Sender | Message | Time |
---|---|---|
17 Aug 2019 | ||
Matthew | on the sync/push stuff; please please understand that matrixβs sync stuff is not intrinsically long polling at all - thatβs just the crappy default http transport | 22:01:33 |
Matthew | https://github.com/matrix-org/matrix-doc/blob/master/drafts/websockets.rst is a proposal and implementation for mapping it to WS for instance and using true push rather than long poll | 22:02:28 |
Matthew | but the coap stuff obsolete that now | 22:02:35 |
Mo π° | In reply to @matthew:matrix.orgI guess I understand your point. http is currently the most pragmatic solution that just works for most clients/servers/platforms. | 22:06:32 |
Mo π° | In reply to @matthew:matrix.orgoh, didn't know that. i will compare it with my current push implementation. i might find something that helps me to improve my implementation π€ | 22:08:10 |
Mo π° | In reply to @matthew:matrix.orgi really have to find out what coap is π sound like a silver bullet. since it's now midnight where I live, i'm going to watch the coap video tomorrow :) | 22:11:43 |
Mo π° | i also have to go to bed now π΄ .... good night π | 22:12:44 |
Matthew | In reply to @forgemo:mos.appprecisely. the only reason we use it is to make it easy to write clients, and because alt transports are immature | 22:13:05 |
Matthew | night! | 22:13:23 |
Matthew | In reply to @forgemo:mos.appthat proposal maps the whole CS API to websockets, so it won't be very efficient for your purpose - it'd need the "only send me events for pushes!" extension to work as you want | 22:15:57 |
Matthew | (plus as you can tell, i think you should use something like CoAP or possibly QUIC rather than ye olde websockets for push :) | 22:16:14 |
ππ joined the room. | 23:11:00 | |
18 Aug 2019 | ||
@progserega:rsprim.ru | In reply to @forgemo:mos.appThank you very match! | 00:38:51 |
@svenvnl:matrix.org joined the room. | 08:43:14 | |
* Matthew pleads for a s/martix/matrix/g on sputnik-messenger.com πΌ | 12:17:54 | |
@tulir:maunium.net | register martix.org? :D | 12:26:23 |
Mo π° | You are killing me πππππ i will rework the whole site later today. .... It's not up to date. I also have to explain how the white label concept is supposed to work. | 12:49:51 |
ππ | Will there be fdroid releases? | 15:14:25 |
ππ | I don't have a google account and can't use playstore | 15:14:58 |
Nico | Als Antwort auf @jomat:asra.grYou could temporarily use https://send.firefox.com/download/029e4ce870d453df/#6YQG_mPrcgGfDMvuxwIZQw . Mo posted that link in Matrix HQ yesterday | 16:10:48 |
ππ | thx! | 17:14:32 |
Mo π° |
In reply to @jomat:asra.grI don't know much about fdroid and what exactly is required to release an app. It requires an app to be FOSS as far a si know ...sputniks code is about 99% Foss. the code base is designed to be a white label app where anybody can build a custom branded version of it. choosing name, colors, theme, artwork and so on. The sputnik messenger I'm releasing on the play store is also using this white label code as a basis ... configuring it with my "branding", connecting to my push server and so on. that outer shell I'm using with all branding related stuff is not open source, though. .... I'm not sure if such an app is allowed on fdroid. I will have to look into it. reason is: I want to make it difficult to build a 1:1 copy of my themed / branded / marketed version, but allow anyone interested to make his own version without any limitations. | 17:22:49 |
ππ | marketed version? what? o.O | 17:54:09 |
Nico | Alternatively you could host your own fdroid repo, which wouldn't have those requirements but is some effort to set it up | 18:48:42 |
ππ | i know, i could, i think i'll abstain from sputnik in the meantime ;-) | 20:59:24 |
Nico | Als Antwort auf @jomat:asra.grThat was more meant as a reply to Mo | 21:24:42 |
@tulir:maunium.net | In reply to @jomat:asra.grthe app is basically lots of modules + a kind of wrapper that contains branding/theme/etc. The default wrapper is proprietary, but there's an example of a whitelabel/unbranded wrapper in the github org | 21:28:06 |
Matthew | In reply to @tulir:maunium.net(i guess you don't know the story of martix.org then - troll extraordinaire: http://web.archive.org/web/20190130000752/http://martix.org/ ) | 21:35:40 |
Matthew | which tragically has gone down after... almost 5 years | 21:36:39 |
Matthew | and meanwhile caused no end of pain. | 21:36:44 |