dangoodmanUT

> This is marketed as competition. But it’s not. It’s a cartel. Each company gets its own protected territory, and consumers get no choice.

Starlink has me extra excited for exactly this: region agnostic competition.

Get rekt, telecom incumbents

> Don’t try to sell developers the stuff that’s fun, the stuff they want to build. Sell them the stuff they avoid or don’t want to take responsibility for.

I think a lot of people in devtools could learn from this right now

@claudecode

Because the model labs can RL against their own scaffold, which will be borderline unbeatable

With unified memory and such a strong os-hardware integration, one would hope that swap could handle this task

I had to close 3 ads before even half my screen was the article

And then it never was more than half…

The terminal is a pretty bad place to have free form input if you need a separate key bind to paste an image than to paste text…

This felt oddly ai-authored, like the kind of “brain dump then ask ChatGPT to sort it into a post”. Probably because of too many bullet points and bold emphasis

I think this is the argument for UIs - it should be self-explanatory since it's singificantly simpler than an IDE

This clearly AI generated website is such a turn-off.

I get that AI can be good at making websites, and someone might not care to spend a lot of time on it, but a website that looks like a slightly modified version of a generic "make an X landing page" from gpt-5.3-codex doesn't scream "I care about what I just made".

Just go super reductionist like planetscale did and have partially rendered markdown, don't put twinkling stars in the background

Yeah, on top of that bringing in social media politics into it is weird, makes it hard to take this as pure/useful criticism

The iroh team keeps cooking, unreal.

I’m excited to have a weekend to just sit down and tinker with iroh, it’s been on my list for a while. I want to make an overlay network like nebula with it

Optane was crazy good tech, it way just too expensive at the time for mass adoption, but the benefits were so good.

Looking at those charts, besides the DWPD it feels like normal NVMe has mostly caught up. I occassionally wonder where a gen 7/8(?) optane would be today if it caught on, it'd probably be nuts.

There's something incredibly ironic about a visual tool for designing TUIs...

There are a lot of reasons for this, but typically “same same reason you would use k8s for customer serverless functions”: can’t scale fast enough, too slow to place workloads, not isolated by default, configuration explosion, limited multitenancy support, and so much more

How have you determined that? You can easily push 6GB/s+, sub ms ttfb with networked filesystems, and hundreds of thousands of iops through fuse.

Woah... a mobile processor and enormous bezels... definitely feels like Jobs would have never let this ship

In review, I think it might have been the workflow versioning being strange, and the lack of any heartbeating/crash detection for longer running activities

Thanks, I didn't mean it as criticism, I guess my 5am brain thought the way it was worded almost came off as like "look at our unique idea" which was a pretty common pattern.

I was referring to this post

Been eager for something that wasn’t temporal (egregious overhead and annoying multiple services), but they do write this like Temporal… doesn’t exist. They use a lot of the same pioneered techniques (like “our own context type”) that they do.

go-workflows has always been the good alternative, but I’m sure dbos is a bit better supported. Dbos always had some weird gaps (I don’t remember why exactly, I just remember saying “oh well I can’t use this then” more than once), but maybe they’ll close them with the go sdk

11 million hours of data is a lot, did you have to synthesize it at all, or was it purely collected?

> because redis began to hit 120 ops/sec and I’d read that any more would cause issues

Suspicious. I don’t think I’ve ever read anything that says redis taps out below tens of thousands of ops…

I can appreciate the effort, but the UI is indistinguishable from tiktok.

> There are no great FaaS options for running GPU workloads, which is why we could never go fully FaaS.

modal.com???

Love this, such a better method

Was hoping this was about network tunnels…

This is awesome. I’ve been experimenting with letting models “play” with different environments as a strong demo of their different behaviors.

hell yes, finally

> We’ll miss the sleepless wrangling of some odd bug that eventually relents to the debugger at 2 AM

No the fuck we wont