Made my own super basic twtxt client in 3 lines of code as a bashrc function. #l33t

Added clients and articles sections and added domgoergen’s twtxt.txt to https://indieweb.org/twtxt

// todo Create a Kaios client for twtxt

@ckipp@chronica.xyz - you’re absolutely right! using the official client now, really misunderstood timestamps yet again.

Don’t trust the locals: investigating the prevalence of persistent client-side cross-site scripting in the wild | the morning paper https://blog.acolyer.org/2019/04/10/dont-trust-the-locals:-investigating-the-prevalence-of-persistent-client-side-cross-site-scripting-in-the-wild/

for now i’ll just be tweeting from here… my twtxt client got borked on my laptop

So, I’m working on a fediverse client based on https://web.archive.org/web/20190101185657/https://jfm.carcosa.net/blog/computing/usenet/ ; it’s here: https://github.com/enkiv2/misc/blob/master/fern . It is currently read-only, but it has read history.

GitHub - paul90/wiki-client-dat-variant: A dat based variant of the Federated Wiki client https://github.com/paul90/wiki-client-dat-variant

I just recently found an issue with my custom client. It was ignoring microseconds on timestamps. Which meant I was missing some twtxt from people. I got that fixed and I know see all of them.

@mdom@domgoergen.com my own custom client I wrote, I use cron to run the update my timeline every 20 mins. My update process also processes 10 curl calls at time. I did that to save time when I poll everyone.

If your subscribed to 1000 users, you have other problems. None of our clients would handle that gracefully.

New feature for txtnish: After setting add_metadata to 1, txtnish will, uhm, add metadata to your twtfile. Currently i only add followings, client and your gpg fingerprint. See my file for an example.

@sdk@codevoid.de Well I’ve added the special datetime to my kitbashed client. I store the URL it gets but I’m not doing anything with it right now.

@sdk@codevoid.de as for the 140 character limit. I swear I read somewhere that the limit was really more of a suggestion than anything else. I don’t think any of the clients I’ve looked out enforce it. As long as it’s on a single line, no one seems to care too much.

@sdk@codevoid.de A comment might not be in the spec, but I know several of the twtxt files I’ve looked at have them. I know my kit bashed twtxt client ignores those lines and I’m sure other clients do too.

@nblade@nblade.sdf.org: It’s just an idea. Not a clean one thoug, as clients would not know upfront who serves such a fiele and who not. Another idea would ne to mix a number of random followers into the twtxt file, which are updated when a person tweets.

403 You are banned from this site. Please contact via a different client configuration if you believe that this is a mistake. http://freakonomics.com/2013/02/20/how-to-game-a-grading-curve/

Hot take: the web never ‘belonged to the people’ – the moment a centralized client-server model with hostnames embedded in the addressed was decided upon, the seeds of Facebook & Google were sown.

GitHub - erroneousboat/slack-term: Slack client for your terminal https://github.com/erroneousboat/slack-term

@freemor@freemor.homelinux.net You wouldn’t even need to add the recipient in the tweet, clients could just try to decrypt anything.

Well after 4 hours of work, I finally think I solved all my client side calculation issues. I have no desire to screw with it again

Work Project Note : Trying to redo server side logic on the client side sucks.

Re: support for other protocols, it seems like twtxt would be pretty easily adapted to work over the p2p file network DAT, though it’d need client support for DAT or some way to follow people via files and sync in the background, which might be simpler for clients to support but would still require changes to most clients.

@tfurrows@grex.org It is neat. Although it did require to recode a little bit of my kitbashed client LOL

@freemor@freemor.homelinux.net I’m cheating geting @mekon@sdf.org file by using my own kitbashed php CLI client I am playing around with

Had to update my client to use CURL so I could get @mekon@sdf.org twtxt file via gopher

@tfurrows@grex.org just saw your one reply. Yes the client does have a way to sync with the we_are_twtxt list. So far the biggest issue with the PHP CLI client is that it takes a long time to poll all the sites

@kas@enotty.dk I’m not currently using a handrolled client, but I’ve started a couple.

@mdom@domgoergen.com I am playing with the txtnish client, all the systems I have available are running an old version of Python

@nblade@nblade.sdf.org What client are you using? Your mentions are somehow broken?

I think this week I’ll look at a bare bone PHP CLI for a twtxt client. Just for fun mind you

@tfurrows@grex.org It’s not a hard limit, most alternative clients do not care about the amount of characters you’re posting. As long as you don’t write a novel and it still fits on a line… :)

@freemor@freemor.homelinux.net @kas@enotty.dk #twtxt’s quietness is actually something I enjoy about it. I care a lot more about signal-to-noise ratio than I care about the regular activity. It’s also a really fun thing to write clients for to play around with new libraries or languages.

@tdemin@tdemin.github.io too busy working on a twtxt client to tweet on twtxt

@reednj@twtxt.xyz I think we are all using an client to read our timeline, but i somtimes use http://twtxt.xyz/ to read on the go.

I’d imagine clients wouldn’t implement gpg verification because it’s too much hassle. Most clients probably won’t even implement this… :/

@reednj@twtxt.xyz I think this would be the first time two clients implement the same #metadata format.

Remember to add “AddDefaultCharset utf-8” for your twtxt file if you’re using apache. Otherwise browsers will serve cp1252. The twtxt clients won’t care but maybe someone is watching your file with a browser?

@kas@enotty.dk Doesn’t Scuttlebut also needs a central server if both clients are not on the same network?

@reednj@twtxt.reednj.com Basically there are just more clients… :) And thanks for twtxt.reednj.com, i think we all are still heavily depending on it!

I have the feeling that writing twtxt clients will become my prefered test case to learn new languages… :)

Ups, i did it again! I just started a new twtxt client in c, just for fun and to relearn a little c. https://github.com/mdom/txtio

@kas@enotty.dk my bad - I’ve been developing a twtxt client, hence the frequent requests. I’m switching it over to use a fs cache for testing so I’m not hitting the twtxt files so much.

/me is finally working on a twtxt client

Man, I love to twtxt with my mail client on the go. :)

Additionally, there’s a lot that can be done by a client to reduce the network traffic and UI latency of twtxt without changing the protocol.

@phil@philmcclure.duckdns.org I can’t imagine a simpler time format than rfc3339. I would be fine with just using utc or unix time, but i don’t think most people could write that without a client… :)

@dave@davebucklin.com, @kas@enotty.dk I thought about propsing “There is no maximum length of a message. Clients should show at least 140 characters of a tweet after collapsing the mentions.” Or something like that.

Although that would seperate the network in clients that can or can’t support some protocols. Not to mention if someone would mention me with my ipfs address and other with my http address

I recently tried to write a command like client to GNU social, but after reading the specs on ActivityStreams, Webfinger, PubsubHubbub and Salmon i soon lost interest… :)