See also:
https://en.wikipedia.org/wiki/Gemini_(protocol)
It is similar to the old gopher: text files, links, and images form a hypertext optimized for reading. Text is formatted like Markdown - but even simpler.
Clients display text, like an eBook, or images / media.
Since it does not use “addictive by design” UI elements, like feeds, timelines, likes and upvotes, colorful and distracting elements, endless scrolling, as well as comments that invite trolling, it feels a lot calmer.
Servers can run on a PC or Raspberry Pi which needs half a Watt of power. No FAANG companies needed. No expert knowledge needed - not more difficult than running a file sharing client.
I think it is the right thing for defense of democracy and sharing your voice in the digital realm.
Edit: If you see comments here which kinda miss the point, appeal to emotions, have faulty logic, or depart from entirely incorrect assumptions: Please keep in mind that big US tech companies can’t say “that’s bad, how will we shovel money with this?”. Please use your critical thinking skills - they are much needed here!
The purpose of these limitations is to make it useless for advertising and user tracking - what the modern web does all the time. It is not a limitation aimed at limiting users but aimed at limiting corporate shit.
Same way as the lock at the door to your home has not the purpose to keep you outside, but the people outside which you do not want in your home.
if you fetch something from a server, you can be tracked. if you get presented content, you can be advertised to. it’s a meaningless limitation.
Well this is what tech companies would say ;-)
But unless you deliberately visit facebook.com or doubleclick.com, you won’t be tracked since gemini transmits a single document from a single site at a time, and it is purposefully made so that your friends and your tribe can run a server.
there is nothing stopping any server from tracking requests on the backend. there is also nothing that stops a server from transmitting that request info to a third party. in this way it’s identical to http. it’s also identical in that there is nothing that stops you from setting up a http server with no tracking.
And the thing is that the modern Internet is not built any more for this. It is a pure perversion of what Tim Berners-Lee wanted and created at CERN.
but gemini doesn’t solve that. the protocol is not the problem.
You keep repeating that.
But tracking and advertising have certain technical requirements to techniques which the gemini protocol does not offer. For example linking to Facebook like buttons or Google Analytics or client-side Javascript.
Gemini does not offer that.
And if you think different, show me a single gemini website with ads.
You can start here, in a HTTP gateway:
https://portal.mozz.us/gemini/mozz.us/
or
https://portal.mozz.us/gemini/taz.de/?
or
https://portal.mozz.us/gemini/geminiprotocol.net
or here
https://portal.mozz.us/gemini/tobykurien.com/microblog.gmi
it doesn’t solve it because the problem is not a technical one but a social one. a malicious actor can still track and advertise to people. that doesn’t mean it’s happening right now.
the onus is on the protocol to prove itself.
Of course, Gemini is also a social endeavour. Such projects are created, nourished and grown by communities.
But technical decisions and design choices also embody values and political goals. You do not build nuclear weapons to combat world hunger and things like Facebook and Grok do not exist to promote democracy.
Yeah but it is all the difference if the server is google.com , or your friend’s.
The whole concept is based on decentralisation.