Ouch! If accurate maybe this is why the compatibility stats have not been published?
Ryan Dahl@rough__sea
deno 2.8 ships tomorrow. bun is definitely faster in many benchmarks - credit where due - but node compat isn't close
English
James Moger
917 posts
James Moger
@JamesMoger
Husband, Father, Guitar Hero, JavaScript student, .NET Dev, Javavore, Linux Enthusiast, Director of Engineering, Cornell alum, Can divide by zero.
Queensbury, NY Katılım Temmuz 2013
842 Takip Edilen160 Takipçiler
Bun 1.3.14 release notes are HUGE! So impressive!🎉
There was an idea many months ago that there would be quantified reporting with each release on progress towards this goal, but the compatibility page looks "the same" as it did last year.🤔
bun.com/blog/bun-v1.3.…
Jarred Sumner@jarredsumner
Bun’s Node compatibility goal: get all of Node’s test suite to pass
English
@dhh @m0nkeypatch Nautilus is out of place in the Omarchy aesthetic. Is there any solid alternative?
English
@jarredsumner @GergelyOrosz Hasn’t there been exploration to use containers?
English
@GergelyOrosz More precisely: the ideal is something like
- complete filesystem CoW clone
- per-host network filtering. per-http request filtering, read-only write locally permissioning by default
- automatic process / etc cleanup
Linux is easiest. But still many, many edgecases.
English
AI agents are far more cable when they have full system access; but when they do, they can mess a lot of stuff up (not unique to any one model). AI harnesses have guardrails: but those can fail.
I wonder if we’ll need OS-level “sandbox primitives” to deal with this better?
Elliot Arledge@elliotarledge
just woke up to opus 4.7 nuking one of my projects during an overnight session. luckily i was able to get it back easily
English
Finished HBO’s GoT today. I’ve also read the books. I don’t think the series ended poorly. I do feel sad about the decisions some characters made but not upset to conclude the eighth season is trash. Looking forward to Martin finishing the written storyline. Could even happen. 🤞
English
@nmstoker @willmcgugan This. Toad demonstrates some very cool Textual usage that archinstall could learn from. Very cool nonetheless.
English
@willmcgugan Nice!
I do get the KISS rationale but feels like they could take it quite a bit further on the aesthetic front without much effort 😉
English
Secure by default should not have exceptions. Drop the top 500 whitelist.
Bun@bunjavascript
`bun install` blocks postinstall scripts by default. This helps protect from supply-chain attacks, like today’s axios incident.
English
James Moger retweetledi
@jarredsumner As another has noted, high performance csv parser would be handy. JWT is another.
English
what large features should I run claude in a loop to try adding to bun
constraints:
- must have a large test suite with positive & negative tests that verify it works
- must have strong evidence of being something people want & would use bun for
English
I’m going to need to write a “How Toad works” document. Too many people assuming that Toad wraps every agent CLI.
Can’t blame them I guess. Everything is so new here.
English
@JoshBodnar1 @wesbos Same experience here. Love the monitor, have had trouble with getting the settings right. Doubt I’m running with correct settings.
English
@use_bruno Proposed. But I still don’t like sequence stored the in request files.
English
I always preferred towers. Amazing this still boots!
Floppy Deep Dive@FloppyDeepDive
This isn't DOSBox. This is Prince of Persia on a real 486 from 1993. No Sound Blaster. Just PC speaker. But when your graphics looked THIS good... it was enough. Real hardware. Real rotoscoping. BOOM! #retrogaming #princeofpersia #dosgaming
English
This is a good question, how do you do SSR with Bun? Let me show a more complete sample here
Mateusz Suchoń@suchy
@cirospaciari btw, how you render jsx on server an return it as html response?
English
1998. Purchased RedHat, boxed, from Electronics Boutique.
sysxplore@sysxplore
What year did you start using Linux? Be honest.
English
James Moger retweetledi
@bunjavascript Wasn’t there a proposal to track percentage of test coverage compatibility on the website for all modules? I know it keeps getting better and it is tracked in some, but this is measurable and notably absent on many described as “fully implemented”.
English