@kat@yarn.girlonthemoon.xyz pandoc is a joy! I haven’t used any Microsoft word processing tools since forever. They want a Word document? Pandoc to the rescue!

@movq@www.uninformativ.de it seems you got plenty of choices. That’s the cheapest of their products, and you are quite right, amazing pricing! I pay Apple $10/month for a shared-amongst-family 2TiB storage space.

@bender@twtxt.net My choices might be a bit limited, at least going by this list: https://www.hetzner.com/storage/storage-box (That would be some incredibly cheap storage.) I’ll probably have to order such a box and then play with it a little bit to see what’s possible.

@movq@www.uninformativ.de there are many other similar backup tools. I would love to hear what will make you pick Borg above the rest.

@bender@twtxt.net Exactly. I suspect it was because of sqlitebrowser also accessing the database in parallel to debug the original issue.

So far, I have not found the exact reason why some replies don’t show up. When I do not filter for unread messages and show all, though, I actually see them. So, there’s that.

@lyse@lyse.isobeef.org 😱😱😱 I am assuming that’s happening with tt2, right?

@javivf@adn.org.es having the extension listed means that it has been discussed and, usually implemented. Now, number 6 and 7 on the list as its stands today are not supported by any of the known clients. I believe their (again, 6 and 7) inclusion on the list has been precipitated, and lax.

@javivf@adn.org.es merged in to the repo of specs:

@prologic@twtxt.net merged to what? to yarn? I though twtxt.dev was solution-agnostic 👀

@kat@yarn.girlonthemoon.xyz It’s more like a cache, it stores things like “timestamp of the most recent twt we’ve seen per feed” or “last modification date” (to be used with HTTP’s if-modified-since header). You can nuke these files at any time, it might just result in more traffic (e.g., always getting a full response instead of just “HTTP 304 nope, didn’t change”).

@quark@ferengi.one Yes, I often write a couple of twts, don’t publish them, then sometimes notice a mistake and want to edit it. You’re right, as soon as stuff is published, threads are going to break/fork by edits.

@prologic@twtxt.net best of luck!!! discover view having no limit sounds scary oh god lol

@kate@yarn.girlonthemoon.xyz I’ll cut a release soon™, but still a few more things to iron out 🤣 One of the new challenges is figuring out what to do with the “Discover” view now that is has an unconfined limit, on my pod (at least) it’s now basically just “noise” 🤦‍♂️

@prologic@twtxt.net alright! i can try that… whenever i am capable of upgrading and figure out how to lol

@prologic@twtxt.net :D

@kate@yarn.girlonthemoon.xyz Been missing your jovial posts too and your friends 🤣

@quark@ferengi.one No editing old Twts that are the root of a thread with replies in the ecosystem. Just results in a fork. Unless the client has an implementation that does not store Twts keyed by Hash.

@movq@www.uninformativ.de Oh dear 😂

@kate@yarn.girlonthemoon.xyz The re-fetch should work just fine 🤞

@david@collantes.us Thanks I’ll fix.

@movq@www.uninformativ.de wouldn’t editing your own twtxts cause the same issue Yarnd (or any other client) has, which is breaking any replies to it? Under which conditions would this work the best? When copying the twtxt.txt file asynchronously? In my case I copy the twtxt.txt file to its web root right away, but I figure I could not do that, which would give me a set period of time to edit without worries.

@movq@www.uninformativ.de json and database put together sounds terrifying. i must try jenny

@bender@twtxt.net very true in both regards!!! <333 happy to see you too, i missed your silly bender icon hahha. my times could be better but things will improve i am sure of it

@bender@twtxt.net awww thank you :‘))) you all are too nice!!! i really wanted to share how i did this because i think i’m the first person to publicly attempt a production instance of dreamwidth code in docker, so i’m glad i did a good job at documenting it!!!!!!!

@kat@yarn.girlonthemoon.xyz woah! That’s something else, kat! Heck, I document pretty much everything (more at work than anywhere else), and I have got to tell you, you put my “documentation” to shame. LOL. Very well done!

@kat@yarn.girlonthemoon.xyz hard times are best shared! Of course, I understand there are times they are best not shared. Either way, I wish the hard time you have been having fades away! And, of course, happy to “see” you! 🥰

@lyse@lyse.isobeef.org It wasn’t our building, yeah, luckily. But I’m pretty scared it might happen some day. I think I’ll put more effort into preparing for that. But whatever I do, it would be horrific to lose all your stuff and the memories attached to it …

@kat@yarn.girlonthemoon.xyz You’re welcome. 😅 (From a hiring perspective, it’s not even important if every detail/step is correct or not. We all make mistakes, all the time; we don’t/can’t know everything.)

