The world can be a pretty boring place, and there is rarely any financial incentive to making it more badass. Why don't we have some sort of crazy 3D matrix internet like in Neuromancer yet? The tech to do this certainly exists, but megacorps can't really profit off it. It doesn't actually make anything more efficient or productive.
The game Cyberpunk 2020 explained how this cyberspace could be created through something called Ihara-Grubb transformation: https://www.oocities.org/Teleran/netrunner/netrun1.html
I want to make this irl. Basically, rather than using hyperlinks and an address bar to navigate the net, the browser would ping all numerically adjacent IP addresses, and populate a circular matrix by using ping times to estimate the relative distance between you and the IP. As you get closer to an IP node, more queries for data about that site will change the shape of the node to reveal more details about it. When you enter the node, you'd simply see the site as you would on a normal browser.
I think that, while this wouldn't really be possible to profit from, and it wouldn't make hacking or getting your dopamine fix any easier, it would actually serve a purpose: create a more exploratory net.
Especially in the last 10 years, the net has been getting more and more centralized. Rather than a massive eclectic patchwork of niche sites, most people seem to go to a handful of one-stop shop sites. This has been referred to as the Walmartification of information. Reddit, stackoverflow, facebook, twitter, youtube, etc. I partially blame this on the domain name system.
Back in the 90s, search tools weren't all that effective and content aggregators weren't a thing. The only option available were the "portals" on your internet provider's homepage, but these almost always sent you to lame sponsored content or run-of-the-mill news. If you wanted to really "surf the net", your only option under the domain name system was type in random words with a .com at the end and hope for the best. It worked better than you might expect, especially for stumbling across porn sites (whitehouse.com anyone?) Obviously, this was not ideal, which is why people started gravitating towards the modern highly centralized net.
I want to change that. The domain name system is an abstraction on top of the IP address system. If you get a single letter wrong in the address, you either get a 404 error or land on a malware site. This is an anti-exploratory property of DNS. If we just navigate the net with raw IP addresses in a true cyber"space" on the other hand, while you may still get a 404 or malware site, it's less likely.
IP addresses also are mappable to real world locations. The actual servers. Domain names are not. This project could also go the route of something closer to the Ihara-Grubb transformation and make cyberspace closer to meatspace.
Is this something any of you free NEETs want to help me with? Any problems you can foresee other than programming novel concepts is hard? Again, the point of this is NOT to make you rich overnight or make it easier to consume infodrugs. If you want to do that, go get a job at Facebook. The main goal of this is simply to make real life more like video games, and the secondary goal is to make a more exploratory internet.
>the browser would ping all numerically adjacent IP addresses, and populate a circular matrix by using ping times to estimate the relative distance between you and the IP
So you're basically pinging a bunch of random IPs and then placing them in a 3D environment where the radius depends on the latency but the position within that radius is random? I guess it could work, but the 3D aspect would be confusing as you're dealing with a 1D line rather than 2D points.
As for getting useful content back that would be way harder. Clearweb sites often hide their IPs behind cloudflare or other intermediaries. You can't get the IP and even if you could it would refuse the connection. There might still be raw ip websites around but I don't know how common they are.
Dank net seems like a better place to implement the system on, but then you can run into all sorts of illegal shit which users wouldn't want and feds would use as an excuse to nuke the program and development team.
In short it can be done but might not work too well.
>the position within that radius is random?
Look at the background in this pic. On possibility I'm envisioning it looking something like that. An "information superhighway" more than a "cyberspace".
>As for getting useful content back that would be way harder.
Why? Literally every other browser does this without issue. At it's core, the project is merely creating an esoteric browser.
>Dank (lol) net seems like a better place to implement the system on
I mean, it would be *cooler* and more cyberpunk to create a new darknet browser that works in a badass way like I described, but I've already never created a browser before and I know darknet coding is out of my wheelhouse. And then there are the problems you mentioned.
Do you know if anything similar to this already exists?
>Look at the background in this pic. On possibility I'm envisioning it looking something like that. An "information superhighway" more than a "cyberspace".
That makes more sense. It could work if you had the addresses.
