twtxt isn't very popular.
I think the feed format itself is neat, other than the tab, but maybe there isn't enough of a niche for it.
I have implemented a twtxt version of the Atom and the JSON Feed feed [1] for my site at https://dbohdan.com/twtxt.txt.
The generator I originally developed for the site creates twtxt feeds, too: https://github.com/tclssg/tclssg.
I haven't seen it in another static site generator or plugin;
please link if you know one.
quite https://lublin.se/twtxt.txt 2024-12-23T03:15:21-05:00 <ROBOT VOICE> THE LAST HUMAN POST ON THIS FEED IS MORE THAN FOUR YEARS OLD. PERHAPS TWTXT CLIENTS SHOULD THEN FETCH THE FEED *VERY* RARELY.
quite https://lublin.se/twtxt.txt 2024-12-23T03:15:16-05:00 <ROBOT VOICE> THE LAST HUMAN POST ON THIS FEED IS MORE THAN FOUR YEARS OLD. PERHAPS TWTXT CLIENTS SHOULD THEN FETCH THE FEED *VERY* RARELY.
quite https://lublin.se/twtxt.txt 2024-12-23T03:15:07-05:00 <ROBOT VOICE> THE LAST HUMAN POST ON THIS FEED IS MORE THAN FOUR YEARS OLD. PERHAPS TWTXT CLIENTS SHOULD THEN FETCH THE FEED *VERY* RARELY.
[...]
[1] This is why you shouldn't include the word "feed" in the name of your standard for feeds.
I was using it for years before moving to Mastodon, then Nostr.
It did work rather well, is easy to code for, and I still have people pulling from my twtxt file, but it does get a bit tiresome managing follows etc with the clumsy apps.
And not having a decent mobile app made it less fun to use on what for me is social media's primary use case - while on the loo hahaha.
It's quite buggy. Any minor change from the norm (config directory, txt file directory) seemed to break it. Finally I went with the standards and it still had trouble. `twtxt following` gives you errors. At first I thought it was because I wasn't following anyone, even though I chose to follow the twtxt news feed, but I never got rid of the error. I got errors about "feed not available" even though the txt files were there (maybe version differences?)
It sounds kinda interesting, honestly, but I give up.
The Twtxt/Yarn community is larger than you think. As the founder of Yarn.social[1] (which itself uses the Twtxt spec and extensions[2]) and operator of the "flagship" instance twtxt.net[3] I often interact with around ~70 folks (_not including news feeds_).
Yeah sure, but as I said the community is actually much larger. It is also very hard to measure because Twtxt/Yarn is what I call, "truly decentralised". However you are right, the search engine/crawler puts the active feeds at around ~1000 or so. So orders of magnitude smaller than any "big tech" social ecosystem, but that's kind of the appeal really.
I will never understand why all these minimalist content publishing systems pretend that visual imagery is unnecessary.
Humans have expressed themselves in images long before they have expressed themselves in text.
Any system that forces humans to express themselves purely in ASCII, monospace, and monochrome is crude and borderline disrespectful towards human expression.
It made sense in 1983, when the technology to power these expressions was not prevalent. But why do this today? I know they exist, but who are the people who confine themselves this way? Is it acknowledged as an act of self-discipline or self-restraint? Some sort of an artistic statement? What is the process through which one arrives at "I will commit to this system, and always only communicate in written speech, laid out in uniform monochromatic typographic format. I will walk into these constraints voluntarily."
Any medieval monk writing manuscripts, any ancient Greek or Roman or Egyptian or Chinese capturing history and literature on parchments would have gasped at such a thought.
Writing evolved after visual imagery to solve the poor ability of visuals to express abstractions. It too is based on imagery, thats what script is after all, but leverages the emergent social pattern of literacy to convey highly encoded information.
This invention (that we now take for granted) has been so exalted in the minds of earlier generations that they would go to extremes to banish imagery (iconoclasts etc.)
It takes its most extreme form in mathematical script: concise expression of the most abstract ideas.
