b0a04gl
Re:OpenFLOW – Quickly make beautiful infrastructure diagrams local to ...
on Tuesday July 01, 2025 @10:39AM (#44434340)
what if we can make these diagrams synchronized with reality. you need the diagram to pull from the same source of truth as your actual infrastructure - whether that's terraform state, kubernetes manifests, or service discovery. that way diagrams become less historical artifacts and more of living documentation
Re:Show HN: Spegel, a Terminal Browser That Uses LLMs to Rewrite Webpages
on Tuesday July 01, 2025 @10:31AM (#44434262)
this is another layer of abstraction on top of an already broken system. you're running html through an llm to get markdown that gets rendered in a terminal browser. that's like... three format conversions just to read text. the original web had simple html that was readable in any terminal browser already. now they arent designed as documents anymore but rather designed as applications that happen to deliver some content as a side effect
Re:Claude Code now supports Hooks
on Tuesday July 01, 2025 @04:23AM (#44431774)
>before this you had to trust that claude would follow your readme instructions about running linters or tests. hit and miss at best. now its deterministic. pre hook blocks bad actions post hook validates results.
>hooks let you build workflows where multiple agents can hand off work safely. one agent writes code another reviews it another deploys it. each step gated by verification hooks.
Re:The new skill in AI is not prompting, it's context engineering
on Tuesday July 01, 2025 @12:57AM (#44430718)
imo i feel it's just reinventing database design principles but for llms. normalisation denormalisation indexing retrieval. same concepts different target. its more of just being good at organising information which is what we should have been doing all along.
Re:Proton joins suit against Apple for predatory practices
on Tuesday July 01, 2025 @12:54AM (#44430701)
every developer knows safari is the new ie6 but we all just shrug and build native apps anyway because what else are you gonna do. leave 50% of your users on the table. classic embrace extend extinguish but in reverse. embrace web standards then purposely not extend them so you can extinguish competition
Re:Rust CLIs with Clap
on Tuesday July 01, 2025 @12:50AM (#44430680)
10kloc for command line parsing. TEN THOUSAND LINES. pico-args does it in 700 lines and probably handles 99% of real world use cases. compile times go to shit binary size bloats and for some edge case you'll never hit.most CLI tools need what three four flags max, maybe a subcommand or two. you don't need the swiss army knife of argument parsing for that. tried replacing clap with pico-args on three different projects last month. 80% reduction in compile time every single time. binary went from 8mb to 2mb on one of them.the "disk space is cheap" argument's acceptable partially but compile time isn't. developer experience isn't. startup time isn't. memory usage isn't
Re:The new skill in AI is not prompting, it's context engineering
on Monday June 30, 2025 @11:13PM (#44430231)
https://blog.langchain.com/the-rise-of-context-engineering/?...
I feel op' s blog is more of duplicate of the above langchain's blog happened a week ago.
Re:Asynchronous Error Handling Is Hard
on Monday June 30, 2025 @02:44PM (#44426523)
in async code ,errors belong to the task ,not the caller.
in sync code ,the caller owns the stack ,so it makes sense they own the error. but async splits that. now each async function runs like a background job. that job should handle its own failure =retry ,fallback ,log because the caller usually cant do much anyway.
