[selfhoster11]

The whole reason why the Web became what it is, is because HTML tags and JavaScript allowed for unbounded feature creep. Gemini purposely specifies a simple protocol that puts boundaries around what's allowed.

You would need a critical mass to get an extended Gemini off the ground now. Between the die-hardness of the early adopter community that realises that this extensibility is why they needed Gemini in the first place, the support for HTML and Markdown mimetypes over the core protocol (to alleviate some pain points) and lack of commercial attention, I have high hopes for Gemini.

> it will figure it out a way to sustain itself economically

Why should it? It's not a platform for commerce. Hosting a Gemini server on a residential connection has a negligible cost (a Pi and a few kilowatt-hours per year). Some small web communities host it for free. For personal expression, Gemini should be very cheap.

[oscargrouch]

When the web was there, it had no historical record to rely on, everything that happened was unprecedent in history because it was the "killer app" of the internet, and by virtue of being the killer app it made the internet something people want.

Now in 2021 we have a lot of historical records of how things shaped and get in the way they are now. Everything that you are saying about the Gemini is exactly how the primary web was in the beginning and yet it changed because a lot of forces happened.

I completely agree that the Web now is a whale and its too fat (BTW this is the point that Gemini is getting right), but i don't agree with the approach at all, as its not possible to have a time machine and make the technological status-quo get back to were it was in the nineties so a project like Gemini can be a successful branch of WWW.

A "skinny web" standard would be much more useful, and here i think that Tim Berners Lee is missing the point by creating something like Solid instead.

Anyway, its a chance missed to point to something for the future, pointing to the past instead and making a lot of good folks lost by following the wrong way.

I consider Gemini part of the "retro technology movement", that while cool and while having its heart in the right place, miss a nice chance to help us all change a very dark future ahead of us of digital feudalism imposed by the big tech goliath monopoly.

We, the tech crowd need to step back and stop following them into that bleak future, Gemini is subversive in that sense, but while the goals are great, its alienatory nature negating the status-quo and forgetting to learn with historical events its not a step into the right direction.

But if at least take some people away from the WWDC's and Google IO's, at least it will be a little bit less people into this army that is leading us into this dystopic future we are heading right now.. and i just hoped that Gemini with a little more contextual knowledge could be at least a stone in their path.

[selfhoster11]

A couple of things to unpack here. The main argument that I would like to make is that hindsight of where we'd end up if we pick the HTTP way matters a great deal for Gemini's adoption.

Back when Gopher and HTTP were slinging it out, the future was unwritten. No matter which nascent vision of the Web we picked, it would be cool. 20+ years later, we ran the capability-maximalist (excuse the term) experiment to its logical conclusion, we know for a fact that a web in the HTTP/HTML spirit leads to a hellscape inhospitable to the privacy conscious.

In that sense, Gemini is not a nostalgic look. It's a recognition that we picked the wrong evolutionary branch and that we must step back and redo the evolution.

I don't think it can win against the Goliaths of Big Tech, but then I think that no-one will. It's a political problem with no technical solution, because ultimately it involves money. The only way to kill those Goliaths is to get the governments off their seats and get them to sling some stones.

Ultimately, I think that Gemini recognises that slaying Goliaths is not something it can accomplish. It refuses to play that game entirely, and instead focuses on carving out little chunks in the hellscape for those that seek them. It's not forgetting to learn from historical mistakes, it's simply abandoning the pretense that the HTTP web can be unseated.