Like I said, cloudflare fucks it up. They hide the IP and use domain magic to isolate the actual server from you. No access to the server means the IP is useless as an address.
Neet.moe uses cloudflare. You can ping them and get an IP and latency, but try accessing that IP and see what happens.
>but I've already never created a browser before
Making a browser is a huge ordeal because you're dealing with the mess that is web 2.0 plus security plus everything around that core. Pinging and making a visual representation is doable, but I think the actual browser part should be relegated to an already existing browser.
>a browser is too hard too make
Yeah, a web extension for an existing browser would probably be sufficient. Any suggestions?
Not a problem because there's definitely a way to get a domain name from a raw IP. Then once the domain is found, the addon would hand the reins back to the normal browser.
>Clearweb sites often hide their IPs behind cloudflare or other intermediaries
This is beneficial for site owners. Even a free Cloudflare account adds a lot of value. Also, if you have complex infrastructure (multiple app servers, load balancers, etc) there isn’t a single IP address that correlates with a single website. There’s always going to be lots of abstraction.
I was thinking more along the lines of a program that embeds a browser or opens an external browser, but I guess you could possibly make something with JS and html5.
I didn't say it wasn't beneficial but it does create problems with OP's IP look up plan.
I do hate cloudflare in that it's a monolithic entity with a lot of centralized power over the web, which they have shown they can't be trusted with, but that's not related to the discussion.
>there isn’t a single IP address that correlates with a single website.
Not a problem. Remember, this project only has two goals:
1) allow navigation of the internet to be spatially oriented
2) allow navigation of the internet to be more viscerally exploratory i.e. you don't have to rely on clicking hyperlinks or typing random domain names, you just wander through cyberspace. The idea is similar to Meme War some other anon from here apparently made
This means websites leading to a variety of IP addresses (which in turn would lead to a variety of returned domain names accessed through the Cyberspace webapp/addon) is a feature, not a bug. The additional info that would inform the shape/color/etc of the website object and would change if the site changes. Shit, you could just manually load each site like normal in the browser in a hidden tab, and use the file size or other properties of the returned webpage document to increase the size or other properties of the cyberspace node.
If you have 30 Mbps internet, it would take a while to populate each chunk of cyberspace. I'm sure many IP addresses wouldn't be able to be used though, which will speed up the loading. As you move through cyberspace, the hidden tabs would be autoclosed. If firefox is too restrictive about their addons, I'll find a browser that isn't.
Example: Take for instance neet.moe. If you ping that, one of the IP addresses you get is 220.127.116.11. When you put that through a reverse IP lookup tool, you get pic related. Not one of those sites is actually neet.moe. *But that's okay* because that's not the point. You can already access neet.moe perfectly well with existing software after all. With regards to that example, there would be a setting to select the highest ranked site or just select one at random or even select multiple and display a few at once.
I understand that a browser extension that does what I want would NOT make the browsing experience easier or more efficient. That's not the point. It's to make surfing the net more exploratory. And to make real life a little more like science fiction.
Your wish reminds me of a 3D imageboard where you walk around and visit threads/posts floating in the air.
I'm not sure if this is the project that I came across (and the site is down), but something like this:
You could make more of a VR/3d game model and then embed websites in an iframe on floating tv panels as you walk around.
I had a high idea once to build something like this. The execution is really simple but I lost interest.
I actually contacted that guy and asked if he wanted to collaborate on my idea, but he said he's too busy. The problem with iframes is most sites don't allow you to display them through iframes. The Meme War guy found a way around it by exploiting the wayback machine, but I still don't think that'd work well for my purposes.
Don't try to make real what belongs to the digital world. It's better that way. It took me a long time to realize that.
also I wouldn't be taking anything out of the digital world, so I'm not sure what you mean. the addon/app I'm describing would still just be a new way to surf the net
1. urls are already the exploratory element
2. here's (an incomplete) yggdrasil network tree: http://18.104.22.168/
here you're not dealing with just addresses but the paths which helps put things in space. traceroute in the normal internet to build paths maybe. probably still a retarded idea.
i remember seeing such a thing running. threads were pillars.