ambitiousslab

I'd like to second Snikket - it's designed for this use case and is very simple to set up.

If you'd rather not use Snikket, check out these recommendations for clients and servers.

Hope it works for you! Feel free to reach out if you have any questions.

Now: terrorists are terrorists, right wing rioters are terrorists, climate protestors are terrorists and misogynistic people are terrorists.

Soon: asylum seekers are terrorists, people who go on strike are terrorists, members of the opposition party are terrorists.

I support reducing violence against women, but prevent is the wrong tool for this problem. If the government actually want to address this instead of just looking like they are, I feel they should take an approach that actually works. We need:

• More consistent and holistic sex education, from a younger age
• Explicitly teaching about sexual violence, the services available and the punishments for doing it
• Investing in local policing, so that there is bandwidth to look into these cases
• Giving more funding to charities who support domestic abuse survivors
• Training for police, so they actually listen to women when they raise concerns at an earlier stage, instead of waiting until it's too late
• Tougher sentencing for any form of sexual violence

Prevent is both ineffective and discriminatory. It increases government surveillance, and raises the burden on GPs and teachers. The National Union of Teachers want to get rid of it, the Communities and Local Government Committee found a multitude of problems that haven't been fixed, and human rights orgs like Liberty and Amnesty International want to get rid of it too. It doesn't work and in many cases has made things worse.

Thank you for taking the time to write this, LibRedirect is so much better!

Unfortunately, I think it just picks randomly. I have had times where it has redirected me to an instance that is down. That said, if you have an instance you know is stable, it does give you a drop-down to always redirect to a specific one.

For me, it's many of the ones people have already said, plus:

• StreetPass (seriously cool - collects the mastodon profile of any website you visit where someone has set up the special link to their profile)
• Video Speed Controller (gives you fine-grained control over video speed, e.g. watching video at 2.6x speed)
• Privacy redirect (automatically redirects to various services, e.g. from Twitter to Nitter - can select a random instance each time)

I'm very sympathetic to this blog post, as it nicely describes why I use XMPP.

But, on a related note, I have noticed an interesting pattern where people talk past each other a little, especially when conflating user freedom and security.

If I'm to generalise, I feel the outlook of XMPP users tends to be more systemic and long-term. We've seen how chat networks come and go, we've seen the dangers of companies promising to serve your interests whilst also being a chokepoint of centralisation. So we tend to de-emphasize papercuts or current issues in clients and the protocol, on the basis that we have the power to fix them if we want to.

I feel that's shown in this blog post - all the points come back to the benefits of user freedom: no one entity controls you, the protocol serves you, you can choose your own clients, and if you don't like it, you can always switch / write your own!

What I've seen is that the people who gravitate towards Signal tend to be more concerned with the here and now - e.g. "how do I get my friend off telegram onto a secure / private service". I feel in many cases that making arguments about federation and the structure of the network won't sway them, as they'll always be able to point to some area where the clients are deficient in the here and now (depending on their interests - papercuts in the clients, different versions of OMEMO being used across the network etc).

I don't really have a solution to this, but I think all we can do is continue to make the clients and servers as good as they can possibly be. I always encourage anyone I manage to migrate to XMPP to send me any annoyances they find in the apps, so that they can eventually be fixed. We need to be ready for when Telegram, Signal, WhatsApp etc. abuse their power, because (as we've seen from the fediverse) that's the only time that "regular people" will care for the arguments that we're making about federation and user freedom.

Agreed, it's licensed under the MPL, a "weak copyleft" license. Each file that is MPL must remain MPL, but other files in the same project can be permissive or even proprietary.

While I definitely think it's better than a fully permissive license, it seems more permissive than the LGPL, which is the main license of WebKit and Blink. So I don't feel it's strong enough to stop it being co-opted.

I really wish there was a GPL-licensed rendering engine and browser, accepting community funding, with some momentum behind it.

I feel Ladybird have correctly identified the problem - that all major browsers and engines (including Firefox) get their primary source of funding from Google, and thus ads. And the donations and attention they've received show that there is real demand for an alternative.

But I think the permissive license they have chosen means history will repeat itself. KHTML being licensed under the LGPL made it easy for Google to co-opt, since it was so much easier to incorporate into a proprietary (or more permissively licensed) codebase.

There is Netsurf, but the rendering engine understandably and unfortunately lags behind the major ones. I just wish it was possible to gather support and momentum behind it to the same extent that Ladybird has achieved.

Different strokes for different folks! I've been fortunate enough that many of my family and friends have been happy enough to follow me.

But I don't disagree with you, Signal has a much more recognisable brand and better user experience. These are things that we need to improve if we're going to get anywhere near the level of adoption Signal has.

How I Got a Truly Anonymous XMPP Account:

• Open my client (e.g. Conversations, Monal, Dino)
• Pick a random server, username and password
• Click register

Sorry, it's a cheap joke, but it still baffles me that Signal requires a phone number, so I felt I had to post it :)

Of course, this is not XMPP-specific either, just my protocol of choice, there are many other open alternatives that also offer such functionality.

Agreed completely!

I think there are two main schools of thought - one is to buy linux-first mobile devices, and the other is to make linux work on android devices people already have.

At the moment, linux on android devices is in a better state. From what I read, calls, battery, suspend etc. generally works well on the OnePlus 6/6T and Pocophone F1.

For the linux-first approach, I'm very fortunate to have both a PinePhone Pro and Librem 5. They are both improving, and I've been daily driving one or the other for a couple of years, but they do still involve some sacrifice. I'm not sure they're ready for "regular people" yet, but they keep getting better.

Personally, apart from buying the devices themselves, I prefer to contribute directly to the projects focusing on the software. I feel you can't go wrong supporting PostmarketOS and Mobian. They are doing heroic jobs and have come a huge way in a few short years. Because they work upstream-first, any benefits one achieves are shared with everyone else in the ecosystem. And any benefits apply just as well to the linux-first phones as to the ones that originally ran android.

It's a long mountain to climb, but we've seen with Linux on the desktop that, over a period of years, things do get better. You used to have to pick your hardware carefully to get Linux installed. Now, you can install it pretty much anywhere, and most likely everything will work.

I feel eventually, the hardest problem in getting people to switch will be that proprietary apps won't work well. I think all we can do is to improve our free software alternatives to the existing ones, especially those with network effects like chat apps, as much as we can.

Every small step we can take is an important one. It's easier to get people to switch to Linux if they already use free software, as free software often does a better job targeting Linux than proprietary software does. Meanwhile, it's easier to get people to switch to free software if they run Linux, as that's the path of least resistance on that OS.

My fear is that they will start off by applying this to Messenger, WhatsApp etc. Then, in a few years, when criminals and tech-savvy people move to XMPP etc, they will say "the laws aren't working, we need to apply it at the OS level instead", and since iOS and Android have a big market share it's very easy for them to do it. At that point, trying to communicate with friends becomes very hard. It's one thing to get them to switch apps, but asking them to switch phone or OS is a whole other hurdle.

I'm trying to contribute both code and money to make XMPP, and mobile linux as good as possible before that can happen. I feel we need to buy time, by delaying and delaying chat control as much as possible, to make the free software, federated systems better and appealing to regular people. And then we can use that technology to buy time to push for political changes. I feel the only long-term solution here will be a political, rather than technical one.

It all feels like an impossible task, but I feel all we can do is try as hard as we can to make the world more like the one we want.