Re: [Standards] Persisting Message Errors (XEP-0280, XEP-0313, XEP-0160)

2019-09-18 Thread JC Brand
On Thu, Aug 01, 2019 at 01:08:49PM +0200, Georg Lukas wrote: > Hi, > > error type stanzas are currently ephemeral, and not taken seriously by > many (client) developers. As one step in increasing the (perceived) > reliability of XMPP messaging, I'd like to make message errors > persistent, so

Re: [Standards] Persisting Message Errors (XEP-0280, XEP-0313, XEP-0160)

2019-08-01 Thread Kevin Smith
> On 1 Aug 2019, at 19:42, Ненахов Андрей > wrote: > > I imagine that a proper solution to this problem might be something > like... reactions? If we do reactions with some kind of attached > message, errors might be a special kind of attachments. Thinking > further, I like it even more:

Re: [Standards] Persisting Message Errors (XEP-0280, XEP-0313, XEP-0160)

2019-08-01 Thread Ненахов Андрей
Our reliable message delivery protocol that we use to replace stream management makes half of message errors mostly unnecessary: we consider that message is delivered to server only when we receive confirmation with server ID and timestamp assigned to this message. If you don't have server ID, you

Re: [Standards] Persisting Message Errors (XEP-0280, XEP-0313, XEP-0160)

2019-08-01 Thread Jonas Schäfer
On Donnerstag, 1. August 2019 13:08:49 CEST Georg Lukas wrote: > Obviously, this should only happen for direct messages and PMs, not for > type=groupchat ones. Should this have been directly below point 4? Otherwise I’m not sure how to interpret type=groupchat in the context of error messages.

[Standards] Persisting Message Errors (XEP-0280, XEP-0313, XEP-0160)

2019-08-01 Thread Georg Lukas
Hi, error type stanzas are currently ephemeral, and not taken seriously by many (client) developers. As one step in increasing the (perceived) reliability of XMPP messaging, I'd like to make message errors persistent, so that users can better gauge which of their messages actually arrived at the