lima

14 ms·

151.
by lima 2y ago
Could just steal the key at that point.
152.
by lima 2y ago
Only if the key is generated off-key
153.
by lima 2y ago
The hard part is not backing yourself into a corner with the feature tree over time. Takes a lot of practice to avoid, and that's something AI could help with. Meanwhile, what their demo does is trivial.
154.
by lima 2y ago
Agreed, sync modelling in Solid Edge and NX can be really powerful, especially when requirements change unexpectedly.
155.
by lima 2y ago
This is data that criminals already publicly released.
156.
by lima 2y ago
Intent matters.
157.
by lima 2y ago
Thank you for this, I learned a lot from this comment.
158.
by lima 2y ago
The RADOS K/V store is pretty close. Ceph is built on top of it but you can also use it as a standalone database.
159.
by lima 2y ago
Lots and lots of heavy machinery uses Windows computers even for local control panels.
160.
by lima 2y ago
The problem is the long tail. Neural networks are not great at handling edge cases and it's not clear whether they will ever be.
161.
by lima 2y ago
What about consumer apps in Local Data Mode?
162.
by lima 2y ago
GitHub org permissions aren't very granular, especially with an org that is tied to enterprise SSO. Makes it hard to grant certain permission to outside members.
163.
by lima 2y ago
OpenVZ, Virtuozzo and friends definitely weren't secure the way gVisor or Firecracker are. You can still do that and some providers do, doesn't make it a good idea.
164.
by lima 2y ago
The funny thing with k8s is that Google doesn't use it (except GKE, and there's a reason it's one cluster per customer). Their internal tooling scales just fine, but all it shares with k8s is some of the underlying concepts.
165.
by lima 2y ago
As a customer of Equinix Cloud... No thank you. Infrastructure vendors are terrible software engineers.
166.
by lima 2y ago
And in hammocks! They don't touch the ground and therefore don't count.
167.
by lima 2y ago
Much of kitty is written in C, particularly the (very fast) rendering pipeline.
168.
by lima 2y ago
The risk with such rewrites is ending up with a Python 3 situation and an ecosystem split. Sounds like YARA-X is (mostly) a stricter subset of YARA, and it's easy to write rules that are valid for both: https://virustotal.gi
169.
by lima 2y ago
Yara is pretty much the industry standard for detection rules now.
170.
by lima 2y ago
Highly targeted attack on a specific machine. Probably not a threat actor but a red team or some other security researcher.
171.
by lima 2y ago
Use Clickhouse - it's specifically built for this use case (low latency OLAP).
172.
by lima 2y ago
It archives a much smaller part of the internet, it't not an IA replacement.
173.
by lima 2y ago
Why would the U2F-on-Ledger route stop working?
174.
by lima 2y ago
By calling them
175.
by lima 3y ago
Talos appears to rely on existing components to a somewhat greater degree in comparison: https://github.com/siderolabs/pkgs
176.
by lima 3y ago
There aren't even any poor security practices. It's just security theatre and the DOC is overreacting.
177.
by lima 3y ago
Not in terms of specific threat model.
178.
by lima 3y ago
Phabricator is what we used before we migrated to Gerrit. The review UI was very nice, the issue tracker is still best-in-class, but Arcanist was a menace. Much prefer a simple "git push" over that client-side PHP contraption :-)
179.
by lima 3y ago
There isn't, you need to contact them so they can delete the offending objects.
180.
by lima 3y ago
Graphite is neat. If you want something lighter-weight, spr[1] is also worth taking a look at as an entirely client-side "solution" to PR stacking. Unfortunately, it is very hard to build good code review tooling on top of GitHub

More ›