Thread 105896301 - /g/ [Archived: 266 hours ago]

Anonymous
7/14/2025, 12:44:18 AM No.105896301
1729765189318488
1729765189318488
md5: 65164e377de6c72f92559d654b2f599e🔍
dead on arrival
Replies: >>105896325 >>105896488 >>105897141 >>105897170 >>105903301 >>105903317 >>105906148 >>105907140
Anonymous
7/14/2025, 12:47:03 AM No.105896325
>>105896301 (OP)
Of all things to switch to....
Replies: >>105896580
Anonymous
7/14/2025, 12:58:53 AM No.105896436
swift-emoji
swift-emoji
md5: 8b42eecdc8ab7d0a884dcf9777824231🔍
based
Replies: >>105896570 >>105896580 >>105897737 >>105901891 >>105907140
Anonymous
7/14/2025, 1:03:59 AM No.105896488
>>105896301 (OP)
Swift is based.
Replies: >>105896580 >>105904037
Anonymous
7/14/2025, 1:12:09 AM No.105896570
>>105896436
what exactly is wrong with supporting unicode?
Replies: >>105896580 >>105900235
Anonymous
7/14/2025, 1:13:17 AM No.105896580
>>105896325
>>105896436
>>105896488
>>105896570
OP here I meant C++
Anonymous
7/14/2025, 1:32:02 AM No.105896750
>more modern than modern c++
lmao doesn't exist, nobody is going to write real code in some meme language that cripples you for some illusion of memory safety which doesn't even exist
Anonymous
7/14/2025, 1:47:25 AM No.105896886
Swift is too slow.
Replies: >>105897037
Anonymous
7/14/2025, 2:05:52 AM No.105897037
>>105896886
> swift is too slow
You can’t make a general statement like that.
I’d say it’s on par with something like ‘go’
But the language itself is very much a cleaned up C++.
As with almost all things, bottlenecks will be identified and replaced (dynamically or at build time) with assembly on some architectures.
Gcc, for example, has stdlib memcpy that is written in assembly using SIMD instructions when it detects it can do so.
Swift is using the llvm/clang optimizer, so, it’s not going to be terrible, but if you look at the from some C compilers (e.g. tcc or pcc) it will be better than that and probably faster if you use those as your C standard.
Also, it depends on how you use it. If you are writing kernel routines (for example), you won’t be using new/malloc a lot, and those things are regarded as slow due to heap maintenance, locking, and non-locality.
Replies: >>105898268
Anonymous
7/14/2025, 2:16:44 AM No.105897141
>>105896301 (OP)
Swift has some cool features but building a browser with it, Im not so sure
Anonymous
7/14/2025, 2:20:02 AM No.105897170
>>105896301 (OP)
Apple chads won AGAIN
Replies: >>105902114
Anonymous
7/14/2025, 2:31:18 AM No.105897276
C++ should only be used for GUI elements
Everything else should be C
Anonymous
7/14/2025, 3:18:47 AM No.105897737
>>105896436
>graphics in source code
An inferior version of Terry Davis' DolDoc format.
Anonymous
7/14/2025, 4:16:06 AM No.105898268
>>105897037
Yeah I thought it should be fast too until I used it and discovered that no, it's actually slow. Maybe it's the referencing counting, who knows.
Replies: >>105904053
Anonymous
7/14/2025, 4:27:39 AM No.105898341
1748451940085825
1748451940085825
md5: 66d629fe4ab5dd2a03964b19b5e55507🔍
>swift
>language even Apple can't use to make good, well, anything
what did he mean by this?
Replies: >>105899341
Anonymous
7/14/2025, 6:58:04 AM No.105899341
>>105898341
it's got to be one of the most perplexing language choices in programming history. my money is on him taking money from apple for it.
Replies: >>105902811
Anonymous
7/14/2025, 9:37:25 AM No.105900235
>>105896570
Nothing. Now, supporting emojis in symbol names, that's pure retardation.
Anonymous
7/14/2025, 1:50:04 PM No.105901889
lmao
Anonymous
7/14/2025, 1:50:37 PM No.105901891
>>105896436
Death to emojifags
Anonymous
7/14/2025, 2:23:26 PM No.105902114
>>105897170
Even Apple isn't all in on Swift, despite their claims. They still hook into Objective-C and even plain C, when they need to get real work done.
Replies: >>105904002
Anonymous
7/14/2025, 3:50:56 PM No.105902811
>>105899341
That would suck, because it would mean Ladybird isn't as independent as it seems. I don't get what's wrong with C++, seems like a much better suited language for a browser than Swift.
Anonymous
7/14/2025, 4:51:22 PM No.105903301
>>105896301 (OP)
swift was {word} but the changes have made it less so
Anonymous
7/14/2025, 4:52:59 PM No.105903317
>>105896301 (OP)
you have to understand he needs something challenging to keep him from relapse and writing a browser in swift should be sufficient
Anonymous
7/14/2025, 6:03:23 PM No.105904002
>>105902114
> write parts in C
That’s always an option, too.
Back in the NextStep days this was common in high-performance stuff, and just have objective C call the C code.
Anonymous
7/14/2025, 6:07:55 PM No.105904037
>>105896488
>Swift is based.
Swift will just Rust by detour eventually, borrow checker and all.
Anonymous
7/14/2025, 6:09:44 PM No.105904053
65538514-24D1-45E2-8328-E26E3883D4EC
65538514-24D1-45E2-8328-E26E3883D4EC
md5: 5aca53cb5b1873ed506b4758fad9d807🔍
>>105898268
It’s not good at sudoku for whatever reason
Anonymous
7/14/2025, 9:31:09 PM No.105906148
>>105896301 (OP)
Is there already a fork that keeps on using c++?
Anonymous
7/14/2025, 11:03:59 PM No.105907140
1730741486064160
1730741486064160
md5: f51d73fbe600aa7714f927252678d295🔍
>>105896301 (OP)
>>105896436
iTODDLERS BTFO
Replies: >>105909157
Anonymous
7/15/2025, 2:28:09 AM No.105909157
>>105907140
based
Anonymous
7/15/2025, 2:31:12 AM No.105909183
they're going to release the first RC as macOS only and make everyone else wait a year then the project will completely fizzle out because no one cares about it except for 5 faggots on hackernews