nijave

21 ms·

421.
by nijave 9mo ago
>more time and resources than anticipated It doesn't say for whom. That could easily be the legal and marketing department to cover the backlash
422.
by nijave 9mo ago
Really I don't think frameworks have kept up and LLMs are the hammer in Law of the Hammer.
423.
by nijave 9mo ago
It's still possible to script out codegen. Frequently use Python to generate code like that. Really I'd rather have AI generate a codegen script that deterministically does the struct from schema generation I've had enough in
424.
by nijave 9mo ago
I like "vibe doc reading" and "vibe code explanation" but am continually frustrated with vibe coding. I can certainly generate code but it's definitely not my style and I feel reluctant to put my name on it since it
425.
by nijave 9mo ago
I'm still on the fence about codegen but it's certainly helping explain code quickly without manually step through and providing quick access to docs
426.
by nijave 10mo ago
https://github.com/yarrick/iodine
427.
by nijave 10mo ago
I still struggle with these things being _too_ good at generating code. They have a tendency to add abstractions, classes, wrappers, factories, builders to things that didn't really need all that. I find they spit out 6 files worth of
428.
by nijave 10mo ago
Mostly agree although I think ease of instrumentation is getting pretty good. At least in the Python ecosystem, you set some env vars and run opentelemtry-bootstrap and it spits out a list of packages to add. Then you run your code with the
429.
by nijave 10mo ago
Citus, sort of Cockroach For HA, Patroni, stolon, CNPG Multimaster doesn't necessarily buy you availability. Usually it trades performance and potentially uptime for data integrity.
430.
by nijave 10mo ago
>How often does the typical user write to the DB Turns out a lot when you have things like "last accessed" timestamps on your models. Really depends on the app I also don't think that calculation is valid. Your users aren&
431.
by nijave 10mo ago
Yes and no. Least privilege has existed in databases for a very long time. You need to implement correct DB privileges using user/roles, views, and other best practices. The MCP server is more like a dumb client in this setup. However,
432.
by nijave 10mo ago
>If you as an adult want to eat to excess In fairness, there are plenty of conditions and imbalances that can cause weight gain besides "eating in excess" >don't encroach on my economy airline seat The things are so dam
433.
by nijave 10mo ago
There's transition tech for 4to6 to handle embedded devices, though. Obviously the gateway or router missing proper v6 support is an issue and a bit surprising ubt hasn't done a good job. Even my mid range TP-Link Archer I had 10
434.
by nijave 10mo ago
Landed on 172.16/22 for this reason however it's not uncommon how an enterprise to use all 3 private classes. One place I worked used 192.168 for management, 10 for servers, and 172 for wifi Using 2 different classes has been a pr
435.
by nijave 10mo ago
>I don't use IPv6 because it solves a problem that I don't have At least here in the U.S., my observation has been it's usually a bit faster and has more efficient routes than IPv4. I assume part of that is using newer equ
436.
by nijave 10mo ago
I do wonder how painfully slow the computer would be if you actually did replace the in-kernel scheduler with an LLM...
437.
by nijave 10mo ago
Seems more likely OpenAI or one of the hyperscalers would continue vertical expansion into chips but likely only supply themselves--possibly making proprietary variants only they could use anyway Seems to be the case with CPUs although I kn
438.
by nijave 10mo ago
Isn't there higher barriers to entry than OpenAI and soda? They were able to compete with "commodity" hardware. A DRAM manufacturer would need expensive machines from ASML, right?
439.
by nijave 10mo ago
Other nations would likely step in before letting China corner the DRAM market
440.
by nijave 10mo ago
Does hot plugging work right yet? Was quickly discouraged when KVM caused crashes and the open issue said "you're holding it wrong, buy edid emulators"
441.
by nijave 10mo ago
Presumably they're talking about the egregious price of NAT on AWS. It's next to free self hosting considering even the crappiest consumer router has hardware accelerated NAT and takes a tiny amount of power. You likely already ha
442.
by nijave 10mo ago
At least a few times a year. Usually looking for old orders either model number or "how old is this thing" or how much it cost over time
443.
by nijave 10mo ago
It tightened its hold by shrinking its portfolio as the article describes.
444.
by nijave 10mo ago
The assumption is nodes are in different fault domains so it'd be highly unlikely to ruin the whole cluster. LMDB mode also runs with flush/syncing disabled
445.
by nijave 10mo ago
Not too long ago I received an empty package from Amazon but luckily it was a low price item and they reshipped it without fuss. Not sure what you'd do in such a scenario if they tried to fight it
446.
by nijave 10mo ago
Customer centric ended a few years ago
447.
by nijave 10mo ago
Search on page (ctrl-f), not search on internet
448.
by nijave 10mo ago
Reminds me of something I heard at a conference to the effect "10-15 years of <tool> experience is usually a red flag because the only people that have that have been pressing <tool run> button over and over again learning
449.
by nijave 10mo ago
What did the terms of service say you agreed to? That'd be a good starting point to understand their liability, if any, and contractual obligations
450.
by nijave 10mo ago
Part of the advantage is a fully hosted service where you don't need to worry about a control plane managing and scheduling jobs. There's some cost involved there--especially if you're offering hosted runners (you need some c

More ›