back
user profile
7777777phil
1,567karma·702submissions·August 28, 2025
recent activity (702 total)
comment
This is just so cool! Not much more to add. Thanks a lot for sharing!! Great work :)
comment
dang, thanks for the thoughtful response. I’ve been following your recent comments on the "curmudgeonly" guideline, and it’s clear this is a priority for the health of the site in 2026. My c…
comment
Hi, if you find the time, please reach out to the email in my bio with the methodology and dataset you used. Did you only test one post or a whole set? Would be interesting to compare the setup. Thank…
comment
self-fulfilling prophecy
comment
You're making a distinction the paper should address more directly. The classifier can't tell the difference between "this API design is fundamentally flawed because X" and "t…
comment
OP here :)
On skepticism being lumped with negativity: partially true. The SST-2 training task treats critical evaluation as negative sentiment. I should clarify that "negative" here means e…
comment
Fair point, `ai_sentiment` should have been defined explicitly. It's the production score from DistilBERT-base-uncased-finetuned-sst-2-english, the same model family as Cloudflare's sentimen…
comment
You're right that a simple Python script would be more cost-effective for this kind of archiving. I went with workers because I was already familiar with the stack and wanted real-time processing…
comment
Yes, probably the core limitation of my analysis (see earlier comment). My classifiers are treating "I don't like React because I don't want to write HTML in my JavaScript" the sam…
comment
The classifiers I used are definitely conflating technical criticism with genuine negativity, and that's a real limitation. When I say "technical critique reads differently than personal att…
comment
Thanks for the pushback this is exactly the kind of peer review I was hoping for at the preprint stage. You are likely correct regarding the sampling bias. While the intent was to capture all. posts, …
comment
Open to the Idea, how exactly would you classify "high energy posts" ?
comment
Hi, appreciate your comment. The sampling is from all posts / comments over the past 35 days, accessed via the API ( https://github.com/philippdubach/hn-archiver ). There migh…
comment
Hi I'm trying to catch up to the comments but definitely appreciate this! I wrote a worker that uses the public API to archive all post and comments and create regular snapshots to track growth e…
comment
Thanks! And appreciate you sharing the Harris clip. I sometimes listen to his meditations: https://philippdubach.com/posts/gratitude …
comment
I don't see any issues on my side Largest Contentful Paint (LCP): 0.5 s
Cumulative Layout Shift (CLS): 0.01
First Contentful Paint (FCP): 0.5 s
Time to First Byte (TTFB): 0.3 s Also not on the Cl…
comment
Related threads on overwhelmingly negative comments:
- https://news.ycombinator.com/item?id=40430263
- https://news.ycombinator.com/item?id=32400521 …
comment
There was this one Black Mirror episode..