As someone who likes Geminispace, the point of using it is that when I’m in a Gemini browser, I can’t get to the Web and the Web can’t get to me, and I enjoy my little hideout. The point of writing stuff there, for me, is to expand what’s there.
That said, I dual post a lot of stuff via smol.pub, so clearly I don’t need Gemini to be not-HTTPS. I just like it. Which isn’t sufficient to justify a whole protocol, probably.
i don’t think it’s supposed to be “better” than anything else or necessarily have a “point” ? my impression is that it’s more of a “why not?” kind of project, like a lot of things on the indie web
i don’t really think any project or hobby needs a justification beyond “i just like it.” liking something is a good reason to do it
It feels like the people working on/running Gemini infrastructure don’t want to actually solve the issues with the modern day web and instead just wanted to be different, for the sake of being different.
but i’m not sure why that has to be a criticism or a bad thing! lots of practices & sites & people on the indie web want to be different for the sake of being different, lots of it is “pointless” or doesn’t “solve the issues with the modern day web”. some things are just kind of neat or interesting or fun.
1.13 Do you really think you can replace the web with this?
Not for a minute! Nor does anybody involved with Gemini want to destroy Gopherspace. Gemini is not intended to replace either Gopher or the web, but to co-exist peacefully alongside them as one more option which people can freely choose to use if it suits them. In the same way that some Gopher users serve the same content via both Gopher and the web, people can “bihost” or “trihost” content on whichever combination of protocols they think offer the best match to their technical, philosophical and aesthetic requirements and those of their intended audience.
An even better answer to this question is “so what if we can’t?”. Gemini doesn’t need to be able to do every single thing that the web can do to be worthwhile. Geminauts can use the web to accomplish tasks which can only be accomplished on the web, and when they’re done they can close their browser down, let their computer catch its breath, then open their Gemini client and spend the rest of their online time enjoying the very different experience of reading interesting or useful or funny content in Geminispace. This is no more weird or futile than using a bike to run errands in your neighbourhood or for fun and exercise on weekends even though you still need to drive your car to work on weekdays, or growing some vegetables in your backyard or herbs on your windowsill even though you still need to buy most of your food at the supermarket. Even part time relief is still relief!
yeah i don’t see anything wrong with something existing just because it can in this context. that a group of people came together and created something that is in use by other folks that enjoy the atmosphere of it is enough to justify it i think. if it isn’t for you, you can safely ignore it
I agree. A bunch of people made a thing to see if they could. Other people use it because they can. That’s the origin story of a lot of Web sites too, tbh.
“Plain text” HTML pages perhaps aren’t as simple as they seem. A lot of the defaults are really bad. Web authors can easily make slow pages that break on mobile.
Now, 10 years later, maybe AMP wasn’t so bad. A subset of HTML that’s fast and has a built-in validator could probably have helped people.
A validator and a separate application sort of do the same “enforcing” job.
As I see it, the point of Gemini is that one can build a simple browser that is guaranteed to display the vast majority of Gemini sites correctly, something you can’t guarantee with the Web. Which makes it a lot less likely to be controlled by large companies because people other than large companies have the resources to make browsers that work. Realistically most people won’t build their own Gemini browser, which is totally OK, the point is that those who want to, can, and there are enough people interested to discourage a monopoly.
That being said, I respect Gemini’s vision but it’s not the right solution for me. I am currently planning out a protocol and format that’s like Gemini but graphical.
To me, the appeal of Gemini is the decentralised nature of it.
It doesn’t have a “town square” or a Facebook, everyone has different capsules (websites) they regularly visit, or use Orbits (Webrings) to just explore.
In other words, it feels just like how the web did before social media!
