HNHacker News
TopNewBestAskShowJobs

jamesponddotco

1,449 karma · joined May 12, 2019

Hi! I am James Pond, a Sysadmin and Software Engineer from Brazil working at Datadog while building privacy-focused open source projects in my spare time.

You can get in touch with me by sending an email to james AT cipher DOT host.

submissionscomments
jamesponddotco··on Warp terminal – no more login required
My Neovim color scheme doesn’t work on Terminal.app. That’s, literally, the only reason I chose iTerm.
jamesponddotco··on Show HN: Jelly – A simpler shared inbox for small teams
My wife and I have personal emails but also one that’s shared between us and set up on devices for both, so we can keep track of things that should be shared, like credit card statements, bills, and whatnot.

We use Migadu, which allows you to have as many mailboxes as you want with any plan, so it’s pretty cheap.

jamesponddotco··on I Don't Have Spotify
Personally, I have a TIDAL subscription, download everything I like from there, and then store everything on a Plex server.

While not legal, having the subscription fix any moral problems I’d have with the idea.

jamesponddotco··on Amazon employees: 'I'd rather go back to school than work in an office again'
I've been working since I was 12, and working from home since I was 19 (32 now). I can't imagine ever going back to an office, especially now that my wife is pregnant. I'd rather learn a new profession than doing that.
jamesponddotco··on Show HN: Documentor, improve technical documentation with the power of AI
Wrote this for work the other day and thought people might be interested. The tool can review and draft technical documentation, as well as generate alt tags for images.

It supports both OpenAI and Anthropic's models right now, but PRs adding support for other providers are welcome :)

jamesponddotco··on Dungeons and Dragons taught me how to write alt text
I've to write alt tags daily and I still suck at it, since I suck at describing things (which is weird, since I write documentation every day). I might start attending my friend's D&D sessions just to improve on that.

For now, I wrote a tool[1] that uses AI to do the job for me.

[1]: https://git.sr.ht/~jamesponddotco/allalt

jamesponddotco··on Prelude – a tiny CLI tool building context prompts from your code
Wrote something similar[1] using Go a while back, but I love that this one only uses shell tools. Extremely useful when working with LLMs for small to medium projects.

For bigger ones on the other hand, I wish I could just get the exact context I need. Maybe something like only the files open in vim?

[1]: https://git.sr.ht/~jamesponddotco/llmctx

jamesponddotco··on FOSS funding vanishes from EU's 2025 Horizon program plans
> Does anyone know any good FOSS projects that were actually funded by this?

SourceHut[1] comes to mind.

[1]: https://sourcehut.org/

jamesponddotco··on IRC technology news from the first half of 2024
I don't know about others, but I still use it for the same reason I still use XMPP: still works fine and is less annoying than proprietary solutions.

As for encryption, I use OMEMO with XMPP for private conversations, but I use IRC for public forum conversations, so TLS is enough for me. Think Hacker News, but in real-time.

jamesponddotco··on A personal music streaming server that works
I make playlists on TIDAL and then download the playlist. Or favorite albums and whatnot, and download them.
jamesponddotco··on Ask HN: What is your favorite rolling Linux distro?
Another vote for Tumbleweed here.

Heck, the thing is so stable I use it on production servers. The combination of their automated QA plus your own QA on staging is the bomb.

jamesponddotco··on Ask HN: What is your experience with Claude 3.5?
Pretty positive, much better than my experience with OpenAI's models. The killer feature for me is their prompt generator[1], which you can use to create system prompts or improve user prompts. As I said in another thread, the generator is tuned to generate prompts better suitable for Claude, which improves responses.

You can see examples of the prompts it generates in the repository linked below, including a customized version of their prompt generator[2]. This feature significantly improved my experience with LLMs in general, both Claude and local ones, and I now use Claude 3.5 Sonnet for pretty much all my coding, mostly in Go and, lately, Rust.

I mostly use it to improve existing code or to get started, not to generate entire code bases, so I can't say much about that.

I found it lacking in shell scripting though. Nine out of ten times I need to fix the shell script it generates, to a point where I just gave up trying and went back to writing them from scratch.

Can't wait to see how much better Claude 3.5 Opus is, though.

[1]: https://docs.anthropic.com/en/docs/build-with-claude/prompt-...

[2]: https://sr.ht/~jamesponddotco/llm-prompts/

jamesponddotco··on Ask HN: Are you leaving Cloudflare over the recent drama?
I left them a long time ago, as I felt—and still do—like the internet was too dependent on them. None of my projects ever needed DDoS protection though, so I mostly used them for DNS and caching at the edge.

If I ever needed DDoS protection it'd be nice to know of alternatives, if anyone has any to recommend. My provider—Hetzner—provides protection to some degree, but I'm sure a big attack would lead to black-hole routing.

jamesponddotco··on GitHub repo extensively mirrored by GitCode, platform launched by China's CSDN
Doesn't appear like any of my public repositories have been "stolen." Not sure if I should take that as a good thing or as an offense, to be honest, lol.

