jacobsenscott

6 ms·

61.
by jacobsenscott 1y ago
It takes a lot of discipline to keep secretes out of dotfiles. You should do that anyway, even if you have a private dotfile repo. Even if they aren't secrets per se - they can still be useful for social engineering. I don't und
62.
by jacobsenscott 1y ago
There are plenty of countries where all government services have "expired". Which one would you like to live in?
63.
by jacobsenscott 1y ago
Yeah. Imagine everything else that's completely wrong in that app.
64.
by jacobsenscott 1y ago
> Also what's the business model? The hype cycle business model never changes.
65.
by jacobsenscott 1y ago
The problem with LLMs is the answer always sounds right, no matter if it is or isn't. If you already know the answer to a question it is kind of fun to see an LLM get lucky and cobble together a correct answer. But they are otherwise
66.
by jacobsenscott 1y ago
When all you get back is a wall of LLM generated text blocking ads will be impossible. This will go the same way as google search results. Probably within six months.
67.
by jacobsenscott 1y ago
I've yet to see an LLM produce anything but disastrous bugs. Remember they don't know they're writing code that can destroy companies and livelihoods. They are just spitting out plausible but always wrong text.
68.
by jacobsenscott 1y ago
AI replacing devs talk is about short term stock pumping and short term COGS reduction. The long tail is someone else's problem.
69.
by jacobsenscott 1y ago
Weekend HN is always like this.
70.
by jacobsenscott 1y ago
Today we have humans being trained on llm garbage - kids using it to do their homework - programmers using it to "learn" how to code, med students cheating their way through med school, etc. So the content the humans are producing
71.
by jacobsenscott 1y ago
This is always because people would rather than twiddle their thumbs in meetings than work - so they drag it out as long as possible. Getting paid for doing nothing is a good deal. Meetings are never necessary, and usually the worst possibl
72.
by jacobsenscott 1y ago
You can partition that over 20 or 30 or more tables on one PG instance and have good performance - assuming a good partitioning key exists. If you need to query all 10B rows you'll have a bad day though.
73.
by jacobsenscott 1y ago
PG requires a lot of expertise to keep running when you get to a billion rows or massive ingest. It can do it, but it doesn't just do it out of box running the defaults.
74.
by jacobsenscott 1y ago
CH excels at extremely high volume writes. You probably can't throw enough data at it.
75.
by jacobsenscott 1y ago
Random chance, and maybe a little skill.
76.
by jacobsenscott 1y ago
Having a garage isn't enough. A lot of people with garages need to work everyday to pay for their garage, and food, and everything else. Bezos and Jobs both had free garages and free time paid for by their parents. I would bet the othe
77.
by jacobsenscott 1y ago
That's because it's glorified search. The postgres docs tell you that without risk of hallucination. You are correct that it won't produce code that does the right thing in that context though.
78.
by jacobsenscott 1y ago
Soon most software devs will just be transcribing LLM trash to code with no concept of what's actually happening (its actually required at shopify now - MS is bragging 1/3rd of their software is written this way), and no new engin
79.
by jacobsenscott 1y ago
This is a marketing blog for some AI startup trying to sell AI voodoo to the insurance industry, so it just needs to sound like it might be useful for finding fraud.
80.
by jacobsenscott 1y ago
This isn't a good way to learn this. If you don't know how rust memory management works you don't know if the llm is just hallucinating the answer.
81.
by jacobsenscott 1y ago
"He would chuck, he would, as much as he could, if a wood chuck could chuck wood" is how I learned it.
82.
by jacobsenscott 1y ago
Well, this is want American Elites want - so they can buy up the pieces at a discount.
83.
by jacobsenscott 1y ago
You might want to read Moby Dick all the way to the end.
84.
by jacobsenscott 1y ago
That loyalty and affection your are feeling is only going one way. I've worked for small and large places. Work is always transactional. The day the CFO at your 10 person startup that "feels like a family" gets some pressure
85.
by jacobsenscott 1y ago
Everything devolves to ad sales. Do you know the minute details about their lives people type into chat gpt prompts? It's a gold mine for ads.
86.
by jacobsenscott 1y ago
s/is/was/
87.
by jacobsenscott 1y ago
That's a marketing page for something called qodo that sells ai code reviews. At no point were the ai code reviews judged by competent engineers. It is just ai generated trash all the way down.
88.
by jacobsenscott 1y ago
These incremental benefits don't translate to an order of magnitude more productivity, or stability, or profitability. Given the choice, as a business owner, future proofing is about being able to draw from the most plentiful and cheap
89.
by jacobsenscott 2y ago
They're all good and bad in different ways. You just need to pick one. Ansible is GPL3, opentofu is Mozilla Public License.
90.
by jacobsenscott 2y ago
In abstract CSV is great. In reality, it is a nightmare not because of CSV, but because of all the legacy tools that product it in slightly different ways (different character encodings mostly - excel still produces some variant of latin1,

More ›