Time  Nick      Message
06:26 paramat   celeron55 some of us had a discussion about forum offtopic recently and feel the section needs some organising: the 3 subforums' titles are confusing because they suggest offtopic is not allowed when the forum section title and 'forum rules and conditions' threads suggest it is. also, these 3 subforums should be merged into 1
07:08 celeron55 paramat: well they're basically a joke
07:08 celeron55 need to make a plan about what to do with offtopic
07:26 paramat   yeah i guessed they were a joke :]
07:26 paramat   some don't get it
07:27 paramat   those who discussed felt it was fine to have offtopic as long as it is firmly moderated
07:29 paramat   but of course it's up to you, it's on your resources
07:42 paramat   personally i wouldn't miss an offtopic section, but i'm sure many like to have one
07:58 paramat   #6457
07:58 ShadowBot https://github.com/minetest/minetest/issues/6457 -- Positional sound: Limit volume when closer than 1 node by paramat
08:06 nerzhul   rubenwardy, waiting for you on #6351 your points are now fixed
08:06 ShadowBot https://github.com/minetest/minetest/issues/6351 -- Implement mod communication channels by nerzhul
09:47 paramat   made some more forum reports for excessive expletives :}
11:52 paramat   ShadowNinja confirms they don't have time for MT currently, so please look at their PRs https://github.com/minetest/minetest/pulls/ShadowNinja . one is approved and just needs a rebase: #3827 two have so many conflicts they may not be worth adopting (easier to start again): #2604 #3653 please consider
11:52 ShadowBot https://github.com/minetest/minetest/issues/3827 -- Fix and clean up NCursesW finder by ShadowNinja
11:52 ShadowBot https://github.com/minetest/minetest/issues/2604 -- Add support for multiple listen addresses by ShadowNinja
11:52 ShadowBot https://github.com/minetest/minetest/issues/3653 -- [WIP] Rewrite filesystem helpers by ShadowNinja
12:55 srifqi    ~tell paramat Done repositioning, IDK what is the best layout. (^_^")\
12:55 ShadowBot srifqi: O.K.
13:46 Krock     paramat, working on the jump node problem
13:46 Krock     fixed the new sneak code... now I got to care about the old one
13:47 Krock     unless we keep that one with all its bugs
13:55 paramat   hmm
13:56 paramat   altering the old move code might change a behaviour those who use it want
13:56 Krock     funny enough this problem crosses with my plans to sync the collision detection and sneak code
13:56 paramat   i think best leave the old code alone
14:34 Krock     #6458
14:34 ShadowBot https://github.com/minetest/minetest/issues/6458 -- Localplayer: Fix disable_jump effect and getStandingNodePos() by SmallJoker
15:08 sfan5     http://sprunge.us/DhAC?diff pushing in 5min
18:00 Hijiri    reminder that #5612 is recnetly updated and ready for review
18:00 ShadowBot https://github.com/minetest/minetest/issues/5612 -- Allow overriding tool capabilities through itemstack metadata by raymoo
18:05 rdococ    ooh
18:06 Hijiri    don't get your hopes up
18:08 rdococ    why not?
18:18 Krock     Hijiri, what's the reason for using JSON?
18:20 Krock     ah, maybe because they're stored in text (serialized metadata)
18:44 Hijiri    I thought it should be readable if it's in metadata
18:44 Hijiri    but I guess the binary format would be fine too
18:49 Krock     no, JSON is a better way because it's going to be serialized again to save the metadata on the disk
19:40 Calinou   maybe use http://msgpack.org/ ?
19:40 Calinou   if you want something JSON-like, but binary and faster/smaller
19:50 Hijiri    If I was going to use a binary format I think I would just use the existing one
19:51 Hijiri    I don't think messagepack is necessary since it's just one format being consumed / produced by Minetest only
19:51 Hijiri    well, one format per protocol version at least
19:59 Krock     binary might be exactly what the metadata serialisation method might make stop working
20:01 Krock     checked it right now. Binary would work too, as a length of the saved string is defined
20:02 sfan5     binary serialization of lua tables would be nice to have
20:05 Krock     a new formspec system would also be nice to have.. but some things might have to wait for a while :3
20:54 Hijiri    thing with metadata is that to me it looked like there is no way to mark the end of the metadata section
20:54 Hijiri    so it is impossible to backwards-compatibly add more stuff to itemstack serialization
20:54 Hijiri    so everything new has to go in metadata now
20:55 Hijiri    maybe 0.5 would be a good place to allow some way to mark the end of metadata?
20:56 sfan5     0.5 would be a good point to stop serializing items as human readable strings
20:56 sfan5     or something similar
20:58 Hijiri    are itemstrings just the serialization format? I didn't realize it was the same thing.
20:59 Hijiri    serializing item metadata at least uses unprintable characters
20:59 Hijiri    also I'm wrong about nothing being addable after metadata
21:00 Hijiri    oh wait, now I see you are right
21:00 Hijiri    metadata itself serializes with non-printable characters, but it gets converted to printable
21:01 sfan5     inventories are serialized as strings too ;_;
21:02 Hijiri    is it a big source of lags?
21:02 Hijiri    the nonefficient serializatino
21:02 Hijiri    maybe the issues people mention about sending lots of metadata
21:02 Krock     sfan5, but for a good reason, when the player data is saved as plain text
21:03 sfan5     indeed