XSF Discussion - 2020-02-12

jubalh: there was also a heated debate, with some people claiming that your approach is the most sensible one

those people lost

poezio still does it that way fwiw, it's on the TODO(tm) to change it

jubalh: so essentially, you need to store the first message id and the last correction's message id and match on either

Welcome to XMPP!

The obvious solution would be to introduce another ID.

Holger, I escape issues by not sending more than one correction to a message :-°

:-)

As an MCabber user, I appreciate that anyway.

hehe

Info> Plugin progress loaded

that’s akin to "ICBM armed" ✏

I don't especially dislike corrections. I do dislike the social crap around it though (and around other features that fondamentally disrupt the normal flow of discussion)

I get the impression most of the things you dislike are about the social crap around it ;)

Probably

https://op-co.de/tmp/progress.mp4

XMPP would be awesome if people wouldn't abuse it for communication.

Holger, yeah

Well.. not exactly

> XMPP would be awesome if people wouldn't ~ab~ use it ~for communication~.

I'm still gonna have to fight people on how to use other solutions..

I just store all ids

And if any match I apply the correction

Yolo message correction

Daniel: that only works if you also store all versions of the message (which is a good idea to do anyway, but not always feasible)

Yeah I'll keep that in mind for the 2023 rewrite

I'm still curious to know if <body/> is the only thing ever that can be corrected

Because that's not how I read the thing

pep.: anything can be corrected

Right so I need to keep the entire message

pep.: it depends ✏

"The receiving client then treats the newly received payloads as completely replacing all payloads of the original message."

(which also makes it work btw with e2ee)

pep.: yes

pep.: if you only store the body, you can replace the body with whatever you extract from the correction

Ge0rG, "I sent you an OMEMO encrypted message but your client doesn’t seem to support that. Find more information on https://conversations.im/omemo"

So if you are correcting a 184 you have to remove one checkmark and apply it somewhere else

Daniel: yeah

except, if somebody LMCs a 0184, you are allowed to do horrible things to them

Send them a LMC of 0184 in return?

I just correct the same 333 display marker over and over. That's how I move it down

Daniel: yes I guessed that Conversations does something different. Because my tests with Conversations worked even before I read that detail.

Daniel, jk? (just to make sure)

jubalh: yes Conversations did the old thing until fairly Recently. And I want to be compatible with myself

pep.: yes

But that'd be funny. Because that's how you could also move it up

And unread things

Daniel: that's an awesome idea. That way, mamfc will only give you the latest Marker!

interesting

Daniel, actually I like that :x

Not especially the LMC hack, but the possibility to move it up again

"Wait no I wasn't done read here actually"

(Of course I still don't want to send that to the recipient, only to myself, but most people probably don't care :))

Maybe the new Inbox answers this

or not

Daniel: whats the old thing?

Using the last ID

Not the original id

ah

Well. I would have much preferred that way

Still don't see an advantage of the "new way"

I don't either

Well think of it as fastening

You need to store all the corrections alongside the original

So far I store only the message/last correction

So it's technically possible to remove a correction right? Or would be, with fastening

On the server side I mean

also, I’m getting better at catching up with deferrals

is it me or is the static content on xmpp.net b0rked?

I'm getting 502

the entire thing seems to be overloaded at the moment

it sends 502 when that happens

so maybe it just struck you for the static contentw hile your non-static request went through

ugh, the disk was full earlier, maybe some cache got messed up

> ugh, the disk was full earlier, maybe some cache got messed up No more xeps! Save storage!

Ah, net, not org

same disk tho, afaik

Yes

And, indeed, it was xmpp.net('s backups) that caused the full disk.