Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
rsyring
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
10 ms
·
121.
▲
by
rsyring
11mo ago
TL;DR: Greater insight → greater exposure to brokenness (maybe a harder time minimizing/ignoring it) → potential for greater sorrow. Ecclesiastes 1:12-18 (traditionally understood to be written by King Solomon, son of David): I the Pre
122.
▲
by
rsyring
1y ago
For those who are going to compare to Codex, make sure you understand what model you are working with: https://cookbook.openai.com/examples/gpt-5-codex_prompting_g... The "Codex" model requires different prom
123.
▲
by
rsyring
1y ago
What job did you switch to? Are you happy with the change?
124.
▲
by
rsyring
1y ago
How to enter a city like a king? Humbly, on a donkey. As one who comes to serve instead of one who comes to be served. What a difference from what we can expect from our current leaders, kings or king wannabees alike.
125.
▲
by
rsyring
1y ago
And they have fun with it: HTMX Sucks mug: https://swag.htmx.org/products/htmx-sucks-mug
126.
▲
Ask HN: How do you run LLM Agents safely?
6 points
by
rsyring
1y ago
|
1 comments
127.
▲
by
rsyring
1y ago
Please note that in my OP, I never mentioned containers. Let's say, as a developer, I need to do some API interactions with GitHub. So, in a terminal, using 1Password's cli tool `op`, I load my GH API token and pass it to my Pyth
128.
▲
by
rsyring
1y ago
I've never tried to use ptrace-read to read secrets from a running process's memory so I can't comment on that part. I had always assumed getting secrets from a running process' memory was non-trivial and/or require
129.
▲
by
rsyring
1y ago
I added a note to my original post about how 1Password's cli tool work. Access to the binary doesn't mean automatic access to any authorized session. I think that's a good start. In my case, I have it tied into my fingerpr
130.
▲
by
rsyring
1y ago
Environment variables are often used to pass secrets around. But, despite its ubiquity, I believe that's a bad practice: - On Linux systems, any user process can inspect any other process of that same user for it's environment va
131.
▲
by
rsyring
1y ago
This should be in the docs or a blog post somewhere. Very clear explanation.
132.
▲
by
rsyring
1y ago
Sometimes value is measured by awareness. I benefited from becoming aware of the behavior because of the article. Yes, it's in the docs, but the docs are not something I would have gone looking to read today.
133.
▲
by
rsyring
1y ago
Updated title with year. Thanks for catching that.
134.
▲
I've Been at NPR for 25 Years. Here’s How We Lost America’s Trust. (2024)
(thefp.com)
5 points
by
rsyring
1y ago
|
7 comments
135.
▲
by
rsyring
1y ago
Thanks for the clarification. I don't use any tools that do remote caching so I guess I just glossed over that qualification. :o/
136.
▲
by
rsyring
1y ago
Turns out caching tasks is an anti-goal: https://mise.jdx.dev/roadmap.html#anti-goals > Remote task caching - turbopack, moonrepo, and many others are trying to solve this (major) problem. mise's task runner will li
137.
▲
by
rsyring
1y ago
> I went looking for an issue to see if they're planning it, but the Mise repo doesn't have issues enabled? And no discussion on the README about why they don't. That doesn't inspire confidence. I'm not sure when
138.
▲
by
rsyring
1y ago
If you know you will use it often, uv has `uv tool install ...`. So, after the first `uv tool install ut` you can just run `ut md5 ...` or whatever. Don't need to keep using uvx. uv also has a couple commands I throw in a systemd uni
139.
▲
by
rsyring
1y ago
> Sculptor is free while we're in beta. Ok, and then what? Honest question.
140.
▲
by
rsyring
1y ago
> uv is another newer Python package manager that gives gives Poetry a run for its money, with super-fast package and dependency management. I don't think uv is just giving Poetry a run for it's money. I think uv is winning an
141.
▲
Testing is better than data structures and algorithms
(nedbatchelder.com)
188 points
by
rsyring
1y ago
|
176 comments
142.
▲
by
rsyring
1y ago
Seriously. If you are considering Tarsnap, or use it but have never tested a restore, don't hand wave this comment away. You might be tempted to think: it's a popular service, it can't be that bad. But, it really can be, and
143.
▲
by
rsyring
1y ago
Apparently, I didn't miss anything after all. :( See sibling comment above about them being just remote binaries.
144.
▲
by
rsyring
1y ago
IMO: make all the laws you want. They generally won't be enforced and, if they are, it will take 5-10 years to make it's way through the courts. At best, the fines will be huge and yet account for maybe 10% of the revenue genera
145.
▲
by
rsyring
1y ago
I've always told my team: focus on being super productive in the 40 hours a week you are working. Then go home and do something that really matters. My belief has been very few lay on their death bed wishing they had given more to the
146.
▲
by
rsyring
1y ago
Wow! I can't believe I missed that. Thanks.
147.
▲
by
rsyring
1y ago
> cost additional developer time, which for most OSS is fairly limited Mostly agree. But something like Podman w/ RedHat behind it is unlikely to be limited in the same way a lot of community OSS projects are. Unfortunately, I disa
148.
▲
by
rsyring
1y ago
Ubuntu is committed to the Snap ecosystem and there is a lot of software that you can get from a snap if you need it to be evergreen.
149.
▲
by
rsyring
1y ago
> What else can they do... They can do what Docker and many other software providers do that are committed to cross OS functionality. They could build packages for those OSes. Example: https://docs.docker.com/engine
150.
▲
by
rsyring
1y ago
I believe rootless containers require Linux user namespaces which have historically been the source of many vulnerabilities: https://news.ycombinator.com/item?id=43517734 I'm conflicted about whether or not it's
More ›