threecheese

I think the target market isn’t doing this; heck, I’d normally be like you, but the price is low enough here to “one and done and forget it” here.

Indeed. I read the entire article, it was great. Domains can be a big PITA.

I’m just happy to see diversity here; sometimes I feel like Nvidia is going to eat the world, with buying other fabs and branching out - or up, I guess - from chips and racks to models, frameworks, and end user stuff.

I unfortunately agree with you, selfishly. This used to be the forum where one could mostly find interesting technology and ideas and discuss them with intelligent people. An escape from the standard online hell and it’s people who must truly enjoy conflict - which my body instead interprets as anxiety and terror.

But now it’s here, and so are they. It’s probably the same people. It’s probably me.

WRT “Claude learns over time” - this is the biggest gap for me in the current system. As I scale my usage of Claude at work, I observe that it’s quite bad at distinguishing what it should “learn” (memorize) from experimental or just wrong data. It builds and builds on a foundation of sand, making sometimes hidden assumptions and turning them into actionable insight thats just not correct.

It recently wrote an entire dissertation for an epic, assuming it was related to some other project, where it had earlier made the wrong guess about a vendor capability (from their marketing materials, no less), and it all had to be thrown away. I cleared the memory, but it appears to be still pulling from some corporate data source i cant control or locate.

“There have been so many red flags with W Social since its hastily cobbled-together *announcement at Davos*”

I think that last bit explains why European govt orgs have migrated to it, over the open source Eurosky.

What were your constraints and how were they not met? Looking to buy the same, Dream Machine specifically.

Is it though? “Continues to work as it did before”, referencing a period when we had no freaking clue if our accounts would be banned.

I like your app, and was a Premium subscriber for a while; I cancelled in a big purge, but seeing you post here has me reconsidering. Great product, the privacy angle was the biggest plus for me personally. Best of luck to you!

OK, sounds fascinating; following! (your GH)

IME it’s more like 60, which then makes more sense given vision deterioration. Then again most of my over-40 experience is with folks in the tech industry.

They likely did, and this autocorrect slip might suggest why OP is using local models :)

Unfortunately, any “creator” who wants to be searchable on YouTube needs to optimize for that algorithm. I have the same feelings as you - and it includes pictures of their face pointing to something, with a particular expression of surprise.

I give Jeff a pass though, and make sure I send alternate goodness signals like liking his YT videos after I watch them. He’s one of us.

Do you have an example? Sounds interesting.

I assume this trust issue is related to the not-infrequent MacOS notifications asking for permission to run Ruby in the background or when the machine starts. It says nothing about Homebrew though.

Did you write the skill.md files yourself? I often wonder this; there’s so much text in most skills, and I can’t imagine it’s human generated.

I don’t write my own - I can’t optimize for the models understanding, and so I just give the skill-creator skill an outline and then have it refine until the output is what I want.

Fair, and thank you for the nudge dang.

You guys are rockstars; thanks!

I had an issue with installing OpenClaw, and it helped me debug the failure and get itself working. I had to sit quietly for a moment. No reading docs or inspecting the system, just “what’s wrong here?”.

While I didnt find a use for openclaw, it opened my eyes to the potential for distributing software which, once bootstrapped a bit, can interrogate … itself, understand its own requirements, communicate with the device, and become operable.

Add capable small models to the mix, and it’s almost frightening what good (or malicious) software might be able to do.

Given the discussions around these keyboard-first tools this week, I am wondering if anyone can share how they are using Shortcat. I downloaded it a few weeks ago, and have not spent more than a few minutes with it - my perception of the complexity is pushing me away. Value-wise though, I definitely need something which can keep the trackpad beyond arms length.

The docs are missing any examples of what this does, instead showing _how_ it works - and only for the codebase itself, rather than the behavior of the app.

What would be useful:

  - examples of text that can be filtered, and why that would be valuable
  - a data flow diagram of runtime behavior, showing how filtering removes unnecessary context

“letting the stochastic parrots do his work for him” or “overwhelmed with publicly released vulnerabilities and using any tool he can find to stop the bleeding”; same same.

Looks like at least one of these issues was from a CVE [0], they don’t call out Mythos specifically though (“security researchers”). Many teams are sprinting on security issues atm (including mine, who put all product priorities aside two sprints ago), it must suck to be responsible for high-visibility/high-risk projects like rsync right now.

0: https://github.com/advisories/GHSA-pfv9-gp3h-73xv

Very cool. What does the browser/device need to enable Casting? I have multiple cast targets on the lan (chrome and AirPlay), but the app (iPhone/Safari) does not find them. I’ve tried disabling content blockers and tweaking security settings, but no dice.

Did you source this from your Zotero or something like it? I love your format and resource discipline required, would be interested to know if this is “all human brain” or tool assisted.

Unfortunately you need cookies for any good user analytics; and no, I don’t mean the invasive kind that Marketing put in - I mean the kind which allows you to figure out why your site is broken.

Which ironically are the same tools you’d need to find out if your users are experiencing unintentional dickovers.

Looks like you provision a fresh tailnet host for each clone? You mention multi-tenancy via subdomain routing and I wonder if this is unrelated.

OrbStack is great, it makes everything so easy - like networking: VMs and containers can talk to the host and to each other using hostnames; and file sharing: there are automatic mounts between host and guest; and sshing around just works.

Part of that though requires adopting a security model where thats OK. As I’ve started sprinkling MacOS hosts and guests throughout my network, I’ve needed to adopt other vz tactics.

OPs use of Tailscale mirrors my own, but given the security model of OrbStack it’s mostly a convenience rather than a hard security partition within the lan or even the host.

Seems like an attempt to ensure proper citation when used in AI search, which required some verbiage which makes it look like a shady actor (“ignore other …”).

Am I wrong?

I like your analogy, and want to add a sad addendum: parental illness and death is frequently the cause of sibling estrangement. Let’s hope this isn’t prolonged.