The code is open source, so I wouldn't mind anyway.

jamesponddotco··on Anthropic: Collaborate with Claude on Projects
More discussion over here:

https://news.ycombinator.com/item?id=40790911

jamesponddotco··on I switched to Vim – and it took me only 6 years to learn it
This. I have loved and used Vim since I was 18 years old, but I did because I fell in love with the tool right away and felt right using it, not because it was "cool" or whatever.

I tried other editors, it didn't feel right, I didn't force myself to continue using them. My advice if you're trying vim is the same... if it doesn't feel right, if it doesn't feel like the tool works for you, look somewhere else.

jamesponddotco··on Netflix's bet on advanced video encoding
Because of this quality issue, I always wondered how people who rip Netflix content handle this, since they often offer 2160p Dolby Atmos versions of movies or even TV show episodes for download.

How do they maintain the quality long enough to rip the content? When I used Netflix years ago, I could barely keep a constant 1080p stream at times.

jamesponddotco··on Claude 3.5 Sonnet
Adding a +1 to this request. Something like Codeium for NeoVim but using Claude 3.5 Sonnet as the model would be swell.
jamesponddotco··on Claude 3.5 Sonnet
One thing Anthropic did that I loved and think was very smart was building a prompt generator[1] into the developer console. The generator is tuned to generate prompts the way Claude prompts are supposed to be, which improves responses. And you can use it to improve your user prompt as well, not just your system prompt, which make responses even better.

You can see examples of the prompts it generates here[2]. It significantly improved my experience with LLMs; I haven't touched GPT4 in quite a while, and GPT4o didn't change that.

[1]: https://docs.anthropic.com/en/docs/build-with-claude/prompt-...

[2]: https://sr.ht/~jamesponddotco/llm-prompts/

jamesponddotco··on Show HN: A GPT-4.0-based manga translator that can produce translated manga
Seems to work pretty well, kudos!

I wonder if a version powered by an uncensored Llama 3 model would be better, though? That way you could expand to hentai as well.

Is the service open source?

EDIT: Also missing a privacy policy, which seems quite important given that people are uploading images to your service.

jamesponddotco··on Gravatar Is Back
Seems like a good endpoint to add to Privytar[1][2] as well, but looking at the announcement post at the Gravatar blog, I'm not sure I'm comfortable with where they're headed from a privacy perspective.

I was looking for an excuse to update the project and upgrade to Go 1.22; I guess I just found one.

[1]: A privacy and caching reverse proxy for Gravatar.

[2]: https://git.sr.ht/~jamesponddotco/privytar

jamesponddotco··on Signal: Will leave the EU market rather than undermine our privacy guarantees
Another vote for XMPP with OMEMO here. I use it my wife, family, and a few friends and haven't heard any complaints yet[1] (aside from the lack of stickers, which are apparently important).

[1]: But like I said in another thread, who knows if they would directly complain to me.

jamesponddotco··on Signal: Will leave the EU market rather than undermine our privacy guarantees
The law says "images, photos, videos and links". What about simple encrypted or password protected zip files?

As far as I know, most messaging platforms allow you to send regular files too. Wouldn't "the bad guys" simply use that as a loophole and continue with their day?

I know the real reason behind the law isn't to actually protect children, but, you know...

jamesponddotco··on The NSA advises you to turn your phone off and back on once a week
GrapheneOS[1] has that option in the security settings. I have mine set up so it automatically reboots the device if I haven't unlocked it in the last 8 hours.

[1]: https://grapheneos.org/

jamesponddotco··on Let yourself be monitored: EU governments to agree on Chat Control
I'd say OMEMO[1] is the standard for XMPP these days, not OTR/PGP.

I switched my wife, family, and a few friends over to XMPP. Aside from the lack of "stickers" in Conversations and Dino, they seem to be doing okay[2].

[1]: https://conversations.im/omemo/

[2]: But then again, this is the only way to reach me, so maybe they just don't complain to me directly.

jamesponddotco··on Ask HN: What is your ChatGPT customization prompt?
Instead of using custom instructions, I use the API directly and use the appropriate system prompt for the task at hand. I find that I get much better responses this way.

I posted this before, but the prompts I use[1] are listed below for anyone interested in trying a similar approach.

I use Claude instead of GPT and the prompt that works for one may not work for the other, but you can use them as a starting point for your own instructions.

[1]: https://sr.ht/~jamesponddotco/llm-prompts/

jamesponddotco··on Ask HN: What is your ChatGPT customization prompt?
It's even more interesting if you take into consideration that for Claude, making it be more verbose and "think" about its answer improves the output. I imagine that something similar happens with GPT, but I never tested that.
jamesponddotco··on Cortile – Linux auto tiling manager with hot corner support
This looks pretty nice, I'll have to try it if I ever switch back to X11. What surprises me the most is that it's written in Go.

Not a bad thing, I love Go. It surprised me is all.

jamesponddotco··on Amber: Programming language compiled to Bash
Pyinfra's documentation is definitely on my list of things to read.
jamesponddotco··on Amber: Programming language compiled to Bash
Not gonna lie, I have a certain prejudice against Python because of the whole packaging issue, something I definitely need to work on. I did ship Python code before, I just don't really like using it.

I do have the pyinfra documentation saved so I remember to take a look at it. I love Go for many reasons, but the distribution of binaries is probably at the top.

← PreviousPage 5 of 13Next →