Commingling imagery and text has its many uses of course: try describing a graph or diagram or a real scene using text only. But for other purposes (literature) its a distraction that forces the brain to switch mode and diminishes the experience.
> Yet one thing that these minimalistic text-based interfaces are terrible for is rendering mathematical notation.
You can notate mathematics using text. In fact the mathematic notation rendered to vector graphics/bitmaps is typically rendered from such text-based notation.
Mathematics is increasingly done using text-based notation (e.g. Lean, Mathematica, SymPy), which I actually prefer because it's typically unambiguous unlike the traditional notation with wildly varying conventions and a lot left implicit or even ill-defined.
They must live and die by their choices but I am defending the category as a whole. Both limitations can be lifted without altering the essence of a distraction free reading flow.
I gather from your text that you perceive images to be presentation and monochrome monospace text to be content.
Isn't all expression in a nutshell an act of presenting content?
Is emphasis content, or is it presentation? What about lowercase and uppercase? IS THAT CONTENT? Or "simply" presentation?
All expression is an attempt for our content to be perceived as we intend. All expression is a presentation of our content, including the very text of it. The text itself is simply one of the dimensions of our content's presentation.
I figure there's value in sharing why one would not want to use something that's being shared/offered here.
Hacker News is a discussion platform. People submit ideas with the expectation that the community would find them interesting enough to discuss them.
When someone submits an idea of text-only content publishing, is it not perfectly reasonable to discuss one's thoughts on that idea, whatever they happen to be? And if those thoughts didn't resonate or apply, would lack of upvotes not conveniently push these thoughts to the bottom of the page where most never reach?
If I was an author of such a platform/idea/product, these are exactly the kind of responses I'd be most interested in reading. Not the feel-good high-fives, but the "what did I miss that others find important"s.
I think it’s turned into the place to submit work when you want as much harsh feedback as possible. You can get upvotes but need to run the gauntlet of “why isn’t it open source/here’s a better app/this is dumb” etc. Top comments are often pretty caustic, so that’s the culture we’ve built. I guess if you want to sharpen your offering really fast, this is optimal.
“Any system that forces humans to express themselves purely in ASCII, monospace, and monochrome is crude and borderline disrespectful towards human expression.”
What is the outcome you’re hoping for, by giving that feedback? Should we kill the authors now or just shun them until they give us pictures?
Edit: every single system makes choices. You either try please everyone or you choose a set of people to focus on. This one isn’t for you, and that’s ok. It may fail and that’s ok too. I don’t think it’s disrespectful to human expression, if anything it’s a form of human expression.
There is a cargo cult of absence of features because old cool developers are used to email, IRC, EMACS, etc (stuff they got acquainted with in their college days 30 years ago).
It's clear from the demonstration video alone that this is not an ASCII system, and indeed reading the doco confirms that the text files are considered to be UTF-8.
This gives some irony to your analysis. Because this system thus permits one of the the very same ancient expression-in-images systems that you are alluding to: hieroglyphs are in Unicode.
I expect that in the 9 years of its existence, almost no-one has ever used them in this system.
Alas, it isn't valid if one goes beyond even reading the doco.
The underlying library that the tool uses, click (https://click.palletsprojects.com/en/stable/), turns out to include the idea that the "tweet" strings can contain ECMA-48 control sequences.
This permits not only boldface and italics, but also (on very old terminals and very modern terminal emulators) underline, strikethrough, faint, reverse video, invisible, and even 8 whole colours. (-:
Again, though, I expect that in the 9 years of the system's existence, no-one has actually used this in earnest.
In part, this is because the default mode seems to be to apply a regular expression substitution to attempt to strip out control sequences, because of course ECMA-48 and ECMA-35 are over-expressive permitting things like OSC, NEL, PM, APC, cursor motions ("layout is content"), insert/delete/erase, and code page changes.
Amusingly, the regular expression substitution is not based upon an understanding of ECMA-35 and is faulty.
