00:46:24  * Fishrock123joined
00:49:39  * Fishrock123quit (Remote host closed the connection)
00:50:16  * Fishrock123joined
00:50:24  * Fishrock123quit (Remote host closed the connection)
00:54:26  * AtumT_joined
00:55:04  * AtumTquit (Ping timeout: 260 seconds)
01:00:58  * Fishrock123joined
01:03:53  * jaseWquit (Quit: Leaving)
01:05:33  * Fishrock123quit (Remote host closed the connection)
01:06:08  * AtumTjoined
01:06:15  * Fishrock123joined
01:06:21  * Fishrock123quit (Remote host closed the connection)
01:06:58  * AtumT_quit (Ping timeout: 264 seconds)
01:07:03  * Fishrock123joined
01:07:10  * Fishrock123quit (Remote host closed the connection)
01:08:20  * Fishrock123joined
01:11:51  * AtumTquit (Remote host closed the connection)
01:13:08  * Fishrock123quit (Ping timeout: 260 seconds)
02:16:52  * keith_millerquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
03:08:06  * Fishrock123joined
03:10:18  * Fishrock123quit (Client Quit)
04:32:33  * jmdyckquit (Quit: Leaving.)
04:34:14  * srl295quit (Quit: Connection closed for inactivity)
05:21:00  * jwaldenquit (Quit: ChatZilla 0.9.92-rdmsoft [XULRunner 35.0.1/20150122214805])
05:34:11  * howdoijoined
07:02:49  * keith_millerjoined
07:25:59  * isiahmeadowsquit (Quit: isiahmeadows)
07:39:46  * gibson042quit (Quit: Leaving.)
07:47:02  * keith_millerquit (Remote host closed the connection)
07:49:04  * keith_millerjoined
08:26:18  * keith_millerquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
08:52:11  * keith_millerjoined
09:34:11  * howdoiquit (Quit: Connection closed for inactivity)
10:22:25  * kosamariquit (Quit: Connection closed for inactivity)
10:25:11  * mylesborinsquit (Quit: farewell for now)
10:25:42  * mylesborinsjoined
10:34:27  * not-an-aardvarkquit (Quit: Connection closed for inactivity)
10:35:30  * AtumTjoined
10:54:13  * keith_millerquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
11:34:08  * Jayfluxjoined
11:37:58  * keith_millerjoined
11:59:20  * Jayfluxquit (Remote host closed the connection)
12:11:45  * jmdyckjoined
12:45:29  * Jayflux_joined
12:45:40  * Jayflux_changed nick to Jayflux
12:45:40  * Jayfluxquit (Changing host)
12:45:40  * Jayfluxjoined
12:53:48  * Jayfluxchanged nick to JaseW
12:56:07  * JaseWchanged nick to Jayflux
12:56:39  * Jayfluxchanged nick to JaseW
13:01:53  * JaseWquit (Remote host closed the connection)
13:03:07  * bradleymeckjoined
13:18:43  * kosamarijoined
14:29:40  * JaseWjoined
14:31:06  * JaseWchanged nick to Jayflux
14:37:40  * Jayfluxchanged nick to JaseW
14:59:10  * JaseWquit (Ping timeout: 264 seconds)
16:07:28  * bradleymeckquit (Quit: bradleymeck)
16:08:37  * not-an-aardvarkjoined
16:11:23  * bradleymeckjoined
16:26:25  * bradleymeckquit (Quit: bradleymeck)
16:30:11  * bradleymeckjoined
16:32:54  * jwaldenjoined
16:33:27  * bradleymeckquit (Client Quit)
16:38:05  * bradleymeckjoined
16:47:55  * keith_millerquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
17:24:58  * bradleymeckquit (Quit: bradleymeck)
17:28:24  * bradleymeckjoined
17:38:34  <ljharb>bradleymeck: wait why would we want it to be per-realm? i'm pretty sure i'd want it to be cross-realm
17:39:27  <bradleymeck>ljharb: aklein was concerned about when we could GC elements of the "trie"
17:40:00  <ljharb>i mean, that makes sense on a symbol vs an object - but why would cross-realm affect GC?
17:41:12  <bradleymeck>for cross realm you can remove entire sections of idempotentcy once the realm dies, even if objects from the realm persist
17:41:23  <bradleymeck>for per realm*
17:41:59  <ljharb>ok - how is that changed if the composite key is cross-realm?
17:42:19  <ljharb>wouldn't it just mean that keys for "objects from the dead realm that persist" also persist?
17:46:19  * bradleymeckquit (Quit: bradleymeck)
17:49:08  * bradleymeckjoined
19:28:01  <bradleymeck>ljharb: sorry on late reply, cross realm means that if *any* object from another realm survives you cannot detroy keys associated with it
19:28:15  <bradleymeck>destroy*
19:28:48  <bradleymeck>still, I'd avoid making keys with components from multiple realms always
19:56:09  <ljharb>bradleymeck: hmm - obviously if any object(s) which were used to create a given key survive, then that key survives too - this should be true regardless of what realm the key is created in - the only difference i can see between `Realm1.compositeKey(x, y)` and `Realm2.compositeKey(x, y)` being `===` (cross-realm) or `!==` (per-realm), where `x === y` ofc, is that when Realm1 *and* its `compositeKey` method are collected, then realm1's key would
19:56:09  <ljharb>become collectible even if `x` and `y` survive. is collecting realm1's key when realm2's key survives so important?
19:56:58  <bradleymeck>if .compositeKey is collected from one realm it can remove all paths it was holding if it is per realm, even if objects survive
19:57:37  <bradleymeck>so you could have objects that have keys, but you can GC keys associated with objects if the realm with the path to the key dies
19:57:52  <ljharb>is there a lot of concern about holding on to those keys?
19:58:38  <ljharb>i guess i'm skeptical that there will be many situations with a) multiple realms, b) key-composed objects passed outside a realm, c) that realm dies but all of the objects survive
19:59:44  <bradleymeck>idk, seems kind of odd to be mixing realms right now but :shrug: no clear winner of either approach I think
20:02:52  <bradleymeck>with per realm you get better granularity for GC, but means multiple backing stores for compositeKey which means you might care about which instance of it you call
20:17:42  * keith_millerjoined
20:17:44  * keith_millerquit (Remote host closed the connection)
20:18:26  * keith_millerjoined
20:28:56  <devsnek>if a value in a composite key is the last remaining thing from a realm just keep the realm around, its still being used
20:35:02  * Jayfluxjoined
20:45:30  <bradleymeck>devsnek: composite keys have null prototypes and are frozen, they don't define a way to keep their realm alive since they don't have any direct paths back to the realm
20:45:55  <bradleymeck>the path is reaped once any component of the path is reaped
20:50:01  * isiahmeadowsjoined
20:51:55  <devsnek>ic
22:12:21  * isiahmeadowsquit (Ping timeout: 264 seconds)
22:14:39  * isiahmeadowsjoined
22:19:15  * bradleymeckquit (Quit: bradleymeck)
22:20:41  * isiahmeadowsquit (Quit: isiahmeadows)
22:32:00  * bradleymeckjoined
22:32:11  * isiahmeadowsjoined
22:34:57  * keith_millerquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
23:01:20  * keith_millerjoined
23:40:09  * Jayfluxquit (Quit: Leaving)