5 ms·
XMPP is supported by a large number of clients, but running a server and getting everyone on clients with comparable featuresets is a nightmare. It’s a cluster
by zenexer 6y ago
XMPP is supported by a large number of clients, but running a server and getting everyone on clients with comparable featuresets is a nightmare. It’s a cluster of disparate standards, and it’s overwhelming. I’m sure it’s doable if you have the time to invest, but it’s not straightforward if you’ve never done it before.
Matrix is pretty straightforward on the server side of things, but the client UX is invariably mediocre. Vector—the official client—exemplifies everything that is wrong with Electron apps. Slow, clunky, poor UI, poor platform integration. With the default home server, it can take seconds for a message to go through. At least it’s far more customizable than Slack; it has an option for everything, which, as a power user, I quite like.
I haven’t tried Mattermost, but it looks like some of the important features aren’t FOSS, at which point it’s just another Slack as far as I’m concerned. I’ll gladly pay for support, but for SSO? Meh, might as well stick with Slack; at least everyone and their dog knows how to use it. (This is, of course, an opinion that stems partially from ignorance; I haven’t actually tried Mattermost, and if I do, I might fall in love with it. But my time is limited, and I can only evaluate so many products in a day.)
Not that Slack is much better here: their threading system has so many UI/UX issues. Ever had a thread with hundreds of messages? For your own sanity, I hope you haven’t. Ever tried to send an image to a thread from iOS? It’s possible, but only by pasting the image into the text field; the normal attachment button isn’t available, and Share buttons in other apps can’t send to threads. And, of course, the recent uptime issues.
- Arathorn 6y agoElement (formerly Riot/Vector), has improved loads over the years, and the default matrix.org average send time is around 100ms these days rather than multiple seconds: https://matrix.org/blog/2020/11/03/how-we-fixed-synapses-scalability/ https://matrix.org/blog/2020/11/03/how-we-fixed-synapses-sca... has details. I suspect you (and the parent) may be running off stale data. That said, Element could certainly use less RAM, irrespective of Electron - and http://hydrogen.element.io http://hydrogen.element.io is our project to experiment with minimum-footprint Matrix clients (it uses ~100x less RAM than Element).
- zenexer 6y agoIt has, and I’ve been using it since its early days. I still use it. It’s still terrible, just slightly less terrible. And, no, messages don’t consistently send in 100ms on the default home server; there are regularly disruptions that cause significant delays, sometimes as much as 10-20sec. That’s a big problem for a federated chat platform. Edit 1: I want to love it; the design is everything I could ever hope for in a chat platform. I even tried to contribute to Vector, but it was such a mess that I eventually gave up. Edit 2: > That said, Element could certainly use less RAM, irrespective of Electron - and http://hydrogen.element.io http://hydrogen.element.io is our project to experiment with minimum-footprint Matrix clients (it uses ~100x less RAM than Element). I'm not sure why this is a priority. Techies complain about RAM usage a lot, but if we have to choose between performance+power and a small memory footprint, we're going to choose the former almost every time. Take Telegram, for example: they have a bunch of native clients that perform amazingly well, although they do gobble RAM. Most of my technical friends use it as their primary social platform. It's not without issues, but it's really hard to go from something like Telegram Desktop or the Swift-based macOS Telegram client to Vector. And those clients aren't made by large teams--most (all?) first-party Telegram clients are each maintained by a single developer, if I'm not mistaken.
- feanaro 6y agoIt's weird that you're calling it Vector when it's now called Element and it was called Riot for years before that.
- zenexer 6y agoThe constant rebranding and confusion over Matrix/Vector/Riot/Element is another point of pain for me. It’s incredibly difficult to communicate unambiguously about Matrix with people who haven’t been following it for years. Does Element refer to the ecosystem as a whole, including EMS? The primary client? The core federation? It’s not obvious from a casual visit to element.io. I suppose if I said “Element web app,” that would be fairly clear, but I’m still in the habit of saying “Vector” from the days of Riot.
- upofadown 6y ago>XMPP is supported by a large number of clients, but running a server and getting everyone on clients with comparable featuresets is a nightmare. XMPP is, well, extensible. If things don't match in the clients then that particular feature just doesn't work. These days all the clients pretty much try to match the feature set of Conversations. That applies to the servers as well. There is a server tester for that: * https://compliance.conversations.im/ https://compliance.conversations.im/ So the tooling is pretty good these days. The great thing about XMPP is that basic messaging always works. That stuff is just too simple not to.
- BlueTemplar 6y agoI recently finally found someone to try Element with and the experience has been great so far. (Except for the need to go through Google's captcha at some point.) It even sent a 20 Mo MP4 like a champ, while Conversations sometimes chokes on not that high resolution photographs...