I promise it isn’t. English however isn’t my native language — I grew up in Armenia, with Armenian as my native tongue and Russian as my first “foreign” language.
Interesting. I'm in a similar situation, but my first foreign language was French. I believe the source of English education makes a difference. In my case, it was Rambo movies :))
In this case, because it's a text file based system that works through the CLI. It is kind of ironic though that the "Demonstration" in the introduction is a video
I used to enjoy using browsers like W3M, if you can just throw assets at your favourite tool of choice, like an image viewer, a video player, it just feels better for me as I don't have to learn a different interface for every site I visit.
> I will never understand why all these minimalist content publishing systems pretend that visual imagery is unnecessary.
I don't think anybody is prevented from putting a link to an image, nor to build a client that autoload linked images (the same way that say, some gemini browsers like lagrange can be configured to show you images by default).
What I like about nostr is the experimentation with different clients, and the fact that your identity is portable to any of them. It's just a really cool model.
I'm working on a nostr-based discussion site heavily inspired by HN/reddit: https://oddbean.com/
No login required, and you can switch to a different client any time you want. We try to keep the bitcoin/politics stuff to a minimum on the homepage. There's a lot of that on nostr, but also a lot of kind and thoughtful discussion about diverse subjects.
The upvotes on the pronunciation in the sibling comments should be considered legally binding and will be the authoritative pronunciation. Upvote carefully.
Huh, in my head I was reading it 'twit-text' (this is not meant to be a pejorative comment), but I guess that is ascribing it another 't' where there isn't one
There is a long history of confusing or weird project names in computing like sqlite, gif, Splunk, Hadoop, Coq, MongoDB (from humongous apparently), yacc, C, R, and X (the window manager; not a lang or the social media site).
Simon Willison has one of the more prolific microblogs going and just did a meta review on his approach including the mechanics (surprise surprise it's Django). Could be of interest to anyone double clicking this thread.
My approach to running a link blog
22nd December 2024
I built a very (very) similar tool called terss (for "terse" and "RSS") years ago in Bash on top of RSS feeds (using core-utils and more-utils — in particular for 2xml and xml2) for interoperability because it made the whole thing more easy to produce and consume with other tools. I'll try to find if I still have it around somewhere :).
Not a bad idea, light enough to implement a telegram bot everyone can use or email or maybe a bitlbee plugin so you can have a twtxt channel.
But, what is it? can you PM someone? participate in a conversation? or is it just that, a microblog? Usability wise what will you do? keep a client running that always checks if new blogs have happened over, how many followers? I'm currently following 500 people, that's 500 connections if this takes off.
It would be nice if an rss client supported it, until then ehh
This is literally just a shared text file with buzz words thrown in the title. You could achieve the same thing with a Google Drive link to a text file.
What problem is this solving? Not a criticism, just genuinely curious. On X, I can follow hackers, I can block/mute and curate as I want. Just not sure who asked for this.
Peertube [0], Lemmy [1], and Pixelfed [2] all run on the same protocol as Mastodon. You can interact across all of them (e.g. Mastodon user replies to Peertube video post), but the presentation on each is very different.
There's more as well. Like Friendica [3] feels a little bit like Facebook, and Hubzilla [4] feels a bit like Google+ used to. Diaspora feels a bit more basic, but similar concept to Friendica.
pixelfed looks nice. Instagram has been the only social I kept around because it's nice to look at pictures but they've recently transitioned to full on Tiktok/yt shorts competitor, I'd love to jump ship to something that's just photographs again.
Never tried lemmy, are the separate servers essentially "subreddits" ? That's cool that there aren't any global admins so each community really can have their own moderation rules.
lol this was not at all obvious to me at first glance from the framing of 'choose a server', but I guess that's the whole value proposition of activitypub is it doesn't matter what server I started on.
