Diseconomies of scale in fraud, spam, support, and moderationdanluu.com4 points·by Liriel·8 giorni fa·3 comments
How to Set Up and Deploy an OpenClaw AI Agent on a VPS3hcloud.com3 points·by Liriel·15 giorni fa·0 comments
Tracing a Slow Request Through Your Django Appblog.appsignal.com1 points·by Liriel·3 mesi fa·0 comments
WireGuard VPN developer can't ship updates after Microsoft locks accounttechcrunch.com9 points·by Liriel·3 mesi fa·3 comments
Monitoring Express Route Performance with AppSignalblog.appsignal.com1 points·by Liriel·3 mesi fa·0 comments
Developers Should – and Shouldn't – Use LLMs in Our Developmenttighten.com3 points·by Liriel·3 mesi fa·0 comments
What Metrics to Monitor in Your Vibe Coded Appblog.appsignal.com3 points·by Liriel·3 mesi fa·0 comments
Study: Sycophantic AI can undermine human judgmentarstechnica.com1 points·by Liriel·4 mesi fa·0 comments
Monitoring Your Node.js App Health on Fly.ioblog.appsignal.com1 points·by Liriel·4 mesi fa·0 comments
Liriel·25 giorni fa·discussSo far I haven't had any trouble, and they have servers right where I need them.
Liriel·26 giorni fa·discussI don't think anyone can guarantee anything here, but I am going to bet that they will be a better-priced alternative.
Liriel·26 giorni fa·discussPrices are going insane, and I am going for an alternative provider in the US.I have 5 app nodes, 2 background workers, 1 DB replica, and 4 dev/staging instances.On new Hetzner US pricing:5 x CCX23 (app nodes): $514.95/mo 2 x CCX13 (workers): $101.98/mo 1 x CCX23 (DB replica): $102.99/mo 4 x CX33 (dev/staging): $39.96/mo Total: ~$759/mo or $9,108/yearOn 3HCloud with equivalent specs:5 x 16 GB 4 vCPU dedicated: $160/mo 2 x 8 GB 2 vCPU dedicated: $32/mo 1 x 16 GB 4 vCPU dedicated: $32/mo 4 x 4 GB 2 vCPU shared: $24/mo Storage ~250 GB total (SSD smart): ~$7.50/mo Total: ~$255/mo or $3,066/yearI will let you do the math.
Liriel·8 mesi fa·discussYou are right, but in practice, a lot of things happen.Docs that get generated end up in a huge PR that people glance over.Sometimes the reviewers aren't senior enough, so they don't even see the issue, etc.
Liriel·10 mesi fa·discussMy pet peeve is people who use LLMs to generate code, never check whether it works, and then submit a PR.As if open-source maintainers don't have enough chores.