tonymet

16 ms·

451.
by tonymet 8mo ago
I question tech community integrity because they will constantly attack Meta or Microsoft for aggressive anti-privacy, Uber for exploitative rates & policies but turn a blind eye to how corrosive & exploitative OF and PornHub are.
452.
by tonymet 8mo ago
we're talking about md files in a git repo, right?
453.
by tonymet 8mo ago
in 100 years ipv4 will be recognized as one of the great discoveries like calculus. ipv6 is a misnomer really. It's a separate, and lesser protocol. Much like other second systems, it was too ambitious and not pragmatic enough. Rather
454.
by tonymet 8mo ago
edit: it's satire. but likely not too far off from the reality in 6 months. > Our process is deliberately, provably, almost tediously legal. One set of AI agents analyzes only public documentation: README files, API specifications,
455.
by tonymet 8mo ago
“Big data” doesn’t have a 5gb memory cap. I’m guessing so many devs started out on 32gb MacBooks that the NEO seems underpowered. but it wasn’t too long ago that 8gb, 1500mb/sec IO & so many cores was an elite machine. I did a lot
456.
by tonymet 8mo ago
IO on a raspi is pathetic. As packaged it’s 100 times slower , with a M.2 hat it’s still 5x slower
457.
by tonymet 8mo ago
Chads never backspace.
458.
by tonymet 8mo ago
very clever and thanks for explaining. for gpu-bound processes, which are common ffmpeg use cases, this is a great approach
459.
by tonymet 8mo ago
cool idea. can you elaborate on IO and how the ffmpeg-server reads blocks from the client? that would seem to be a big blocker
460.
by tonymet 8mo ago
Given the 10x+ productivity rate, it would be reasonable to establish a higher quality acceptance bar for AI submissions. 50-100% more performance, correctness, usability testing , and one round of human review. If a change used to take a
461.
by tonymet 8mo ago
I appreciate you mentioning the mood effects because I've heard those are pretty common too. Also weight gain , bloating and skin issues.
462.
by tonymet 8mo ago
The video was helpful and addressed many of the clinical and ethical concerns I have. Can you consent to a procedure that has a 30+ year impact? Are these treatments completely safety tested if long term studies are not being done? What a
463.
by tonymet 8mo ago
thanks . yeah i had a close family member do the same and it was a pretty long and painful process. it was > 10 years ago but I think the concerns are still similar.
464.
by tonymet 8mo ago
thanks for sharing that's helpful. I've heard similar to more moderate ill effects from the therapy
465.
by tonymet 8mo ago
Employers encouraging egg freezing by offering egg freezing benefits is an abysmal conflict of interest. Employers reap tremendous medium-term benefits and the woman bears all of the long-term risks -- in this case the biggest risk of all
466.
by tonymet 8mo ago
how were the adverse effects during the hormone / endocrine therapy for her?
467.
by tonymet 8mo ago
you can try doing fstrim + Optimize-VMD first . i've found it is good up to 20% savings. On my last run my 75 gb wsl disk saved about 15gb. But wiping and re-installing trimmed about 60 gb
468.
by tonymet 8mo ago
it's illustrative
469.
by tonymet 8mo ago
what if authors marked up AI content with special classifier Unicode characters to clarify how it was used. From full robot, down to minor grammar or markup assistance?
470.
by tonymet 8mo ago
reinvest in your relationships with friends and family. they may have been neglected during your relationship.
471.
by tonymet 8mo ago
maybe it was WSL1 ? WSL2 IO is excellent as long as you stay within the FS. And with VS Code WSL support , all IO & commands are remoted into WSL so there are no Windows/Linux compatibility issues.
472.
by tonymet 8mo ago
this is great and I have a feature request: A "refresh VM" feature that exports your home dir + packages into a new VM instance. I've tried Optimize-VHD but renewing the VM this way frees up disk and speeds up the VM as well.
473.
by tonymet 8mo ago
I mean it's trivial for any attacker to discover admin tasks and know where to place malicious code for the admin tasks to execute it.
474.
by tonymet 8mo ago
Somewhere in early MacOS design guidelines I remember it saying that settings/prompts were a design failure that put a burden on users. Something really has been lost on the design team. I know the landscape is different – more liabil
475.
by tonymet 8mo ago
Admin tasks are public in phabricator so it would be trivial to review chores and place malware in the chore's scope
476.
by tonymet 8mo ago
there's a very active tech discussion on the Wikipedia discord you can join here https://en.wikipedia.org/wiki/Wikipedia:Discord
477.
by tonymet 8mo ago
It's Wikipedia's 25th birthday but their security discipline is still very much circa 2001. No code signing, BOM / supply chain security. Only recently activated 2fa for admins (after another breach). Most admins are anons.
478.
by tonymet 8mo ago
that's a common attack vector -- like leaving malware usb sticks on the ground, knowing an admin will pick it up and insert it. Phabricator reveals the ops tasks that WMF admins perform, so attackers can drop malware in common location
479.
by tonymet 8mo ago
Consumers are overwhelmingly giving consent to surveillance via their apps. So when the government would like to it's a bad thing? why not improve policy so the government has guidelines for doing so, so they don't to conceal the
480.
by tonymet 8mo ago
exactly. These are supposed to be tools and behave predictably. It's maddening when focus is stolen and interrupted by an unexpected prompt. Often for something I already consented to. [edited]

More ›