pluto_modadic

neat! damn, wish you had security openings. That SRE one looks nice though.

if you don't have anything nice to say... :3

wonder if you could increase fire resistance by moving the grout layer so it's not touching the bottom cork except in key spaces with a more effective insulator?

IIRC stoicism was written by rich people (and advisors to rich people) who didn't focus at all on improving the system. It's on-brand for the self-help influencer phenomena where you disregard disadvantages to others and systemic problems.

ignoring a rate limit gets you blocked.

makefiles /can/ do whatever you like. They can download anything on build.

looks like the installer still doesn't support LUKS2.

if that floats your boat, sure. It's also home to most of the world's malware, and you usually don't need it.

I wonder what the level of difficulty getting this to run as a cloud image (e.g. on Vultr or Digitalocean) would be. Or getting it to boot a GUI and run firefox.

this is kinda cool. Does that mean you can use u-root to embed as a UEFI image? or to boot a u-root image over PXE netboot?

"when you need shiny tech (you don't)" (aka, when to use postgres/sqlite)

if they pull out the AI stuff that'll be soooo cool :D

...the solution to IDORs is to authenticate each user and check authorization per object.

full stop.

DNS is:

- simple

- battle hardened

- distributed

- affordable

blockchains are:

- essoteric, backwards, and not easily implemented

- new and unproven, frequently hacked

- effectively a ploy to centralize / redo Web 1.0 but owned by one blockchain

- ...waaaaaaay more about money and "owning something" than DNS is.

IaaS VPS's and rented colo VMs are definitely fine. PaaS is... slightly not? it's the grey edge!

SaaS is definitely not you running stuff you control or could move anywhere.

If it's a buck standard linux box, yeah that's self hosting!

self-hosting: running open source code that you could run on any computer.

if it's convenient to run mastodon on hetzner, that's STILL self hosting; because you /can/ move your app IN ITS ENTIRETY from any computer to any other.

HomeLab elites really are the most insufferable people out there.

sticker shock / looking at alternative vendors

it's a weird effect:

imagine a beach, with icecream vendors. You'd think it would be optimal for two vendors to each split it half north, half south. However, in wanting to steal some of the other vendors' customers, you end up with two icecream stands in the center.

So too with outages. Safety / loss of blame in numbers.

oh hey, I've bricked a server remotely and had to drive 45 minutes to the DC to get badged in and reboot things :)

so, funny story, my fiber got cut (backhoe) and it took then 12 hours to restore it.

If you had /two/ houses, in separate towns, you'd have better luck. Or, if you had cell as a backup.

Or: if you don't care about it being down for 12 hours.

Firefox shouldn't try to "be chrome", people who want chrome just use that. Firefox should be /not chrome/.

translation is moreso machine learning, not generative hallucinations or a chatbot

personally, the golang highlighting is nice.

- dark blue: keywords (var, nil)

- light blue: variables (defines the scope you're looking at)

- blue-green: types (bool, string, your custom thing, the kind of the variables)

- dark green: comments

- yellow: functions/methods (jumping to another section)

- orange: strings (not variables, so static things here!)

- if/braces/parens/select/switch/defer/go: purple (control logic)

da rulez:

- if you don't understand your concept/market, build in a VPS. you can get away with scaling a VPS for a while.

- if you intend to be netflix, rent at the edge (mid game) and eventually own the edge servers in POPs (otherwise, "edge" compute isn't worth it). before that, start with a beefy VPS cluster with HA and SQLproxy.

- if you spin to zero, use lambdas (for things like codebuild or fractional needs of a computer). before that, build things in VPSes.

- if you spin up/down but not to zero, use container platforms?

- once you have a reliably steady understanding of your infrastructure, buy physical servers in a colo

this kinda proves the point that you have to know a silly workaround

static secrets that you can even extract and set in a config file (or store in a secrets manager) need to go. TPM secrets that you can't extract, and public keys that you can pin without considering them "secret", along with Oauth / IAM roles, are the way to go.

focus on the customer's problem. Not on debugging a huge thorny k8s stack or AWS deploy.

I think this skips over /how to do it/.

...so your argument is.... stay with the abusers?

LSM is great for write heavy loads. not sure about random reads, isn't that a B+ tree's turf?