Nitter: Alternative Twitter Front-End | Hacker News Reader

Nitter: Alternative Twitter Front-End

github.com

274 pointsducktective50 comments

When Nitter first came around, it was incredible. It really goes to show just how easy it would be for Twitter to provide a fast and accessible version of their site if that was their goal. Too bad it isn’t and Nitter instances will forever run into rate limiting.

Page says “Uses Twitter's unofficial API (no rate limits or developer account required)”. What is this unofficial API?

I would guess that it's some kind of reverse-engineered internal API that official apps use.

It means using the "guest token" (gt) for sending GraphQL queries. The gt is public and is distributed via the twitter.com public web page. The gt is the same for every member of the public.

To see the gt, read the contents of twitter.com

For example, using Chrome

    chrome view-source:mobile.twitter.com  
    Ctrl-F gt=

or curl

    curl https://mobile.twitter.com|grep -o gt=[0-9]*

To retrieve tweets three steps are required: 1. retrieve the "guest token" (gt), 2. retrieve the "REST ID" (rest_id) for the twitter.com user and 3. submit a GraphQL query to retrieve the user's tweets.

Javascript provided by Twitter in the twitter.com can do these three steps automatically (but this requires using a Javascript-enabled browser), or it can be done without a browser, e.g., with a different scripting language (personally I use the shell).

For example, some folks use Python:

https://pypi.org/project/twint/

Wow!

But there must be some kind of IP based rate limiting or something, right? At least for DDoS protection even if it's not intended to prevent scrapping.

Never have I been more happy to be wrong. It used to be impacted by rate limiting quite a bit, but it looks like the situation has improved, probably due to this unofficial API.

The old rate limit errors were false flags caused by a bug in the API token parser. Nitter instances have yet to run into a real rate limit.

I see. Thanks for the information.

Ditto for YouTube - I remember being able to watch videos on it 10 years ago, with even older hardware and software, and the site was far more responsive than it is now.

Yet it seems if you tell the average web developer that those sites should be accessible with software and hardware from that era, or even older in the case of Twitter (rendering short snippets of text and some small images doesn't require the latest technology, seriously), they'll become confused or indignant, or try to use some absurd argument to justify it.

Twitter and YouTube are great examples of a lot of the disturbing trends in web development today. Whatever it is, it's not "progress"!

I don't "contribute" to either one, nor do I have an account; I just go there to read or watch what others have contributed.

> how easy it would be for Twitter to provide a fast and accessible version of their site if that was their goal

You can spoof your useragent to a mobile handset and visit https://mobile.twitter.com/ if you like a more bare-bones Twitter interface.

Yes, but as of last December (I believe Dec. 15), it still required JS to be turned on.

>Too bad it isn’t and Nitter instances will forever run into rate limiting.

Not if you run your own for yourself.

privacy redirect is a browser extension that can automatically redirect twitter links to the nitter instance of your choice. it also handles youtube->invidious, instagram->bibliogram, reddit->teddit, and more

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

chrome https://chrome.google.com/webstore/detail/privacy-redirect/p...

It's a shame Mozilla butchered Firefox mobile, cos this still doesn't work on it.

Werks on my machine.

There's an anal process to enable it, and I think it only works in beta/fennec, but it works.

It does work on Firefox Android but the extension is not part of the default set. You need to use a version of Firefox that supports Custom Add-on Collections like a Nightly build or the F-Droid Fennec build. Once you've got this installed you need to enable the debug menu by tapping the Firefox logo on the 'About Firefox' screen five times (see [1] for more info), add the Firefox ID and repo name for a repo with the required extension in the 'Custom Add-on collection' section and install the extension the normal way. Luckily the author of PrivacyRedirect has created just such a repo [2] which you can you to install the extension without the need for a Firefox ID (which I do not have and do not want, I'm still waiting for an easy way to run a Firefox auth/sync server on my own hardware) or having to create such a repo yourself.

Once installed you'll quickly find out the 'More options' page - on which you can configure which servers are used for redirection - does not appear. Fortunately the actual URL is entered in the location bar, it just doesn't open the page. Simply press the button, then close the 'Settings' page using the arrow on the top-left of the screen, then select the location bar (which should now contain a moz-extension: URL) and press enter on the keyboard.

The extension works fine, at least for me, on Firefox Developer Nightly on Android.

[1] https://blog.mozilla.org/addons/2020/09/29/expanded-extensio...

[2] https://github.com/SimonBrazell/privacy-redirect/issues/140#...

Sounds like people were getting tricked into installing dodgy extensions, so they had to build a Proof of Nerd system.

PoN - add some blockchain and build a currency out of it - NerdCoin.

No, I assume the real reason to be the fact that many extensions need to be adapted to mobile as can be gleaned from the non-appearance of the 'More options' page in PrivacyRedirect. Mozilla just chose a rather clunky and ham-fisted way to keep untested extensions from running on mobile. They could have added a simple option to allow untested extensions from the standard repo with a disclaimer of such a configuration not being supported.

I recommend Mull! Fennec plus arkenfox defaults and they've got an upstream repo (DivestOS official) for quicker updates than F-Droid's builder (same day vs. up to a week)

Also, all your extensions need to be in a collection together, so this isn't the most useful.

You can use mine, which contains all the extensions I use, including privacy redirect, plus a few bonus extensions I don't normally use but which I think are useful: https://addons.mozilla.org/en-US/firefox/collections/1383278...

>instagram->bibliogram

no longer works.

My experience as well. Invidious also breaks all the time, except for the most popular channels around.

Its a shame, is there an equivalent out there?

