eatonphil

It's an extremely early project but it's been helping me out a bunch already. Would love to hear your bug reports or requests if you try running it on your codebase.

[0] https://theconsensus.dev/p/2026/06/27/the-feature-in-oxcaml-...

* I think you can get this in Clang and maybe GCC with `__attribute__((nonallocating))` though.

Foundational Cryptography in Spark adacore.com

1 eatonphil 3 days ago 0 comments

The Web Can't Afford a JavaScript Runtime Monoculture redmonk.com

3 eatonphil 4 days ago 1 comments

Protobuf in Python and TypeScript is good now buf.build

3 eatonphil 4 days ago 0 comments

Comment · 5 days ago

> For real judgements on database products, I'd look to https://jepsen.io/. That author tends to be refreshingly straightforward in a field where many vendors love to cheat by different names

As far as I'm aware, Jepsen solely focuses on correctness (which is more than enough for someone to do!), not performance.

Comment · 12 days ago

I was not precise there sorry. They were not crashes but Lightpanda failed to render them. It was these URLs

https://duckdb.org/2025/07/04/ducklake-02.html

https://ducklabs.com/news/2024/06/03/duckdb-version-100-rele...

https://neovim.io/doc/user/dev_arch

https://www.twelvelabs.io/blog/%EB%B9%84%EB%94%94%EC%98%A4%E...

Costco Haul Culture and Land Use urbancourtyard.substack.com

4 eatonphil 14 days ago 1 comments

Comment · 15 days ago

I have a research folder of some findings comparing Chrome, headless-shell, lightpanda, jsdom, etc. Lightpanda still did quite well, though with much smaller multiples compared once you've mildly tuned Chrome/headless-shell, but it crashed more often than Chrome and headless-shell.

I haven't yet had time to write it up on The Consensus. So here are some findings you can take or leave.

  Configuration                       CPU s  Mem MiB  Median  Matched  Errors
                                      /page   (peak)    page  (of 40)
  Chrome, default flags                2.31    1,229    4.4s      ref       0
  Chrome, tuned                        1.98    1,244    3.1s       40       0
  Chrome, tuned + block all trackers   1.31    1,338    2.2s       37       0
  headless-shell, today's settings     2.30      748    3.9s       39       0
    + stop at network idle             2.16      772    2.6s       39       0
    + block ads/analytics              1.79      748    2.1s       39       0
  headless-shell, tuned                1.67      723    2.1s       39       0
  headless-shell, tuned + no frames    1.29      696    1.9s       39       0
  headless-shell via Playwright        2.58      805    4.2s       38       0
  Firefox via Playwright               6.94    2,270    5.6s       38       0
  WebKit via Playwright                4.76    4,125    5.2s       36       0
  Lightpanda, default                  1.31      540    3.6s       35       4
  Lightpanda, tuned                    1.04      509    1.3s       35       4
  jsdom                                2.66    2,102    4.3s        4       8
  Plain HTTP fetch, no JS             0.002       97   0.05s        0       0

Comment · 16 days ago

hahaha that is a good one.

Comment · 17 days ago

For sure, there are ~23,000 articles in the feed (which only goes back 5 years, but grows naturally as people publish and as we index more companies in software infrastructure).

The point is to index interesting jobs and articles to 1) help people follow the technologies they're interested in and 2) to help us figure out what to write about.

Articles get tagged with mentioned infrastructure technologies (e.g. Postgres [0] or Lean [1]) and get tagged semantic categories too (e.g. compilers [2] or network protocols [3]). (And all the same tagging happens for jobs as well.) The weekly newsletter [4] features articles and jobs from a few key semantic categories. But you can also build your own RSS feeds or email watchlists for whatever you're interested in.

There's a lot of filtering done to keep out marketing content, listicles, etc. And checking AI tells is the next stage of that.

Limiting to only good quality articles lets us build broader intelligence like the ranking of programming languages [5] and databases [6] and helps us figure out which companies are most focused on the technologies and so on.