the connections on what are essentially closed gardens or, at best, isolated islands. Over there exist hundreds of servers (relays) run by volunteers where new features are proposed by users and actually implemented (success or failure depends truly on themselves).
Other good features being delivered on NOSTR are for example the anonymous IPFS (blossom files) where you can host complete static sites across the relays. There is truly plenty to discover there and to build, feels like the 90s and has been fun.
I just tried 0xchat. It's full of useless crypto shit, missing translation errors, their website is down and their "Oxchat Open Group" is basically dead.
In spite of my earlier comment, I do think it is worthwhile to stick around a little longer to see how it goes. As Arathorn mentioned, the latest ElementX release just came out and they are saying it is production ready. I got the update yesterday (I was using it in beta before) and I did notice some of my rooms syncing faster in the ui. Once SchildiNext rebases I am hoping it will be something I could finally recommend to people, at least if voice/video aren't needed. If voice/video is a requirement for you, then I could totally understand going to discord, although I am hoping that with the native matrix calls it could change that once the ux is equivalent.
if it would only work. but it doesn't. occasionally messages can not be decrypted. it happens randomly, and there is no clue why. had to give up using matrix with one friend because it just happened to often. the key management is also way to complex. there should be one key that protects everything. that can be a key that unlocks other keys, but all this complexity should be invisible to the user.
Messages that can't be decrypted have been a big problem however from personal experience and matrix themselves saying it should be fixed with recent Element X version this should be a thing from the past.
People don't want "a distributed platform that isn't irc", but a platform they like. IRC fails primarily due to requiring a persistent TCP connection, and secondarily due to not supporting image sharing, and tertiarily, due to not supporting emoji reactions.
Requiring a persistent TCP connection is simply a complete dealbreaker on the mobile phone platforms which make up, like, 90% of computers in existence. Fixing that would go a very long way, but it requires a lot more server-side storage since messages must be stored on the server until a client reconnects.
Data privacy requirements go along with server-side storage. If messages are stored for a few minutes that's probably okay, but that's too short to implement the feature. If they're stored for weeks, and especially if images are stored, then for legal reasons there must be a way to moderate them. On another front, at least the server only needs to store one copy of the message for as long as the slowest client might need it; it might as well just store the entire channel history for a fixed time. These considerations lead to a design like Slack/Discord/Mattermost/etc, and trying to make it decentralized leads to a design like Matrix.
You may think of IRC bouncers which don't have these issues, but notice the responsibility structure of an IRC bouncer is completely different. With a bouncer, the responsibility for storing the message lies with the reader, not with the network, so completely different considerations apply.
Sharing images and videos is just useful sometimes, even though 90% of the time it only serves to increase spam. And I don't get why users are extremely horny for emoji reactions, but they are. A platform can exist just fine without either of them - every social platform develops a unique personality in response to the constraints and nudges the platform imposes on users - HN doesn't have images, videos or emojis and that's a good thing for HN. It's the persistent connection thing that's the real killer, because it means the platform is simply unusable 90% of the time.
A version of IRC tailored to slower connections might look identical to (the non-binary segment of) Usenet. However, even Usenet has light moderation, that differs per server, to avoid being clogged with spam and binaries.
sharing images and other media is vital in most groups where i participate. we don't just socialize. it's not like HN. we actually work on stuff. and very often images are needed to convey certain information.
though even social channels work better with images. my family channel would be very limited without images. most of what we talk about is things we see, and can share images about.
emoji reactions are another matter. but given that it is possible to just use unicode emojis, i think that reactions reduce clutter.
Social media communities adapt to the constraints of their platform or don't use it. You seem unaffected by the inability to share photos on HN, indicating that's okay for sites that want to be patronized the way HN is. It wouldn't work if you wanted to create a meme channel, but HN doesn't want to be a meme channel and neither does IRC.
it's fitness for purpose. HN works perfectly for what it is used for. it would be unusable for a family or work channel unless that work is programming or something else text heavy which doesn't require sharing graphics. same goes for IRC.
the point is that usecases where media support is needed do exist.
It's still has a bit of a work-in-progress feeling, but for desktop usage Cinny has proven to be quite a good Discord-like interface for Matrix.
I'm not using it right now because I switched my server to authenticated media and Cinny doesn't seem to have that implemented (yet), but that's mostly because I'm ahead of the curve on that change.
Where does someone begin with Matrix? The homepage mentions servers but how do you choose one? Is it like the ferdiverse with its hundreds of mastodon servers?
Yes, it's like the fediverse. The "main" server by the developers is matrix.org (similar to mastodon.social), but there are others too.
You can also host your own server from different providers as described in this post: https://matrix.org/ecosystem/hosting/
It was literally in the top five search results for “programming” in the “public channels” page. With this little care or moderation I will stick to Discord.
The web app has been evolving too, but less dramatically - we need to enable instant sync there.
In terms of custom emoji and selfserve moderation: both are the direct casualties of having to focus exclusively on servicing the government customers who actually pay for Element development. Since the end of 2022 we have had to completely change focus to first generating $ in order to be sustainable, and that meant parking everything but Element X, Web, Call and Synapse. The Libera mess is also effectively a symptom of that.
P2P, Dendrite, Thirdroom etc have all been shelved for over a year. (Weird to see so many year-stale comments on a year-stale blog post).
well, hopefully it is just a slump, and as you may have seen we’ve been doing releases on a best effort basis. it isn’t the first time dendrite has had a hiatus. we did try to spell out the situation here: https://matrix.org/blog/2023/12/25/the-matrix-holiday-update...
Most status page products integrate to monitoring tools like Datadog[1], large teams like github would have it automated.
You ideally do not want to be making a decision on whether to update a status page or not during the first few minutes of an incident, bean counters inevitably tend to get involved to delay/not declare downtime if there is a manual process.
It is more likely the threshold is kept a bit higher than couple of minutes to reducing false positives rates, not because of manual updates.
Nah, _most_ status pages are hand updated to avoid false positives, and to avoid alerting customers when they otherwise would not have noticed. Very, very few organizations go out of their way to _tell_ customers they failed to meet their SLA proactively. GitHub's SLA remedy clause even stipulates that the customer is responsible for tracking availability, which GitHub will then work to confirm.
reply