The best solution I guess would be to use a burner account with bibliogram, but currently this feature doesn't exist.

Nice! I'll give this a shot.

There's also 'UntrackMe' for Android: https://fedilab.app/wiki/untrackme/

This might be a shot in the dark, but: is there a Twitter "frontend"* that is as clean/compact as HN or even old.reddit? I get it was made with phones in mind, and that its primary use case is real-time news and short messages from influencers, but, as someone who primarily consumes social media on a desktop and whose interactions with the platform consist of reading tweets from niche accounts, i've found the way Twitter displays conversations to be -much- worse than what we have here on HN.

*I use "frontend" quite liberally here, i mean, honestly maybe a browser plugin that shrinks a tweet to half its size, indents replies much like HN comments are indented, and loads more tweets in a single page, maybe that would be enough to make Twitter 90% more pleasant to use.

There certainly are. I quickly wrote one 'mvp' for myself and it is the primary way I interact with twitter since several years, but it does the bare minimum I need and I never had time to clean it up and publish it (also I do not know whether anyone would pay for a cleaned up, purely chronological twitter feed with configurable filters?)

Probably wouldn't pay for it. Release it as open source.

Tweetdeck offers a great interface and it's also an official client run by Twitter. Better information density. I have 20+ different columns with different focuses or "priorities".

The ability to get RSS feeds of twitter users and twitter keyword searches from nitter is the only reason I still pay attention to that ecosystem at all. It took a long time to find a semi-stable nitter mirror but things have been running smoothly for almost a year.

If anyone's trying to tame their twitter, I'd recommend `refined-twitter` chrome extension and `tweak-new-twitter` extension. https://chrome.google.com/webstore/detail/tweak-new-twitter/... These two significantly help me remove useless features (e.g. Trends, Moments, recommendations) and even stuff like retweets which I mostly don't care about.

Old discussion: https://news.ycombinator.com/item?id=21849744

Thanks! Macroexpanded:

Nitter: Alternative Twitter Front-End - https://news.ycombinator.com/item?id=21849744 - Dec 2019 (38 comments)

Could this be used as a base to fork the existing ecosystem away from centralised control and back to its users? Say using decentralised storage and a synched side chain?

That would be amazing. Even an opt-in system that automatically mirrors your tweets to Mastodon or something would be useful.

twitter ratelimits nitter servers, so you'd need every individual user (or a good portion of users) running a personal server and then you'd have to collate the data somehow. and at that level of effort it would be easier to just convince people to use mastodon.

It can work as a migration path for those on the fence, and that could still work in a decentralized manner.

Example: I run a Mastodon instance on communick.com. I am thinking of running a nitter instance AND a secondary instance of Mastodon (or Pleroma) that acts as twitter mirror. I can then let users of the main communick instance to follow people on twitter, but still keeping them inside the Fediverse. The idea gets a bit more complicated if we want to make it two-way bridge, but I believe that can be done with a bit of Pleroma's MRF magic.

The beauty of this solution is that it does not require any change from those that have no interest in moving out of Twitter, but it does break the garden walls.

It feels like there could be a way to automagically solve collation as part of decentralising the storage. If there were deterministic way to create the decentralised hash for example, and then it could simply attempt to load it before committing it on fail. The user could be provided a constant heads up on side chain synch.

Have never run a nitter server, so maybe this is not possible - but could one be included as part of a web client? Ie, user surfs to "twinitter.com", receives a nitter server that takes their twitter credentials and starts posting their activity to the new chain and optionally to twitter also. If the user supplies no credentials, the server would simply run like a user-led crawler, adding more data to the new chain in what could be presumed to be some kind of optimum fashion, as it would always be the content users of the dual chains were interested in.

This isn't actually true, the "Instance has been rate limited" error was a false flag. It turned out to be a bug in the API token parser, found and fixed by a contributor. Nitter instances have yet to run into a real rate limit.

You can definitely follow Twitter users. There's BirdsiteLive, a Twitter->Fediverse bridge which allows following Twitter accounts by reference from Mastodon (or any other Fediverse server). I'm using that to follow a small handful of users.

https://github.com/NicolasConstant/BirdsiteLive/blob/master/...

Nitter also supports RSS, so you can follow Twitter accounts in your feedreader.

What you can't do in either is reply back to Twitter profiles. BirdsiteLive sets the bridged toots (equivalent of tweets) as private, which on Mastodon renders as unsharable.

I’m a little sad to see this on the front page, because even though it’s just a matter of time before Twitter nixes this workaround for their recent user hostile change, it might have otherwise lasted a bit longer.

Nitter has existed for roughly 2 years already and I think Twitter knows about it even without this HN post :)

Quick and dirty bookmarklet that will replace twitter links on the current page with equivalent nitter links, or if already on a twitter page, redirect to nitter:

http://ix.io/3wSM

To use: Create a browser bookmark with its URL field set to the contents of the above ix.io page. Then, when on a page with twitter links, click your new bookmarklet before clicking the links. Title bar text will briefly say what the bookmarklet did.

It's fun how it works the first three times, then less fun when it complains about being "rate limited" on the fourth time.

Any rate limit errors are false flags caused by bugs, they aren't related to load or real rate limits

I'm not sure why this is down-voted despite it being posted by what looks like the original developer according to their username: https://github.com/zedeus/nitter

I have no idea, maybe "false flag" was too incorrect? You're right, I am.

"False flag" is pretty close to loaded language these days.

"Spurious error" would probably be a better term.

Thanks, BTW, amazing tool.

Is it just me who thinks UI/UX of the web app feels like MySpace?

I choose to interpret this as praise and hope that was intended :)

Reply on news.ycombinator.com