write async blocks like isolated tasks. contain errors inside unless the caller has a real decision to make. global error handler picks up the rest
Re:Reverse Engineering Vercel's BotID
on Monday June 30, 2025 @02:40PM (#44426489)
why is bot detection even happening at render time instead of request time. why can't tell you’re a bot from your headers, UA, IP, TLS fingerprint. imo making it a surveillance. 'you're a bot, ok not just go away, let’s fingerprint your GPU and assign you a behavioral risk score anyway'
Re:Gridfinity: The modular, open-source grid storage system
on Monday June 30, 2025 @02:34PM (#44426432)
gridfinity makes repeatability physical. can dial in a layout that works , can clone it across benches ,rooms ,even teams. becomes a shared spatial logic. that alone cuts setup time and mental load more than any new tool
Re:There Are No New Ideas in AI Only New Datasets
on Monday June 30, 2025 @02:31PM (#44426401)
if datasets are the new codebases ,then the real IP can be dataset version control. how you fork ,diff ,merge and audit datasets like code. every team says 'we trained on 10B tokens' but what if we can answer 'which 5M tokens made reasoning better', 'which 100k made it worse'. then we can start being targeted leverage
Re:I write type-safe generic data structures in C
on Monday June 30, 2025 @02:27PM (#44426369)
what happens if two types have same size and alignment but different semantics : like `int` vs `float` or `struct { int a; }` vs `int`? does the type tag system catch accidental reuse . aka defending against structual collisions
Re:Entry-level jobs down by a third since launch of ChatGPT
on Monday June 30, 2025 @11:06AM (#44424263)
i always felt an analogy of entry level jobs being a distributed human pretraining layer. they were loss functions. company burns time + money so humans can converge. but then now who's will be the next generations pretraining. youtube or bootcamps
Re:The provenance memory model for C
on Monday June 30, 2025 @11:02AM (#44424206)
provenance model basically turns memory back into a typed value. finally malloc wont just be a dumb number generator, it'll act more like a capability issuer. and access is not 'is this address in range' anymore, but “does this pointer have valid provenance”. way more deterministic, decouples gcc -wall
Re:Show HN: New Ensō – first public beta
on Monday June 30, 2025 @10:51AM (#44424083)
imo it's more of a thinking constraint journaling tbh.. friction like edit lock, coffee noise, fullscreen etc just makes me stop editing while thinking lol not letting me kill the raw draft midway.. more tools shud do this subtractive ux
Re:Show HN: TokenDagger – A tokenizer faster than OpenAI's Tiktoken
on Monday June 30, 2025 @10:44AM (#44423997)
if dagger builds a byte level DFA for special tokens and resolves overlaps via longest match, how does it handle inputs with partial matches at chunk boundaries, say a stream ends mid token like <|endo , does it buffer forward or require lookahead
Re:NativeJIT: A C++ expression –> x64 JIT
on Monday June 30, 2025 @01:03AM (#44419598)
how deterministic is the emit really. if i feed same expression tree twice,same node layout same captures. do i get exact same bytes out every time (ignoring reloc) or not. if output produced is byte stable across runs for same input graph ,that opens up memoized JIT paths.worth checking if current impl already does this or needs a pass to normalise alloc order
Re:What LLMs Know About Their Users
on Monday June 30, 2025 @12:55AM (#44419537)
models often pick up more than that,to the finest trait like interaction rhythm can bias outputs subtly ,so two users giving same prompt might still get slightly different completions. google's user‑llm paper breaks this down well,they show how user history gets converted into embeddings and fed back via cross attention ,so even if you dont pass prior chats or preferences explicitly,the model still knows you in a structured way
Re:Performance Debugging with LLVM-mca: Simulating the CPU
on Sunday June 29, 2025 @12:32PM (#44414343)
llvm-mca's always was one of those tools i bookmark but never touch, this post finally made it feel usable, seeing uop breakdowns and bottlenecks right in the cli was super clarifying
Re:Implementing fast TCP fingerprinting with eBPF
on Sunday June 29, 2025 @12:29PM (#44414320)
why do fingerprinting always happens right at connection start ,usually gives clean metadata during tcp syn. but what is it for components like static proxies or load balancers or mobile networks ,all of these can shift stack behavior midstream. this can make this activity itself a obsolete
Re:Bloom Filters by Example
on Sunday June 29, 2025 @12:21PM (#44414247)
i got into bloom filters while debugging cassandra read spikes ,lot of sstable lookups even when key not exist ,didnt make sense at first ,then realised bloom filter on each sstable meant to skip disk ,but default fp rate was high like 0.1 or so ,too much for our case ,most reads were cache miss anyway so those false positives were killing us ,changed it to 0.01 ,bit more memory it consumed but way less useless reads ,lbrought p99 read latency by good 16-18%
Re:Show HN: Octelium – FOSS Alternative to Teleport, Cloudflare, Tails...
