Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
dinvlad
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
27 ms
·
301.
▲
by
dinvlad
4y ago
Those are good points. Still, it’s fairly manageable, after certain adjustments. Also, we’re using the new (Go-based) version of TH that’s both much more performant and validates secrets against endpoints. I suspect their SaaS offering is a
302.
▲
by
dinvlad
4y ago
Awesome, thanks for the heads-up! We will definitely look forward to it, either to replace or enhance our current in-house solution.
303.
▲
by
dinvlad
4y ago
As told earlier, we have thousands of repos. And our teams are thousands of users on GH.
304.
▲
by
dinvlad
4y ago
Thanks, I've already figured out how to run Trufflehog for free on our thousands of repos.
305.
▲
by
dinvlad
4y ago
Nah thanks, I'm already running Trufflehog for free on all of our multiple orgs' thousands of repos. I think we would consider GG if its pricing was acceptable for non-profits though.
306.
▲
by
dinvlad
4y ago
This is definitely a marketing piece. And they charge a LOT for it, so it's not a solution for the common masses.
307.
▲
by
dinvlad
4y ago
I think the fundamental problem is, a lot of orgs just don't care about security, as it doesn't affect their bottom-line. Even breaches are only a temporary hit on the PR. Proper way to address that might just be legislation, with
308.
▲
by
dinvlad
4y ago
I wish hosted GitHub made pre-push hooks available to the public. Would make this a much easier problem with free scanning tools like Trufflehog. Or alternatively, if GitHub Secret Scanning was available to all public repos, instead of requ
309.
▲
by
dinvlad
4y ago
Man, how much Wonder these ads invited back then, purely from the novelty factor. Hard to get excited about tech the same way these days, because it’s all become so ubiquitous and banal, and there’s less and less novelty now. Sounds pessimi
310.
▲
by
dinvlad
4y ago
Too bad distributed DBs are very expensive atm. Their multi-region offerings are only good for companies who can spend thousands on a database each month. I’m actually thinking that with GDPR and similar regional lock-in laws, the future mi
311.
▲
by
dinvlad
4y ago
Could you share a link please? I've only seen similar demos of a Doom game migration on LXD (I believe). EDIT: very interesting, thanks TIL! Here's at least someone from RH playing with it in the context of K8s: https://
312.
▲
by
dinvlad
4y ago
The best futuristic feature by far I've heard of is live-migration of workloads between hosts (read: your compute and memory gets transferred to another VM and resumes without having to restart). This would be amazing for, e.g. auto-sc
313.
▲
by
dinvlad
4y ago
High salaries though? :-)
314.
▲
by
dinvlad
4y ago
Out of curiosity, which TC ranges do you normally offer?
315.
▲
by
dinvlad
4y ago
Highly-paid jobs? Would be good to hear some examples.
316.
▲
by
dinvlad
4y ago
This. The more we like to think about things and be creative, the more our minds go to any number of imaginary situations, many of which can be negative. It's kinda the opposite of the age when we were all hunters and gatherers and our
317.
▲
by
dinvlad
4y ago
Same here - since upgrading to NVME SSD my boot times were 3-5 sec (!) with Win10, and 5-10 sec with Win11. I think there's at least something they're doing right, and normal Linux desktop has a long way to go still (mine still ea
318.
▲
by
dinvlad
4y ago
Mind => blown. There are very few original ideas, it seems. Just slightly better execution maybe.
319.
▲
by
dinvlad
4y ago
I think the silver lining is at the time of their peak Sears etc. seemed unstoppable. And it took only a few decades for that to be undone. And that is basically guaranteed to happen to any behemoth simply by their inertia and inability to
320.
▲
by
dinvlad
4y ago
Or start your own bootstrapped SaaS - the only investment is time and effort. These are some of the most enduring businesses, imho.
321.
▲
by
dinvlad
4y ago
Now I don't understand your comment, sorry :-)
322.
▲
by
dinvlad
4y ago
I think lawyers probably miss a lot of these technicalities (which explains why they can require something as technically unreasonable as Schrems II), but I just wonder if someone more technical can actually push for something like that, to
323.
▲
by
dinvlad
4y ago
Tbh, I've been trying these same instructions for Pop_OS! and that's why it didn't work. Maybe Arch itself is much better in that regard.
324.
▲
by
dinvlad
4y ago
Unfortunately, this is what I've heard happened to a medium-sized org at least. It's less about the support itself, but more so on the strategic/partnership side. But I understand, such things are expected and change over tim
325.
▲
by
dinvlad
4y ago
Tbh, k8s "system" namespaces also consume quite a bit (particulary if you wanna run a minimal system) - at least 0.5 vCPU on each node and something like 0.5-1GB of RAM. This is only important for the smallest systems, but still i
326.
▲
by
dinvlad
4y ago
Please, don't give ideas to whomever among their lawyers reads this later :-). We don't need another Schrems II..
327.
▲
by
dinvlad
4y ago
I think revenue comparisons with non-indie companies miss the point, because a lot of Indiehackers live a different lifestyle altogether, where their "Ramen profitability" of $1-2k MRR can already allow them to live very comfortab
328.
▲
by
dinvlad
4y ago
Exactly, a lot of them live in remote areas like Bali with low costs of living, and have better housing etc. compared to a lot of salaried employees here.
329.
▲
by
dinvlad
4y ago
He built all/most of those in public, btw. Which is part of how he was able to develop a lot of well-deserved trust. This strategy has a very good reasoning, but I do think it's hard because not everyone is able to develop a huge
330.
▲
by
dinvlad
4y ago
Oh, after trying my hand at a custom scheduler on bare-bones VMs, I dreamed of using K8s for bin-packing too, but in reality they still end up just running a bunch of Java services that are extremely poor at RAM reporting, and so really are
More ›