lxgr

39 ms·

961.
by lxgr 8mo ago
What annoys me most about OpenClaw after trying it for a few weeks is that it cosplays security so incredibly hard, it actually regularly breaks my (very basic) setup via introducing yet another vibe coded, poorly conceptualized authenticat
962.
by lxgr 8mo ago
Personally, if I could run capable-enough inference on hardware I control, and could rely on the harness asking me for mechanistic confirmation before the agent can take consequential actions, I'd do it immediately.
963.
by lxgr 8mo ago
That's a fair point, and I guess the marketing problem here is intrinsic: If the problem is trivial, off-the-shelf solutions abound; if it's idiosyncratic, almost nobody will be able to relate (as you can't assume that people
964.
by lxgr 8mo ago
Oh, I thought that support was only present in some iOS 13 beta and then disappeared again? (At the OS/driver level; I’m pretty sure the hardware supports it.) If it’s really available, that would be amazing!
965.
by lxgr 8mo ago
The core new functionality here seems to be MASQUE proxy support, which is genuinely useful and probably not possible to provide in a plugin. Of course the UI itself could be extracted into a plugin, but I'm not really a fan of the pat
966.
by lxgr 8mo ago
I just tried it. It seems to be based on MASQUE and provided by (at least) Fastly.
967.
by lxgr 8mo ago
As another Firefox user, I much prefer modern browser-centric solutions like iCloud Relay over VPNs, which seems like the wrong layer for what I'm using them for both in terms of implementation complexity and achievable privacy. Happy
968.
by lxgr 8mo ago
The trick is to use a plausible Slack message as your sudo password :)
969.
by lxgr 8mo ago
It’s not available on iOS devices though, right?
970.
by lxgr 8mo ago
For bots?
971.
by lxgr 8mo ago
OpenClaw's core achievement is that it was first, and that's not a moat. The code/product itself is an absolute nightmare of overengineering, riddled with bugs and undocumented behavior changes across versions.
972.
by lxgr 8mo ago
Unfortunately, Telegram is also a no-go from a security/data privacy perspective for many use cases.
973.
by lxgr 8mo ago
I'd definitely agree for normal bikes and e-bikes capped to speeds under 30 km/h, but at least in NYC, these delivery bikes often go ridiculously fast. I don't think they should be considered the same regulatory category. The
974.
by lxgr 8mo ago
Really? I thought API key usage was always billed per token, not via monthly allowances?
975.
by lxgr 8mo ago
> What’s nice is not that it’s just me, but that it’s everyone. This made it sound like you enjoy me being offline, and that seems pretty selfish (as long as I don't annoy you somehow with my Internet connection, and on that, see
976.
by lxgr 8mo ago
Host a “phones off” party, go to a sauna, go for a hike with friends with self control etc., but please don’t hold me hostage (connectivity wise) in a cramped metal tube for your sense of nostalgia. Planes are just about the least pleasant
977.
by lxgr 8mo ago
It’s pretty good, but the latency is inherently high since Viasat is in GEO.
978.
by lxgr 8mo ago
> their ungodly expensive product Do you have any idea how much other satellite operators charge per megabyte or Mbit/s?
979.
by lxgr 8mo ago
Wouldn't a widespread ground-based infrastructure outage also take out (or at least severely degrade) Starlink in the affected region if people were to widely use it as a backup solution?
980.
by lxgr 8mo ago
Depends on the size of the system you need to secure. If kilobytes of storage and very limited computing power works for your use case, you can get very secure (smartcards and secure elements remain essentially undefeated at the hardware le
981.
by lxgr 8mo ago
I think it’s uncontroversial that cheap, unlimited 5G exists in some places.
982.
by lxgr 8mo ago
TFA specifically calls out not wanting to depend on 4G/5G coverage, which is anything but ubiquitous: > It has the advantage of working pretty much anywhere with a view of the sky so no relying on mobile network coverage. I'm a
983.
by lxgr 8mo ago
> Set IPv6 Connection to SLAAC (this is critical - SLAAC must be used, not DHCPv6) [...] Set 'Prefix Delegation Size' to 56 Is this also A UniFi bug, is Starlink doing IPv6 address assignment in a weird way, or is this a normal
984.
by lxgr 8mo ago
It’s extremely rare compared to v4, where it’s more common than not. I haven’t seen it with a single consumer ISP, and why should they use it?
985.
by lxgr 8mo ago
> Honestly ISPs really missed an opportunity to essentially provide IPv6-only as a service This is in fact how many 4G and 5G networks work today. I’m sending this reply via one right now. For wired connections, many DOCSIS cable ISPs us
986.
by lxgr 8mo ago
Firewalls exist in IPv6 too, so hole punching still has a reason to exist as well.
987.
by lxgr 8mo ago
> The Spotify DJ is a recommendation engine. Is it? Having recently tried Spotify's presumably related "prompt based playlists" feature, I've been wondering: Is the "AI DJ" maybe not a recommendation engine,
988.
by lxgr 8mo ago
> NAT Behavioural Requirements for TCP TIL, thank you! I've been looking for this for quite a while after hearing it indirectly referenced recently, but only found host-side specifications for TCP simultaneous open. Do you happen to
989.
by lxgr 8mo ago
Are you even acknowledging my example? Where does it exist in your bimodal model of reality of "my firewall" and "somebody else's firewall"? What provider would you suggest somebody wanting to make VoIP calls on the
990.
by lxgr 8mo ago
Love it when random people tell me whether my use case is legitimate or not without apparently even knowing it exists! Take mobile data connections, for example: Most people don't want to pay for metered (by the byte) inbound traffic t

More ›