Hacker News: HelloNurse

New comment by HelloNurse in "CTTI is Exponential, RTTI is Linear"

HelloNurse — Thu, 03 Sep 2026 08:20:43 +0000

Behind the article's dichotomy of CTTI and RTTI there is the catalogue of concrete features (e.g. printf-like string formatting) and the respective algorithms with an exponential or linear cost in the number of types. The choice between optimized but polymorphic and lean but indirect implementations and the choice to do the work statically at compile time, dynamically at run time, or never are more independent than the article suggests, and the constraints more complex. For example, the known types at compile time that need to be handled could be all important, or an infinite (sometimes merely impractically huge) superset of the types that are actually used for a certain input, or anything in between.

New comment by HelloNurse in "Salad Theory"

HelloNurse — Wed, 02 Sep 2026 08:39:54 +0000

  >  An “untossed salad” is definitely not a salad.

I definitely agree, but grandfathering in this specific exception suggests a definition variant: a salad can be mixed and shuffled (possibly from an ordered initial state), and in this case it remains a salad; it isn't damaged, and on the contrary it usually improves (typically by increasing the entropy of solid component positions and distributing liquid condiments more uniformly).

New comment by HelloNurse in "FFmpeg 9.0"

HelloNurse — Tue, 04 Aug 2026 14:09:03 +0000

Given that it is easy to feed FFMPEG arbitrary input to exploit any vulnerability, popular codecs are very likely to be already secure enough, while unpopular ones are far more dangerous and deserve more attention. It can be assumed that an attacker would target the worst, easiest to exploit codec they can find in the whole of FFMPEG.

New comment by HelloNurse in "Introduction to Data-Oriented Design [pdf]"

HelloNurse — Tue, 28 Jul 2026 10:55:59 +0000

It is usually possible to design a representation where cows and pigs are the same thing, either Animal class instances or equally unified other data structures, and their differences are data (e.g. weight, length, one of several colour patterns).

New comment by HelloNurse in "Introduction to Data-Oriented Design [pdf]"

HelloNurse — Tue, 28 Jul 2026 07:57:56 +0000

You don't seem to grasp that handling individual entity state has a cost that has to be minimized and amortized.

A generation counter offers a good deal in many common cases: for the fixed small cost of incrementing the generation counter when an entity ID is assigned a trivial test, also fixed cost, can disambiguate successively recycled entities that share the same number and tell which one is current.

This adds up to work proportional to the number of distinct entities plus, assuming obsolete references are eliminated, work proportional to the number of references to entities, which is clearly very good. Do you recommend a different way to solve the same problem? There is at least one obvious one.

New comment by HelloNurse in "Teaching Kids Forth"

HelloNurse — Mon, 27 Jul 2026 11:55:45 +0000

> start incorporating DUP and OVER and ROT and PICK and friends

If one understands adding and consuming values at the top of the stack, putting the stack in the state you want it is certainly a small logical step without new theoretical baggage.

Word definitions and the accompanying pointer tricks are more likely to be challenging.

New comment by HelloNurse in "Introduction to Data-Oriented Design [pdf]"

HelloNurse — Mon, 27 Jul 2026 11:45:17 +0000

No amount of ideology can fill L1 cache.

New comment by HelloNurse in "Introduction to Data-Oriented Design [pdf]"

HelloNurse — Mon, 27 Jul 2026 11:43:06 +0000

I certainly don't know "what my program does" if it's a game where, at every frame of the simulation, thousands of entities might or might not be created, deleted and recycled depending on player inputs.

Can you describe a superior replacement for generation counters?

New comment by HelloNurse in "We have proof automation now"

HelloNurse — Mon, 27 Jul 2026 09:00:34 +0000

Focus on formal verification is a technique to "harden" an unreliable LLM that produces meaningless slop from arbitrary text into a compiler that produces a correct program from a concise, formal input.

"Writing formal specs" means simply programming, taking a step upward in programming language abstraction level, and whether it's going to be easy remains to be seen.

