Hello community! Happy New Year to everyone.
I would like to share this interview that we published, both in English:
and in Spanish:
It was the result of an email interview with @GaelDuval
¡enjoy it!
interesting read; thank you Soldan and Gaël.
The article alludes to something that I’ve thought about a lot, though it doesn’t really seem to go into detail about it. That’s fine, because it was out-of-scope, but I think the tangent might be interesting to talk about here…
ROM modding is still more headache than it should be (Ronnie isn’t busy churning out ISO images of Windows or Linux on a per-laptop basis), but if the /e/OS Easy Installer doesn’t just-install properly, it’s possible to buy a phone with /e/OS out of the box…and while a handful of apps still linger a bit in terms of the cat-and-mouse compatibiity game, the overwhelming majority of Android apps just-work. I’m not saying it’s a “solved problem”, but I am saying that there is an onramp.
My actual-thought is about the broader ecosystem. I’m pretty guilty of it as much as everyone else..but if I get /e/OS and then spend 90% of my time in WhatsApp and Instagram or Gmail or Tiktok…I’m not sure how ‘unplugged’ I’ve gotten from The Algorithms and The Data Collection. The next step after replacing Google-ified Android is replacing the apps on top of it that do the harvesting…but while some can be relatively easy (Signal and Telegram aren’t exactly difficult to use or wanting for functionality over WhatsApp), things get dicey pretty quickly when we start talking about media content. If there’s such a thing as Spotify for Creative Commons licensed music…I haven’t come across it. I know a handful of content creators use Odysee or Rumble, but neither platform seems to be gaining traction in a meaningful sense over Youtube. Mastodon has seemed to get a bit of a following vs. Twitter/Bluesky/Threads, but its federated nature seems like it requires either one extreme (here’s a docker-compose.yml for you to self-host to convince your friends log into!) or the other (here’s a super popular server that’s just differently-centralized!). I’m not touching the third-rail of “Netflix Alternatives”, and in a depressing irony, I think that e-mail was about the closest thing to a distributed social network we saw, but it’s pretty much homogenized between Microsoft and Google and maybe Yahoo, with everyone else splitting up a single-digit percentage.
I think the problem we’re running into is that these sorts of services require either a whole lot of cooperation from everyone to do peering in a practical way, or it’s just centralization-with-extra-steps…but social media and video and music only really works if there’s a collection of content to enjoy…and I’m not sure how to meaningfully solve this problem.
Discuss =)
4 Likes
Hello @voyager529 thank you for your comment and happy new year 
I completely agree. That’s precisely why I mention in the title of the article that the problem is the smartphone ecosystem. All those apps are a crucial part of that ecosystem, and until something changes significantly in the way those “social media” apps work, there’s not much that can be done. Well, actually, a lot can be done, because just talking about these issues and trying to bring them to the non-technical public helps a lot. That’s what I enjoy most about communicating these issues, trying to bring them to places where no one thinks about them. In my personal case, I use eOS and I don’t have any proprietary social networks on my phone, only WhatsApp for obvious reasons…the so-called “necessary evil.” But for everything else, when I have to use those types of networks, I do it on my computer, where I have more control over what I do.
For my part, I’m waiting for some radical change that will bring down these applications. I think that at some point, data, like commodities, will cease to be so valuable, and all that digital extractivism won’t make much sense.
3 Likes
Happy New Year to you as well, Soldan! 10/10 for the discussion starter and the awareness being brought to the situation!
You said:
Well, actually, a lot can be done, because just talking about these issues and trying to bring them to the non-technical public helps a lot. That’s what I enjoy most about communicating these issues, trying to bring them to places where no one thinks about them.
And, I think that awareness is a good first step…but I do think that there are fundamentals that are very difficult to solve even in theory, regardless of how many people know about it.
I think the most distilled variant of the best possible method of social networking was - and still is - Usenet. Users can be on a server, who then in turn federates with another server, and posting on one newsgroup replicates to all of the other servers. Want to run your own NNTP server? you can! …don’t know why you’d WANT to, but you can peer with other Usenet servers if you want, and then write posts to newsgroups on your server, and it peers to everyone else, and vice versa. There’s still a level of control that exists by server admins, who can opt out of replicating specific groups if they want, and if they do, users can move to a different server who does replicate the blocked group without losing access to others. Users can move between newsgroups with their username and participate in discussions based around topics. All of that is brilliantly engineered.
…but there’s a reason why Reddit is functionally the Usenet of the modern web, ‘accessibility’ being one of them. Newsreaders are their own special breed of applications that, oddly, all still keep their Windows95 motif. I don’t think this is a bad thing personally, but the closest thing to a browser-based frontend was Google Groups, with all the headache involved with that. The next problem was spam. So. Much. Spam. We made killfiles to help mitigate it, and communities even sprang up, ironically, to share their killfiles…but while that mostly-worked in the 80’s and 90’s, the only reason why AI spambots don’t completely obliterate Usenet today is because there is no overlap between “people who use Usenet in 2025” and “people who fall for spam”. Something like Mimecast or Postini could help, but again, centralized authority…and one person’s ‘spam’ is another person’s ‘sincere, glowing review of a product’. Similarly, there’s no mechanism to upvote/downvote/prioritize posts over others, and it’d be impossible to bolt onto the existing standard for the same reason - it requires a centralized authority to ensure that a spambot isn’t upvoting its own spam by itself, instead of telling every peered server to register another upvote, then changing its username, and telling them to upvote again, ad infinitum.
The issues with trying to come up with a solution to the social network problem is that they basically all end up in one of a few fundamental camps:
- It’s a trustworthy centralized server/authentication service when they do it (Odysee/IRC/Pixelfed/BlueSky)
- It requires a bunch of technical people to set up a disparate set of servers (Mastodon/Rocketchat),
- It’s prone to spam/ID theft (Usenet)
- It’s got very little content to convince others to join (Diaspora/Retroshare)
- While the algorithms of Meta/Tiktok/Google are highly questionable, there’s either no meaningful means for one to curate what they see so there’s a needle-in-the-haystack problem, or it requires one to write their own Regex to do so (…basically all of the above)
If there is a practical solution to solving this paradigm, I wish to subscribe to your newsletter, because this is basically the point at which I feel defeated.
2 Likes
This topic was automatically closed after 15 days. New replies are no longer allowed.