dustfinger

Here are the last two sentences:

Governments don’t do anything about the status quo, for a number of reasons: it inconveniences them to look too deeply into the darker corners of their own financial systems, and they make money from printing their own currencies and don’t much care how that cash is used. But most of all, they don’t do anything about it because they haven’t got a clue.

The last one couldn't be farther from the truth, and the first one couldn't be farther from a lie.

I think I have watched Star Trek: TNG all the way through 3 times with my kids already. They also love Deep Space 9.

It was happier, with plenty of opportunity for everyone and society felt like it was held up on common ground. It defnately felt bigger, because it was.

Any, and I mean any, attempt to remake this show is doomed to failure

I second that. Please, for the love of all good things, do not remake the X-Files, or Firefly.

It would be interestingi to know if AI is less likely to follow rules if the instructions provided to it contain foul or demeaning language. Too bad we couldn't re-play the scenario replacing NEVER F*ING GUESS! with:

**Never guess**

   - All behavioral claims must be derived from source, docs, tests, or direct command output.

   - If you cannot point to exact evidence, mark it as unknown.

   - If a signature, constant, env var, API, or behavior is not clearly established, say so.

Only a receiver with the right equipment can pick up the hidden message.

So all an eavesdropper has to do is setup the right equipment then? I guess it is only invisible until the technology becomes more widely available.

We’ve combined lab-grown neurons with silicon chips and made it available to anyone, for first time ever.

There is a line somewhere here that I personally feel we should not cross.

Appreciate this breakdown and the disclosure.

+1 for highlighting that PR quality is the bottleneck. Garbage-in/garbage-out is exactly what I ran into and it’s why I’m planning to introduce PR templates so the why/what changed/impact is consistently present. For sparse PR bodies, I also optionally add truncated diff context for the LLM summary so the output isn’t just a long list of raw PR titles.

Also agree there is a split between dev-facing changelogs vs user-facing release comms that need to land where users are. What I built is aimed at the "developer-consumer" audience, people using the library, not contributing to it: it renders into our docs and is meant to be readable as a curated changelog, not a raw list of commits.

I agree that the best quality notes are the ones hand written by a thoughtful human. In my case we had about two years of history with no curated notes, and writing that by hand would have meant significant time investment vs shipping fixes and features. The generator helped us get coverage fast, organized the notes chronologically and categorically. I specifically designed the generator with your your concern in mind, in that it preserves manual edits as well as omissions, so we can gradually curate it into something we are proud of.

I agree with the philosophy of curating release notes for the consumer of the release. When I first started looking for a release notes strategy, I was considering towncrier for that exact reason. You are also right that commit messages are not intended for the consumer of the release, but a dialogue between developers.

Your points are well received and largely why I went PR-based (title/body with optional GitHub metadata) instead of commit-based. A PR title and body tend to be focused on the deliverable, whereas commit messages are narrowly focused on the code change at that moment with developers as the intended audience.

Re: git-cliff, I honestly hadn’t evaluated this one, but it looks solid for commit-driven changelogs. I like the rationale behind conventional commits being parsable and templates enforcing consistency. What constraints pushed you toward git-cliff vs writing release notes by hand, and do you have a config/template you have found works well for surfacing breaking changes?

Yeah, that matches what I have seen: if the upstream metadata isn’t reliable, automation can amplify the mess.

I tried to avoid relying solely on contributors to accurately label or tag things correctly. The script is tag-driven only for release boundaries (version tags), while categorization is derived from PR title & body with optional GitHub metadata. The script is idempotent and preserves edits/omissions so you can correct the few bad ones post-generation.

If you are curious, I am happy to share my script and would be genuinely interested whether it reduces the manual cleanup for your workflow. Also, if you run it with `--ai --github` and a PR body is sparse, it fetches a truncated PR diff and uses that as extra context for the LLM summary.

+1 on separating "how to upgrade" due to breaking changes from "what’s new". A dedicated BREAKING.md / MIGRATIONS.md is a really good idea.

One thing I am trying to do is make the generator surface breaking/migration items explicitly, but I still think anything that requires human judgment (migration steps, caveats) should be hand-curated in a dedicated document like you suggested.

I hear what you are saying, there is a risk that auto-generated release notes end up as PR-title soup. I put a lot of effort in my script to mitigate against that.