I actually liked bitclout, seemed like it would work really well for artists' fundraising for an album release or a tour etc. featured no-fee "tips" of all sizes right next to the like button and kept users addicted with day trading mechanics tied to artists' popularity. Money flowed freely across borders and moderation was going to be handled by clients.
afaik there was never any rugpull of the base currency, but it turned out to be a platform ideally suited to rugpulling and impersonating celebrities. Plus if you thought drama on Twitter was bad just try pinning your bank balance to individuals' reputations and see how well everyone gets along.
perhaps someone can implement it in the bluesky pds as an additional feature https://github.com/bluesky-social/pds put caddy in front of it for brotli compression so it serves fast on 1k2 links
Twitter is precisely the reason we need these decentralized, niche platforms. It's a return to the norm prior to social media centralizing everything.
They won't avoid the same fate, because decay and entropy are inevitable. But they will last longer simply because fewer people know about them and use them.
These things are good or bad because of the community of people who use them, not the technology per se. And that community is shaped by both culture and moderators.
The medium is the message. Any microcblogging platform that tracks and displays engagement inherently optimizes for short form swipes and "rebuttals" and outright lies. The "real time feed" nature optimizes for taking zero time or effort to confirm anything that anyone says. Nearly everything that "breaks" on twitter and doesn't make it to actual reports was an outright lie.
You can't put a necessary amount of nuance in 140 characters.
I disagree. The shape of the instrument makes some sounds easier to make than others. Pianos are inherently polyphonic while trumpets play one note at a time. Anything made in the shape of twitter will result in twitter like behavior.
The original Web was the work of genious, but everything since is either adtech enshittification or cumbersome hacks.
The "simplest" requirement - decentralized discovery of what is out there, is essentially still unsolved. That is one of the main reasons we have centralized platforms of all types instead of some RSS on steroids design.
The second (and more difficult in sociopolitical terms) requirement that is still unsolved is the "active" web, decentralized POST-in on somebody else's server. Here you have social challenges (identity, spam, moderation, fake news etc).
We really need a good society adapted Web 3 evolution, because Web 2 has been a disaster that keeps on giving. But it will require genious at least commensurate with the original.
How do you know what’s happening elsewhere? Other than having relations, it falls under reporters to propagate news. How do you meet new people? You go to special events and gatherings. The web is already linked, but we have special nodes like search engines, directories, and forums that are information hubs.
Creating a website was always easy. The minimum html you need is very small, and all you have to do is copy the files with an ftp client. Then tools like wordpress came and it became even easier.
What social medias have done was to put everyone in the same space. First there were walls and you just have to build your own information hub. Now, the platform is providing you its own like a private television service (including ads) whether you like it or not.
No need to invent a new version of the web, we can always go back to what was working and is still working.
You must like it for political reasons, because from a technical standpoint the new Twitter is a catastrophe. It doesn’t even work properly. The other day I typed in the URL of an account I wanted to check, from a browser where I wasn’t logged in, and I got an endless loop of redirects. Watching Musk tear down Twitter infrastructure over the last two years has been like watching the Notre Dame fire, except if it was set on purpose. It was a miracle of human accomplishment and now it’s a shell of its old self.
The directory at https://git.mills.io/yarnsocial/we-are-twtxt gave me an error, but I found it in the Internet Archive (19th September 2024):
https://web.archive.org/web/20240919022045/https://git.mills...
Here's a live example from that list: https://niplav.site/twtxt.txt - and that one shows ones its following, this one has recent posts (from December 2024): https://txt.sour.is/user/xuu/twtxt.txt
The last commit to https://github.com/buckket/twtxt/commits/master/ is October 2023, so I don't think this project is 100% thriving at the moment.
Update: Aha! Found https://twtxt.readthedocs.io/en/latest/user/registry.html and via it https://registry.twtxt.org/api/plain/tweets which shows some recent content across the network.
Also https://registry.twtxt.org/api/plain/users looks to be a list of users, though I couldn't figure out how to paginate it (using ?page=3 doesn't seem to work, despite that being listed on the https://registry.twtxt.org/swagger-ui/ page)