Nitter and other Internet reclamation projects | Hacker News Reader

Nitter and other Internet reclamation projects

drewdevault.com

129 pointsdecrypt27 comments

I spent a few days experimenting with recent open source p2p and anonymizing networks (tor and i2p). I'm impressed the thing still works well. Some networks became much smaller (kad and gnutella2), but given they're distributed/decentralized, they still work. SoulSeek is alive. Tribler allows to find torrents without needing trackers, it also adds a anonymization layer. I2p gives some intuitive names to hidden services while tor has never been so easy and fast to use.

On the web, personal blogs and sites, planets and fora still exists. Neocities is incredibly cool. There are a number of alternatives to google and they're constantly improving.

We still have IRC and a few newer services like xmpp and matrix.

Ad blockers take some time to configure but work well and there are good interfaces for annoying services like youtube and twitter.

The bad internet exists mostly for commercial, news sites and streaming services, but there are alternatives and there's enough people using them to keep them alive and sustainable. It is not like the good internet stopped existing, it is more like the bad internet became more popular.

Try Scuttlebutt aka SSB too.

Firefox add-on to automatically redirect similar websites:

Twitter -> Nitter

Reddit -> Libreddit

YouTube -> Invidious

Instagram -> Bibliogram

Google Maps -> OpenStreetMap

https://addons.mozilla.org/en-GB/firefox/addon/privacy-redir...

I'm using this for about a month now and the only problem is the default instances that are used are getting rate-limited/blocked; so you will have to find or host your own fresh instance.

This is one of the major problems with distributed / federated systems.

The name itself (twitter, youtube, etc) is meaningful to users. These "default instances" need ways to distribute the compute & network load while allowing users to only remember "nitter".

Are there any projects that address this? Would be curious to help.

It doesn’t need a separate project , all it needs are hosters to just agree to be behind a load balancer together.

That.. that sounds like a project :)

It doesn't need to be "separate", but it does seem that many of these systems could use the same load balancer design.

This is my experience as well. I've typically got to refresh the average linked tweet about 3-5 times before it cycles to a federated instance that can actually render the page.

https://news.ycombinator.com/item?id=28575855#28576302 (last week)

[autoliteInline] The amount of cruft on the web just blows me away, whether it's a weather or real estate or recipe site. We're living in a world of shit.

[wizzwizz4] Treat the crappy site as an API and make a better interface. Like the SimpleWeb project: [https://simple-web.org/]

Copied from the page, which features several projects in the "reclamation" category:

• SimplyTranslate: A frontend for Google- and LibreTranslate (and in the future, potentially other Translation Engines as well)

• SimpleerTube: A frontend for SepiaSearch and PeerTube

• SimplyNews: (No known Instances) A frontend for numerous news websites

• FreeBay: (Inactive) A frontend for eBay

• PornInvidious: NSFW! A frontend for xvideos.com

> Medium et al, via an open source readability-as-a-service platform

I usually throw Medium articles into Outline[0] for readability. There is also Archive.today[1] where the article is usually mirrored already. My only issue being that Archive.today could go down since it's expensive to run such a service.

> Facebook

> GitLab and GitHub

As for Facebook, I think the only way such a frontend to that would work is to use a Facebook account to scrape walled garden content and present it externally to users. I'm not sure Facebook would like that however since it hurts their bottom line. They would probably code against such tools, and then we have a whack-a-mole scenario.

As for Gitlab & Github; I see no reason for a frontend since they don't hide things behind a login prompt and their UI is pretty intuitive (albeit a bit bulky and bloated).

[0] https://outline.com/

[1] https://archive.ph/

> As for Gitlab & Github; I see no reason for a frontend since they don't hide things behind a login prompt

They do. Not at the Facebook scale, but they already started moving in that direction: unfolding issues with many comments requires login on GitHub, searching for code too - it does not prompt you to log in, in just brings 0 results to the unlogged

archive.today doesn't work if you use 1.1.1.1 for DNS. It's very peculiar to me that a service that helps the web be more open feels it necessary to break the web by refusing clients who use a standards-compliant DNS.

https://news.ycombinator.com/item?id=28459600

I tried Invidious (it was very easy to set up with Docker Compose) but the UI looks terrible. At one point I wondered whether CSS wasn't loading, for some reason.

I know OSS is "don't like it, don't use it", but why do designers rarely contribute? Spending an hour or two just adding proper spacing to elements would help immensely.

It's not "why do designers rarely contribute?", it's that almost only programmers contribute. As to why, I have honestly no idea. These days I would say that it's the culture influencing new people, but I wonder how it started.

OSS is not "don't like it, don't use it", OSS is "don't like it, change it!"

A list of similar sites:

https://github.com/mendel5/alternative-front-ends

These tools should be halfway homes for a decentralized alternative.

For example, when you make an account on Nitter, maybe it lets you post to both Twitter and also to a decentralized protocol at the same time.

Eventually, you can get most of your content straight from the decentralized protocol.

This would be a great way to get more content elsewhere.

nitter has no accounts, it's read only.

Would be interested to hear about the user hostile features of github/lab? I've used both quite a bit and they seemed fine to me?

Everything social, anything that shames me for not engaging more with the platform, and poor user experience when using a browser that doesn't support Javascript.

Should have been disclosed: The author makes a competitor to GitHub/lab

Yeah, it's really annoying that Twitter is following Instagram on the "fuck unregistered viewers" path. Looks like they're really desperate for the last bits of growth…

All of these services are more useful, more accessible, and more inclusive than their corporate counterparts. They work better on older browsers and low-end devices. They have better performance. They aren’t spying on you. In short, they are rejecting the domestication of their users that the platforms they interact with have been trying to do.

No matter how much I want to like and recommend these projects, mainly for those last 2 sentences, unfortunately the rest of picture this paints doesn't seem to be completely true for me: while Nitter for example is ok, I've tried to switch to e.g. Invidious so many times and in the end it's just a pain to use. Most of the time it loads slower, if it even loads, and every x days I have to try another server, etc. So in the end it is simply worse than the counterparts on those fronts, and I have the impression using e.g. youtube in private windows gives most of the other benefits as well.

I don't find gitlab/hub are user-hostile (the author runs a competing service). For other service, the answer is the fediverse, but author stopped publishing on mastodont (please continue).

The irony of medium is that it started as a clutter free expression platform.

I like the idea of nitter and bibliogram but they are so unreliable that I always end up giving up on them.

Reply on news.ycombinator.com