00:11:24  * akirosequit (Ping timeout: 244 seconds)
00:14:08  * akirosejoined
01:50:55  * keith_millerquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
01:52:16  * keith_millerjoined
02:03:54  * keith_millerquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
02:20:59  * akirosequit (Ping timeout: 268 seconds)
02:21:52  * akirosejoined
02:32:53  * jwaldenquit (Quit: ChatZilla 0.9.92-rdmsoft [XULRunner 35.0.1/20150122214805])
03:12:41  * keith_millerjoined
03:14:39  * keith_millerquit (Client Quit)
04:25:23  * gibson042joined
04:28:31  * akirosequit (Ping timeout: 246 seconds)
04:30:44  * akirosejoined
04:42:01  * gibson042quit (Ping timeout: 250 seconds)
04:48:26  * jwaldenjoined
05:30:50  * jwaldenquit (Quit: ChatZilla 0.9.92-rdmsoft [XULRunner 35.0.1/20150122214805])
05:40:46  * jmdyckquit (Remote host closed the connection)
06:38:32  * akirosequit (Ping timeout: 250 seconds)
06:39:07  * akirosejoined
06:50:30  * akirosequit (Quit: 👋🏻)
06:53:42  * akirosejoined
07:21:19  * keith_millerjoined
07:30:01  * keith_millerquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
07:35:06  * keith_millerjoined
07:44:42  * keith_millerquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
08:45:47  * keith_millerjoined
08:47:40  * keith_millerquit (Client Quit)
08:49:04  * keith_millerjoined
08:50:18  * keith_millerquit (Client Quit)
08:53:59  * keith_millerjoined
08:57:51  * keith_millerquit (Read error: Connection reset by peer)
09:06:04  * keith_millerjoined
10:19:07  * keith_millerquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
12:29:29  * keith_millerjoined
12:43:44  * keith_millerquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
12:46:31  * keith_millerjoined
12:51:45  * jmdyckjoined
13:26:13  * keith_millerquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
14:00:52  * keith_millerjoined
14:01:59  * keith_millerquit (Client Quit)
14:40:21  * gibson042joined
15:05:16  * keith_millerjoined
15:19:50  * gibson042quit (Ping timeout: 268 seconds)
15:34:38  * gibson042joined
16:42:28  * keith_millerquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
17:07:20  * keith_millerjoined
17:16:53  * keith_millerquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
17:18:33  * AtumTjoined
17:30:42  * keith_millerjoined
17:48:50  <devsnek>has there ever been discussion of arbitrary precision floating points
17:48:56  <devsnek>0.1f + 0.2f would be 0.3f
17:50:28  * keith_mi_joined
17:52:23  * keith_millerquit (Remote host closed the connection)
18:04:14  <leobalter>devsnek do you mean non 32bit precision floating operations?
18:04:37  <devsnek>leobalter: could be more or fewer bits
18:04:48  <devsnek>depending on the number being represented
18:04:53  <devsnek>same as how arbitrary precision ints work
18:05:42  <leobalter>I'm not sure if this would go with a new number type
18:06:21  <leobalter>these non bit operations might be heavy enough for engines, in a way implementors might now desire this feature enough to make a case
18:06:26  <leobalter>this is my guess
18:06:48  <devsnek>if you've used gmplib, its the same idea
18:06:49  <leobalter>at the same time I admit I get annoyed by 0.1 + 0.2
18:07:19  <devsnek>gmplib has an arbitrary precision for the mantissa and one "word" of precision for the exponent
18:07:32  <devsnek>usually 32, 64, or 128
18:08:38  <leobalter>yes, and the counter argument from implementors would be asking the advantage of having this natively. So far, it doesn't seem like it would be necessarily faster
18:09:00  <devsnek>i suppose you could use bigints to represent these
18:09:07  <devsnek>but then you lose using operators
18:09:07  <leobalter>maybe, and this is just a guess, we could have a minimal work on Intl to have currency operations, for addition, etc
18:09:26  <devsnek>i think it would definitely have a perf improvement to be a native feature
18:09:50  <leobalter>not really, engines mostly reuse their C++
18:10:01  <devsnek>i'm not sure what that sentence means
18:10:02  <leobalter>and the abstractions wouldn't be that amazing as well
18:10:28  <leobalter>engines are generally based on C++
18:10:32  <devsnek>right
18:10:45  <leobalter>unless you have something native from there
18:11:03  <leobalter>you would have an abstraction similar to what you can already do in JS today
18:11:30  <devsnek>you can make a bigint out of a system of reallocing Uint32Arrays
18:11:43  <devsnek>you can make a bigfloat out of a system of bigints
18:12:09  <bradleymeck>devsnek: littledan was looking at bigdecimal types last i knew
18:12:27  <devsnek>gmplib is definitely faster than anything you can write in js, they're using platform specific assembly and simd and whatnot
18:12:55  <devsnek>bradleymeck: as something for the language?
