Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
efxhoy
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
11 ms
·
91.
▲
by
efxhoy
3y ago
You can’t run video cards in the new Mac Pro. https://www.macrumors.com/2023/06/11/apple-exec-discusses-ma...
92.
▲
by
efxhoy
3y ago
I agree. I think the rare frustration when you want the timestamp greatly outweighs the much more frequent but ever so slight cognitive convenience from the “X time ago” format.
93.
▲
by
efxhoy
3y ago
We run the exact same setup, in my two years at the company we’ve only had one migration related issue and that was due to different minor versions of postgres between CI and staging. Now if postgis could get those official ARM docker image
94.
▲
by
efxhoy
3y ago
I had to implement a workaround for that for our DW which imports from two different app databases. IMO this should be part of import foreign schema but then you would have to qualify the names somehow if you were importing from more
95.
▲
by
efxhoy
3y ago
I work at a company trying to get more people into our app. The logic is completely flawed: "we want more engagement, app users show more engagement than web users, so if we get more users from web into the apps we'll get more eng
96.
▲
by
efxhoy
3y ago
I wrote simulation software at a research department previously. 99.8% of the problems were perfect for numpy/pandas and learning a compiled language wasn’t worth it for the 0.2%. If I was to work the same problems again I’d do it in t
97.
▲
by
efxhoy
3y ago
For work we always pick the cloud version. Chances are whoever is running that cloud version knows their shit better than we do and if we self host it it's on AWS anyway so probably not much cheaper if scale is any issue.
98.
▲
by
efxhoy
3y ago
Have your query as a function in the database. BEGIN, do the migration, replace the function, COMMIT. Postgres DDL is transactional so at no point does the function stop working.
99.
▲
by
efxhoy
3y ago
We decided to use DynamoDB as a cache for our expensive-to-compute statistics from the data warehouse so we can show them to users without putting a big read load on the data warehouse. DynamoDB scales really well right? And big statistics
100.
▲
by
efxhoy
3y ago
RME?
101.
▲
by
efxhoy
3y ago
Neat tool! I prefer using environment variables for my connection info to avoid passwords in my shell history or plaintext config files. The standard[1] ones from libpq work in psycopg2 if you pass an empty connection string. Then you can k
102.
▲
by
efxhoy
3y ago
I guess my "don't bother unionizing" take is extremely dependent on where you live. I'm in Sweden where we already have very generous benefits and labour friendly laws.
103.
▲
by
efxhoy
3y ago
As long as tech workers can credibly threaten to quit I think that’s easier than unionizing. Unions are great when you have a single big employer (a monopsony) like the public sector or a dominant single company, like a factory in a small t
104.
▲
by
efxhoy
3y ago
I use copilot and chatgpt to help me write terraform code, which i’m still very new to. It’s decent at showing me which IAM role policy data attachnent thingy I need to add next. It’s not that it’s particularly amazing, more that the google
105.
▲
by
efxhoy
3y ago
We use Sagemaker at work because AWS. I don't really like their style of APIs but it works.
106.
▲
by
efxhoy
3y ago
It’s the biggest and most capable rocket yet and it flew higher and faster than the second most powerful (the N1) ever did. It’s progress. Success or failure of the test itself is secondary. We’re getting closer to the stars.
107.
▲
by
efxhoy
4y ago
What prompt did you use? I immediately smelled chatgpt from the comment too. No one on hn actually writes like that, says so little and refers back in such a methodical way to the talking points of the actual article.
108.
▲
by
efxhoy
4y ago
If you want to estimate causal effects you need a randomly assigned treatment. When studying the effects of wealth a lottery gives you exactly that. The critique would be that the findings don’t necessarily generalize outside of the lottery
109.
▲
by
efxhoy
4y ago
> But nobody knows how GPT works. They know how it was trained, because the training scheme was designed by humans, but the algorithm that is executed during inference was not intelligently designed but evolved, and it is implicit in the
110.
▲
by
efxhoy
4y ago
They mentioned their mrsk tool sshes to the boxes to deploy so the action probably runs the tool and does just that?
111.
▲
by
efxhoy
4y ago
Absolutely. I'm pretty new to infra and I learned in our big company AWS environment. I've dabbled a bit with home servers and stuff but I have no idea how to manage a bare metal host "properly".
112.
▲
by
efxhoy
4y ago
How did you setup the k8s cluster? Is it running on bare metal or VMs?
113.
▲
by
efxhoy
4y ago
I spent some time working in a prison, it was very eye opening for someone like me from a pretty sheltered background. Most of the inmates were decent enough people that made bad choices but there were a few who were just horrible sadistic
114.
▲
by
efxhoy
4y ago
black has an optional jupyter component that lets it format notebooks with pip install "black[jupyter]"
115.
▲
by
efxhoy
4y ago
One thing I really like about RDS is the query visualizer. It shows which queries ran when and for how long and what they spent their time waiting for, CPU or disk or locks or whatever. It’s really good. Are there any free tools that show a
116.
▲
by
efxhoy
4y ago
Do you have books to recommend about shell scripting?
117.
▲
by
efxhoy
4y ago
Sure, it’s definitely possible. But getting a consistent layout, copy paste behavior and locale setup I like across editor, terminal, tmux, file explorer and browser in a Swedish ISO layout is a surprisingly big chore. I know a few Swedish
118.
▲
by
efxhoy
4y ago
The default keyboard layout is a big reason why I keep using macs tbh. I’ve used windows all my life and used a bunch of linux distros but the only os where my fingers are comfortable is on a mac. Swedish ISO btw.
119.
▲
by
efxhoy
4y ago
I write mostly batch ETL stuff. All plain psql and bash. We don’t have a good testing setup to be honest. What we do use instead: Plenty of constraints, uniques and foreign keys and not nulls. Enum types. Visuals, dump to csv and plot some
120.
▲
by
efxhoy
4y ago
I learned this from the sopranos: https://youtu.be/J2TkCgLooRU
More ›