If you are willing and interested enough to take a quick look, here is what my script generated for our 2025 changelog (no hand-curation yet, this is the raw output):

https://raw.githubusercontent.com/confident-ai/deepeval/refs...

I am curious: does this still seem too noisy in your opinion, or is it getting closer? And what would you want to see for breaking changes/migrations to make it actually useful?

I now have 2024 & 2025 generated; to fully hand-curate two years of history just wasn’t practical, so I’m trying to get the "80% draft" automatically and then curate over time.

That has been my concern as well. The script I wrote tries to bucket entries into categories, including "Backward Incompatible Change" so those are easier to spot. Since it is automated I am trading some accuracy for time saved, which seemed like the only practical choice for me, since I had to backfill a lot of history, but it’s been surprisingly decent so far.

I am also planning to add some PR templates so contributors include the context up front, which should make any release note generation more accurate.

Are you using any tooling to help with changelog curration? I know towncrier is all about fragments, so contributors must had write a brief summary of their contribution, which would be more in-line with your preference.

I am curious what are people using for release notes in their own projects? Towncrier, GitHub Releases, something else?

If anyone tries my script on their repo and runs into issues, I am happy to help troubleshoot. Also, the output is actually plain Markdown (no JSX). The only Docusaurus specific bit is the YAML frontmatter header. If you are not using Docusaurus, you can just strip that header and rename .mdx to .md.

I realize cache segregation is mainly about security/compliance and tenant isolation, not protecting secret prompts. Still, if someone obtained access to a company’s prompt templates/system prompts, analyzing them could reveal:

- Product logic / decision rules, such as: when to refund, how to triage tickets

- Internal taxonomies, schemas, or tool interfaces

- Safety and policy guardrails (which adversaries could try to route around)

- Brand voice, strategy, or proprietary workflows

That is just off the top of my head.

I wonder if there is valuable information that can be learned by studying a companies prompts? There may be reasons why some companies want their prompts private.

Ouch!

What we need is a "Little Bobby Tables" vanity plate that exploits a buffer overflow in speed cams.

True. how about https://ycombinator-status.com or https://status.ycombinator.com. They can drop that link on their official ycombinator.com site just so it is easy to tell that it is legit when they are not down. They can maintain the status for all of their official sites there. Well, with the exception of ycombinator-status.com of course. I guess they can post that status on some-sketchy-domain/status-right-now - LOL!

Maybe ycombinator does have an official status page somewhere, but it is not easy to find if that is the case.

It would be a nice feature if hackernews had a simple news.ycombinator.com/status page that I could trust. I don't like loading those some-sketchy-domain/status-right-now sites.

Thank you, I will check that out.

If you are an emacs user, you might be interested in org-drill, which of course is plain text and uses space repetition algorithm:

https://orgmode.org/worg/org-contrib/org-drill.html

When I saw the title, I thought it was going to be about exercise :-P

Hmm, I am getting a 403 when trying to load the link.

While that is partly what I meant, my notion was really about being creative, inventive and having a lot of fun hacking code or hardware without it being about work. I really meant these notions of hacking:

- https://stallman.org/articles/on-hacking.html

- https://stallman.org/articles/happy-hacking.html

- http://www.catb.org/~esr/faqs/hacker-howto.html

Despite all the critisism, C++ has been my favorite language for 20+ years. Nowadays I code 99% of the time in python and previously TypeScript, but all my personal for fun projects are in C++. I just enjoy coding in it so much more. I know I will get some heat for this, but I would love to live in an idealistic world where computers were still for hackers and it was all just about fun not profit. I was so inspired by the writings of Eric S. Raymond, I always hoped I would experience some of that. ~SIGH~. Maybe in my retirement.

Can you elaborate. I just use it by itself and think it is wonderful, but I am curious for some examples of using it with other tools.

EDIT: I just looked at the link posted by lvass and I understand now. So cool! I am going to update my workflow.

I would agree that the usefulness is reduced if you have tight vertical space constraints (3 lines of text).

Don't underestimate the advantage of being able to move the cursor anywhere in a few clicks without having to take your hands off the keyboard. It is much faster than a mouse. Also, you can save your elbows the pain of constantly reaching over for the mouse.

avy does more than just jump the cursor to a specific place. It also allows you search for, copy and move text round without needing to move your cursor to that text. It is extremely easy to use and very efficient.