on Sunday June 29, 2025 @12:13PM (#44414205)
what if this wasnt something you add after infra but the checkpoint you start with. right now you spin up a vm or db then wrap vpn or firewall around it. but imagine writing access rules first in way : 'team ml can hit service x' or 'web app can hit this backend' and the system wires infra from that.. infra becomes a side effect of access intent. access isnt something you cant guard always( as things move fast, breaks fast), it's may become seed where you can design with.
Re:I made my VM think it has a CPU fan
on Sunday June 29, 2025 @12:03PM (#44414140)
Mitre ATT&CK's T1497.001 (VM Detection) lists SMBIOS checks as a known vector means its open for injection anyways.
i did one little expirement on faking VM's powersupply. done it with 'HotReplaceable=Yes' and 'Status=OK', and you suddenly look like a $5k baremetal server.
cmd used
pip install dmigen dmigen -o smbios.bin \ --type0 vendor="American Megatrends",version="F.1" \ --type1 manufacturer="Dell Inc.",product="PowerEdge T630" \ --type39 name="PSU1",location="Bay 1",status=3,hotreplaceable=1
Re:The Perils of 'Design Thinking'
on Sunday June 29, 2025 @08:07AM (#44412437)
it assumes the system is already asking the right question. you walk in with sticky notes ,empathy maps ,whatever. but the brief's scoped wrong from the start. you're figuring out how to make the form smoother ,not whether the form even needs to exist. so you end up optimizing bad premises. no amount of ideation fixes that
Re:Solving `UK Passport Application` with Haskell
on Sunday June 29, 2025 @08:01AM (#44412398)
i was questioing why haskell in first place and then realised we can flip the whole model and treat the form as a constraint system where the user defines the outcome they want and the system solves for valid inputs so instead of users guessing what fields to fill the system already knows what combinations are legal this only works when your logic is deterministic and declarative which haskell supports natively, this may be one of the reason but suits well for this usecase
Re:More on Apple's Trust-Eroding 'F1 the Movie' Wallet Ad
on Sunday June 29, 2025 @07:49AM (#44412338)
because it's one of the only apps i've not disabled notifications for. you could've muted already tv+ ,mail ,stocks ,news ,fitness ,whatever. but not wallet cuz you leave that on for flights ,covid passes ,payments ,and some legit actions.
Re:The Unsustainability of Moore's Law
on Sunday June 29, 2025 @07:08AM (#44412124)
the law delivered enough headroom that systems moved on. once compute got cheap to rent and scale ,there was less pressure to push frequency or density every cycle. so focus shifted. the gains kept coming ,just not in the same shape.
Re:Using the internet without IPv4 connectivity
on Sunday June 29, 2025 @07:00AM (#44412087)
ipv6 only machine still reaches ipv4 sites because dns64 upstream is just faking AAAA records ,makes it look like everything is native ipv6. this part of the trick is happening somewhere else which's not controllable. if dns64 breaks or stops doing the mapping properly then this might break
Re:JavaScript Trademark Update
on Sunday June 29, 2025 @01:02AM (#44410487)
oracle holding javascript just froze it ,no one else can touch it ,they not doing anything with it but that nothing is exactly what kept it stable without fights or rebrands or ai company trying to trademark over it. later then it can also open lane for anyone to grab and next holder might not be this quiet. oracle doing nothing ended up protecting it more imo
Re:Sirius: A GPU-native SQL engine
on Saturday June 28, 2025 @02:57PM (#44407195)
query plans are tightly coupled to the engine that emits them : cost models, memory layout, parallelism strategy, codegen behavior all vary. substrait enables structural portability, but the actual execution efficiency depends on engine specific rewrites. a plan optimized in duckdb might underperform in sirius unless it's reshaped. how this handled for now?