New comment by HelloNurse in "I regret migrating to Codeberg"

HelloNurse — Fri, 24 Jul 2026 12:25:25 +0000

I agree with the distinction between using a LLM as a tool to make personal contributions to healthy projects and expecting Codeberg to host arbitrary slop, but I'm afraid you are assuming good faith.

The "developer" could be a bot, or a tool-assisted criminal or activist, and they might be using LLMs to enhance their ability to add stealthy malware, plant patent or copyright violations, impose not quite shared technical choices, and so on.

Many open source projects ban AI contributions just because they are an efficiently low effort way to waste the time of the actual maintainers.

New comment by HelloNurse in "I regret migrating to Codeberg"

HelloNurse — Fri, 24 Jul 2026 07:17:49 +0000

  >  It seems like many ‘vibe coders’ don’t realize that they don’t actually have a community around them.

This is not about demanding a community, as the article argues about at great length: it's only a polite way to say that many ‘vibe coders’ are annoying, negative net value assholes. Community would ensure that they are not aberrant lone wolves.

New comment by HelloNurse in "Alternative(s) to run CUDA on non-Nvidia hardware"

HelloNurse — Tue, 14 Jul 2026 15:58:57 +0000

Being fast and being as easy to program as CUDA are two different things.

New comment by HelloNurse in "Show HN: Beautiful Type Erasure with C++26 Reflection"

HelloNurse — Tue, 14 Jul 2026 15:51:52 +0000

C++ is in the transitional phase where major bad things are being taken away very satisfactorily, usually by providing a simpler and more general replacement (for example, auto instead of long and pointless type declarations or modern initialization protecting against implicit conversions and surprise constructor overloads), but most progress of elegance and grace come from new features that enable something traditionally terrible or impossible (for example the gradual generalization of templates, culminating with concepts, the gradual extension of constexpr, consteval etc, and the new reflection).

New comment by HelloNurse in "Why American ambulance rides are so expensive"

HelloNurse — Fri, 10 Jul 2026 12:20:12 +0000

Unable to be willing. The combination of not knowing any better, political parties and media that don't care or exert a contrary pressure because they belong to billionaires, and powerful health sector businesses appears to be very effective.

New comment by HelloNurse in "What ORMs have taught me: just learn SQL (2014)"

HelloNurse — Sat, 04 Jul 2026 16:03:00 +0000

They don't consider the ORM the second class citizen it actually is: an optional simplified alternative to normal queries, that can be used for the easy cases.

New comment by HelloNurse in "Half-Baked Product"

HelloNurse — Fri, 03 Jul 2026 15:25:09 +0000

One of the points is that, for a startup, not flailing around is very difficult and unlikely: these patterns of failure and "classic bullshit" are the universal norm. This piece "articulates a point of view or personality that you had never before considered": that you and your startup are no better than the characters in the tale, and equally doomed.

New comment by HelloNurse in "Half-Baked Product"

HelloNurse — Fri, 03 Jul 2026 15:04:35 +0000

It can be accurately simulated with appropriate brushes and rags, water and soap, and a sufficient number of hands.

New comment by HelloNurse in "Half-Baked Product"

HelloNurse — Fri, 03 Jul 2026 15:00:53 +0000

Considerable arrogance is required to think you can improve on mature products enough to conquer a large market share. Such arrogance should be supported by having, if not a demonstrably fantastic prototype, at least an obviously good idea; otherwise you are following the example of Juicero or Theranos.

In the article, the "smart" oven is only a speculation (maybe it works, and maybe someone will pay for it) and as such it is appropriate as a relatively low effort and low risk experiment on the part of an established oven maker (develop rudimentary automation and offer it as a very mildly disruptive feature at a modest price increase).

New comment by HelloNurse in "Half-Baked Product"

HelloNurse — Fri, 03 Jul 2026 09:37:31 +0000

Brilliant autobiography.

New comment by HelloNurse in "Solid and Clean Code never felt solid or clean to me"

HelloNurse — Wed, 01 Jul 2026 19:21:43 +0000

On the contrary, the new trend of telling a bot to write code automates disregard for architecture.