Hacker News: fpoling

New comment by fpoling in "Quick impressions: A week of using Codex more than Claude"

fpoling — Fri, 21 Aug 2026 20:56:11 +0000

If one needs to dig history, then git log -p and similar are way more useful and reflect the real history rather than LLM inference about it. Comments reflecting the goal are more useful as neither history nor code itself covers it.

New comment by fpoling in "Judge Rejects Google's Attempt to DMCA Its Way Out of Being Scraped"

fpoling — Mon, 27 Jul 2026 23:49:33 +0000

Index is just a part of the model. The nice property is that the index for LLM does not need to be updated often so LLM plus index can be static. This even works for the latest news as LLM can fetch major news sites to learn the latest stories and then fetch individual articles for the final answer.

New comment by fpoling in "Judge Rejects Google's Attempt to DMCA Its Way Out of Being Scraped"

fpoling — Mon, 27 Jul 2026 22:02:21 +0000

The model just asked the crawler to get a recent copy of relevant pages and used that to give the answer. That is why AI companies experimenting with browsers or at least agentic extensions to browsers as that allows to fetch on behalf of the user directly from the device with a residential IP and ditch expensive to maintain crawling infrastructure.

New comment by fpoling in "Stinkpot: SQLite-backed shell history"

fpoling — Sun, 26 Jul 2026 15:13:48 +0000

With SQLite one automatically gets reliable persistent storage. Try to emulate that with a text file with history that eliminates duplicates. It is surprisingly hard to archive.

New comment by fpoling in "Stinkpot: SQLite-backed shell history"

fpoling — Sun, 26 Jul 2026 08:38:17 +0000

SQLite gives indexes that makes searching and locating previous entries much faster. This is relevant when the history is over 100K entries and one wants to avoid duplicates in history files.

New comment by fpoling in "Hyprland 0.55 announced the switch to Lua for its config files"

fpoling — Mon, 20 Jul 2026 21:40:50 +0000

GN tool that is used in Chromium to generate ninja files has much more restrictive language than CMake or make. And even its more complex features are mostly workarounds or escape hatches to deal with legacy setups.

New comment by fpoling in "Almost Always Unsigned"

fpoling — Thu, 09 Jul 2026 21:57:52 +0000

The article over-downplays the need for sentinel values. Surely Rust has Option that also spreads into C++ these days as std::optional. But for plain C using -1 or negative values to denote sentinels or error code is rather nice idiom.

The argument will be more valid if array indexes will be 1-based like in Fortran/Matlab/Julia as then 0 becomes extremely nice sentinel values. But C is C and needs -1.

New comment by fpoling in "Road to Elm 1.0"

fpoling — Mon, 06 Jul 2026 22:15:58 +0000

Yet Elm due to its simplicity does not need a lot of examples. Language itself is small and blessed way to architecture things consist of few patterns. Then things are composable. This is ideal for LLMs.

The price for that is verbosity. What in other languages are hidden behind async and complicated libraries in Elm has to be done explicitly resulting in boilerplate. But that is never a problem for LLMs.

Then Elm compiler has one of the best error messages. This helps agents a lot.

Then Time Machine debugger is another big help for agents as they have access to the full execution trace of the program. The trace is much smaller than would be necessary for other languages and typically fits into content window.

New comment by fpoling in "PostgreSQL and the OOM killer: Why we use strict memory overcommit"

fpoling — Fri, 03 Jul 2026 19:03:00 +0000

Java allowed to handle out-of-memory rather well if one wanted to even 25 years ago. Basically one allocated a buffer on a startup taking 5% of memory that the application was supposed to use and made all threads to catch the oom exception. When handling that the buffer would be released, GC would be forced and a special flag would be set asking app to cancel any memory-intensive tasks until enough memory would be released and the buffer can be allocated again. It worked extremely well.

New comment by fpoling in "PostgreSQL and the OOM killer: Why we use strict memory overcommit"

fpoling — Fri, 03 Jul 2026 18:48:47 +0000

Microsoft just followed VAX/VMS that does not overcommit. And there is a noise on Linux mail lists to implement process builder pattern which VAX had like 50 years ago…

New comment by fpoling in "PostgreSQL and the OOM killer: Why we use strict memory overcommit"

fpoling — Fri, 03 Jul 2026 18:35:34 +0000

You can also use cgroup to set allocation policy for a particular program.

New comment by fpoling in "Spain Orders Blacklist of Palantir from Public and Private Companies"

fpoling — Thu, 02 Jul 2026 18:58:19 +0000

And in Spain most immigrants are from Latin America with close enough culture and language to avoid most integration problems.

New comment by fpoling in "Since Linux 6.9, LUKS suspend stopped wiping disk-encryption keys from memory"

fpoling — Thu, 02 Jul 2026 18:45:17 +0000

Interesting idea. On the other hand on the latest SSD with hardware encryption the raw disk speed under Linux can be over 5 GB/s so on my laptop with 64 GB of RAM the full restoration from disk takes like 45 seconds. With LUKS it is like 2 times slower. Which is not a problem at all. So I do not see much value in memory encryption in suspend.

New comment by fpoling in "Since Linux 6.9, LUKS suspend stopped wiping disk-encryption keys from memory"

fpoling — Thu, 02 Jul 2026 17:09:35 +0000

On my laptop with Fedora I just configured Linux to hibernate to disk after 15 minutes of suspend. Powering memory off ensures that bugs like this Debian-specific would not matter.

Plus what Debian extension to Linux tooling does although nice in theory, but in practice if one really worries about cold-boot attacks, then all keys and important documents has to be wiped out from memory, not only LUKS keys.

So hibernating is really the only proper way to protect against cold boot.

New comment by fpoling in "LastPass notifies users of yet another data breach"

fpoling — Thu, 25 Jun 2026 15:36:45 +0000

I worked for a big company that switched from 1password to Keeper. The transition was smooth and I do not see why it shouldn’t be as long as IT knows what they are doing.

New comment by fpoling in "Mexican government unveils a prototype for a new homegrown, ultra-affordable EV"

fpoling — Mon, 22 Jun 2026 20:49:02 +0000

And they are so much nicer. Way less noise and pollution and much better maneuverability on narrow streets or on slippery roads in winter.

New comment by fpoling in "Identity verification on Claude"

fpoling — Sun, 21 Jun 2026 19:53:01 +0000

Many such IDs are designed for local physical verification, like proof that the mobile phone owner is above certain age or has a valid driving license, they are not designed for remote verification.

New comment by fpoling in "Prefer duplication over the wrong abstraction (2016)"

fpoling — Sun, 21 Jun 2026 19:22:40 +0000

With LLMs the cost of duplication is much lower both to write and maintain. So abstractions needs much higher justification.

New comment by fpoling in "Prefer duplication over the wrong abstraction (2016)"

fpoling — Sun, 21 Jun 2026 19:19:48 +0000

With LLMs the cost of duplication is much lower and LLMs

New comment by fpoling in "Computed goto for efficient dispatch tables (2012)"

fpoling — Sat, 20 Jun 2026 15:47:45 +0000

I wonder are there any compilers that recognizes switch-in-loop pattern of interprets and optimizes it to be branch-predictor friendly using multiple indirect jumps?