This sounds like a good idea. The main thing that bugs me about Gemini is how painfully bland of an experience it is to browse Gemini sites. HTML and related technologies are a total dead end for web freedom in my opinion, so I agree with the decision to leave all that behind with Gemini (except maybe the HTTP(S) protocol, I don’t know anything about that stuff), but I feel like they overcorrected a little. I know for a fact that it would be possible to design a layout format that would allow for a great deal of creativity and flexibility without having a novel-length specification. I’ve been wondering if most of the types of layouts that people do with HTML/CSS could be done with some kind of simple binary space partitioning system with cells that contain text and images, since that kind of what flexbox is like anyway. Images and background colors/textures alone could probably make up for most fancy CSS styling stuff in a way that is good enough (as long as there was alt text for accessibility).
The strength of the web platform is that decades-old pages can still work today if they conform to standards. The weakness is that it is too free because users clicking links don’t know what they’re going to get when they click.
A Gemini user knows what they’re going to get (plain pages) in the same way a user using the YouTube app knows what they’re going to get (videos).
The reason the rate of link-clicking has fallen so low is because it is so often a bad experience and users have been “trained” by random negative reinforcement not to click links.
I don’t think that would be as much of a problem without JavaScript, autoplaying audio/video, and certain CSS features like style changes on hover. Browsing any kind of audiovisual content on the Internet runs the risk of stumbling across something extremely unpleasant, but beyond that, I think it’s mostly the fault of the features I just mentioned. What I’m thinking of is something more like a book or a PDF document where text and images can be arranged in any way, but it would be impossible to make something like a pop-up or a screamer (although it’s been a good eight years since I’ve come across one of those naturally).
That being said, though, I also think it might be cool to see sites that are primarily focused on sharing videos, images, or some other kind of media move away from web browsers entirely. A site like YouTube doesn’t have any particular need to be chained to a web browser and the same functionality could be provided by a standalone, non-browser-based app. Social media sites and other forms of media sharing sites are moving towards being based around standardized protocols anyway, so it makes me wonder how much browsers are really necessary for the way people use the Internet nowadays. It makes me wonder if it would be possible to make a generic, standalone, non-browser-based client that could be used to browse any instance of a federated platform like Mastodon, PixelFed, PeerTube, etc.
I agree, but what bugs me the most, as an artist, is that even though images and videos are fine, they’re basically discouraged, and Gemini doesn’t really do anything to make it easier for browsers to get videos and images working without bringing in complex, difficult to maintain third-party libraries.
My planned solution to this is to make my document format into an extremely restricted VM. VMs on a webpage are often considered a security nightmare but that’s because they’re basically allowed to do anything and everything. Not this one, and in fact it’s designed with limited extensibility. It does just enough to decode and play media, provide some basic interactivity, and create page layouts. It always runs in a fixed memory block, so you know exactly how much memory a page needs to run, and references to memory are wrapped within its bounds. I have a bunch of other details but I’d rather not flood this thread with them.
The point of this is that not only can pages have videos but it also has everything it needs to decode those videos, and the browser implementer has to do zero extra work to get it running compared to getting page layouts working. Shipping a whole codec with the videos may theoretically result in larger files, but I definitely believe there are ways to minimize that extra increase in size through a variety of techniques.
I agree with a lot of the reasons brought up here, but I wanted to mention an additional one that appealed to me when I first discovered Gemini. Because it is limited to plain-text, it’s very difficult for advertisers to co-opt it. You can’t set up shopping carts or flashy banner ads, and you can’t redirect or load a ton of tracking scripts in the background. It’s very likely that Gemini will always remain a hobby project because it can’t be used by venture capitalists, and I think that’s a very good thing.
Frankly, this sounds like a great use case for anonymous FTP, BitTorrent, or some other p2p tech.
Of course, the problem with FTP is that the protocol was never designed with security in mind. BitTorrent requires trackers, and trackers are accessible via the web and can be knocked down by the authorities. Other peer-to-peer (p2p) tools generally don’t run on smartphones.
I hadn’t articulated it, but this is part of what I mean by “the Web can’t get to me” while I’m perusing Gemini. I will definitely not face banner ads or get tracked by a Meta Pixel everywhere I go, and I love that.