@movq@www.uninformativ.de HELP THIS IS GENUINELY SO SWEET THANK YOU ;_; omg i felt so nervous posting this because i was like what if i get something wrong but then i did it anyway and i felt so free… like woah i did all of this

@kat@yarn.girlonthemoon.xyz As someone who has a say in hiring decisions (every now and then – I’m not an executive nor an HR person 😆): This is gold. Writeups like these tell me/us so much about job applicants. It’s much more valuable than “a CV without gaps” or “know your algorithms” or whatever. Instead, it shows how you work and that you understand what you’re doing, and that’s the most important part. 🥇

@bender@twtxt.net NOOOO i self host an XMPP server and also revolt but as much as i love XMPP (gajim client reminds me of using skype as a kid highkey) i don’t use it much and revolt is a bitch to maintain. like i broke revolt file uploads and it stayed that way for months until literally last week lmao. i never bothered with matrix tbh maybe i should’ve but it seems not worth it

@prologic@twtxt.net hm would there be any loss with the re-fetch option? i wouldn’t mind either but i’d like to hold onto what i got if possible! but if it IS possible but also really annoying to do i’ll just do the re-fetch of feeds because i’m lazy af LMAO

@bender@twtxt.net been having a hard time! but i am trying to be more active <3

@movq@www.uninformativ.de that is a great quote god damn

@kat@yarn.girlonthemoon.xyz i did this in september but it took me this long to do the write up part lmfao

@movq@www.uninformativ.de Hopefully at your neighbors and not your building. :-S Keep your pecker up!

I just heard the fire brigade respond here with their compressed air sirens, too.

If you let fire take effect properly, it helps to reduce density.

And, of course, I am getting a 403 when trying to edit the file. <– @prologic@twtxt.net.

Ha! I stand corrected, didn’t scrolled long enough. Indeed, it should be added (you will need an account on Mills’ Gitea), noted.

@eaplme@eapl.me you wrote:

“That PHP snippet could be merged into https://twtxt.dev/exts/twt-hash.html”

Why, though? AFAIK @andros@twtxt.andros.dev’s client is on Emacs, @lyse@lyse.isobeef.org’s is on Python (and Golang, for tt2), @movq@www.uninformativ.de’s is on Python, and @prologic@twtxt.net’s is on Golang. All the client creator needs to know is in the documentation already, coding language agnostic.

@andros@twtxt.andros.dev I give you not creating another file, but then I’d vote for commenting out DMs. See https://eapl.me/timeline/post/z5e2bna

It’s easier to find the DM in comments from your side, than asking all the client maintainers to add the regex =P
You can even use a Modified comment, such as
#! <DM content>
Or something like that

This approach is retro-compatible with current and older clients.

just a note that we are doing that on PHP: https://github.com/eapl-gemugami/twtxt-php/blob/master/docs/03-hash-extension.md#php-72

That PHP snippet could be merged into https://twtxt.dev/exts/twt-hash.html

@hacker-news@feeds.twtxt.net this is the 4th one in the row - I think. Loosing count of them.

@klaxzy@klaxzy.net Jellyfin FTW … I guess!

@movq@www.uninformativ.de this settles it then. Hopefully @andros@twtxt.andros.dev can make slight corrections, and broken threads will be a thing of the past! Thanks!

@david@collantes.us @andros@twtxt.andros.dev The correct hash would be si4er3q. See https://twtxt.dev/exts/twt-hash.html, a timezone offset of +00:00 or -00:00 must be replaced by Z.

(That said, there’s a bug in jenny as well. It only replaces +00:00, not -00:00. 🤡)

@movq@www.uninformativ.de Lol! Yeah, nobody wants to see you bring your coal-powered forklift into the gym. :-D

@andros@twtxt.andros.dev maybe @movq@www.uninformativ.de, @lyse@lyse.isobeef.org, @prologic@twtxt.net, and @xuu@txt.sour.is can explain what’s going on here, or which approach is the right one. One thing seems to be happening: your approach is breaking “conversations” (threads), as the other clients seemingly do not agree.

@david@collantes.us If I run

printf '%s\n%s\n%s' 'https://aelaraji.com/twtxt.txt' '2025-04-16T22:49:11+00:00' "Am I tripping or `rsync` is actually THIS effing faster than `scp`!!? 🫨" | b2sum -l 256 | awk '{ print $1 }' | xxd -r -p | base32 | tr -d '=' | tr 'A-Z' 'a-z' | tail -c 8

I have xqfsv6a. It is raw text

But… If I change de date to 2025-04-16T22:49:11Z I have si4er3q.

@andros@twtxt.andros.dev the hash on @aelaraji@aelaraji.com’s last message (as I type this) is:

[si4er3q] [2025-04-16 22:49:11+00:00] [Am I tripping or `rsync` is actually THIS effing faster than `scp`!!? 🫨]

So, si4er3q

@prologic@twtxt.net @bender@twtxt.net

What is the hash of the last message from?: https://aelaraji.com/twtxt.txt

@bender@twtxt.net For example:

If you can see this twt in any feed…

xxxx-xx-xxTxx:xx:xxZ	!<bender https://twtxt.net/user/bender/twtxt.txt> U2FsdGVkX1+QmwBNmk9Yu9jvazVRFPS2TGJRGle/BDDzFult6zCtxNhJrV0g+sx0EIKbjL2a9QpCT5C0Z2qWvw==

It is for you. Any other possibility must be ignore (hidden in your timeline).

If your client doesn’t have the posibility to decrypt the twt, hide all direct message. It is all :)