18:13:26  <bradleymeck>yea, but bigint is new. he did give a presentation on different ways to store big decimals though
18:13:41  <devsnek>definitely no shortage of ways to represent them
18:16:43  <littledan>well, Andrew Paprocki gave that presentation (though I wrote the slides)
18:17:49  <littledan>each language does it differently, and no one seems to care. My takeaway was that we should just use IEEE 128-bit decimal, since it's standard, there are libraries for it, and, just, no one complains about how all the other programming languages use observably different bespoke decimal formats, so I bet we don't actually need BigDecimal
18:18:00  <littledan>(whereas people really do run into overflows of ints)
18:18:18  <littledan>that was fun research!
18:30:48  * keith_mi_quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
18:51:46  * keith_millerjoined
19:00:41  * keith_millerquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
19:20:43  <jschoi_>Does anyone at the 2018-11-29 TC39 summary session have details regarding what got discussed during the function-call syntax discussion (https://github.com/tc39/tc39-notes/blob/master/es9/2018-11/nov-29.md#--function-calls)? TabAtkins?
19:21:33  <TabAtkins>I do but am not in a position to write them out right now, I'll ping you later
19:33:10  <jschoi_>Thanks!
19:53:09  <devsnek>littledan: if you were modifying a script to use import() why couldn't you just go all the way and make it a module and use static import
19:53:53  <devsnek>you would probably need to modify it anyway to deal with import() being async
19:55:39  <ljharb>because i have scripts that require sloppy mode and also require builtins
19:56:09  <ljharb>however, for all of those, they require a *sync* means of getting those builtins, so `import()` isn't an option anyways ¯\_(ツ)_/¯
19:57:12  <devsnek>yeah I don't really understand how import() would play with a stdlib
19:59:34  <zenparsing>the ability to dynamically import anything that can be statically imported (and in basically the same way) is a nice quality
20:02:53  <ljharb>totally agreed
20:03:25  <ljharb>that's a current gap in the quadrants right now in the language - there's no way to synchronously import a Module into a Script.
20:05:19  <devsnek>i kind of feel like its not a bad thing that scripts are being left behind
20:05:49  <devsnek>(as long as they keep working as they do now)
20:07:29  <ljharb>that's certainly not a unique attitude
20:07:44  <ljharb>and yet, `import()` works in Scripts
20:08:26  * jwaldenjoined
20:09:37  <Domenic>It's a one-way valve
20:09:45  <Domenic>Get you out of classic scripts land, into module scripts land
20:09:49  <Domenic>But you can't go back, very intentionally.
20:13:03  <ljharb>it's fine for normal dynamic import, but it presents a problem for builtin modules
20:23:10  * keith_millerjoined
20:25:34  * keith_millerquit (Client Quit)
20:39:28  <Domenic>The less stuff we can have work in classic scripts, the better, IMO.
20:39:41  <Domenic>I just wish we'd also put a secure contexts restriction in there too.
20:39:52  <Domenic>I guess we still can for built-in modules.
21:08:46  * keith_millerjoined
21:59:48  * keith_millerquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
22:02:25  * keith_millerjoined
22:15:10  * keith_millerquit (Quit: My MacBook has gone to sleep. ZZZzzz…)
22:17:03  * keith_millerjoined
22:35:11  * AtumTquit (Quit: AtumT)
22:45:40  * AtumTjoined
22:54:23  * ephemera_quit (Read error: Connection reset by peer)
22:55:05  * ephemera_joined
23:42:02  * efaustjoined
23:55:43  * AtumTquit (Quit: AtumT)