valleyer

> Plus, if you live in a red state, they've likely refused to take the subsidies

No. Individual states can refuse Medicaid expansion, but that does not have any bearing on the health insurance marketplace / premium tax credit ("subsidies"), which states cannot opt out of.

If you're all staring at the TV, you can at least share your thoughts on the thing you're all watching together. If you're all staring at your phones, your minds are in different places. It doesn't serve the same role at all.

The federal government spends $850 billion per year on the DoD. Why can't it find money in there to buy toiletries and spaghetti for the soldiers and families stationed overseas? Why should communities have to pitch in (beyond their contribution via taxes)?

In this case, you cannot simply force Congress to appropriate money to a reincorporated CPB -- unless you were to get a second ruling from a judge that the rescission was unconstitutional.

The Trump EO was deemed unconstitutional because he specifically called out that it didn't like the "left-wing propaganda" (his words) in PBS/NPR programming. Congress's rescission is ostensibly for budgetary reasons -- even if we all know in our heart that they were following Trump's orders.

What we can do is elect a Congress that will revive the CPB. Here's hoping.

The CPB, the legal entity that the government actually funded (and which in turn supplied some of the funding for PBS/NPR and its stations) had its funding rescinded by Congress (under HR4 last year), and has since shuttered.

It's not clear how, even under that recent ruling, that rescission will be undone.

Thanks for the info on the VuPoint. In general, I actually don't mind capture cards not handling deinterlacing, since (as long as I correctly understand the format it's spitting out) I can handle it myself with e.g. ffmpeg. Color problems are a dealbreaker though.

I've looked at going down the RF capture route but haven't dipped my feet in yet. Maybe one day.

Looks like Santa Clara is using the VuPoint Digital Video Converter -- not familiar with this one, so I'll have to look into it.

https://www.sclibrary.org/services/other-services/vhs-conver...

Super, super cool initiative.

I would be interested to know what capture hardware they're using. As someone who took on this project for my own family's videos, I ended up using the Canopus ADVC-110, which captures composite (or S-Video) NTSC (plus stereo audio) and generates DV, which you can then capture over FireWire.

It worked well for me since it didn't require any non-built-in drivers on macOS (though I hear Tahoe drops FireWire support entirely -- boo), and it therefore was easily interoperable with FFmpeg, VLC, and custom AVFoundation code I wrote.

Unfortunately, I don't think the ADVC-110 is made anymore, and my experiments with various USB Video Class devices (which would be similarly interoperable), mostly based on MacroSilicon chips like the MS210x, were utter failures in terms of quality.

> How do you make a contribution to reduce the debt?

Hilarious.

That's not a distinction that actually exists in the real world. This makes me wonder what other made-up distinctions they are claiming in industries I'm less familiar with.

That's not a distinction that exists in industry.

Certainly the OS can't optimize it to a no-op, since `dd` makes separate read and write syscalls.

I suppose your `dd` implementation itself could do so, but I don't know why it would.

Thoughts on that: (1) you'd need a way to visualize the magnetic fields, (2) the data is frequency modulated, (3) due to helical scan, the video field lines do not line up evenly one over the next as they did so nicely in the Laserdisc / CED (there'd be a skew).

So I don't want to say it's impossible, but I think it would require a lot more creativity.

Nit: ISPs don't have access to path names / query strings of browsing activity. Those are encrypted by TLS. (They're also not part of "DNS".)

Honestly, <span content-filter-level="adult">fuck</span> that.

What's a computer?

Addiction. Misappropriation of others' secrets. Perverse incentives.

I don't think it's inherently impossible, but Apple at least seems to do a really poor job at local cache management on iOS for Photos and Messages attachments. I am constantly amazed to find my non-tech-savvy relatives deleting stuff from their phones to free up local storage.

As long as a mutation isn't strongly maladaptive, it can evolve prior to its being useful.

No -- use doors.

The first sentence of the article should make that clear. In any case, it's a pretty well-established abbreviation in US policy discussions.

You are essentially describing the system that Codex (and, I presume, Claude Code et al.) already implements.

If that part were running sandboxed, then it would be impossible for it to contact the OpenAI servers (to get the LLM's responses), or to spawn an unsandboxed process (for situations where the LLM requests it from the user).

The LLM inference itself doesn't "run code" per se (it's just doing tensor math), and besides, it runs on OpenAI's servers, not your machine.

Here's OpenAI's docs page on how they sandbox Codex: https://developers.openai.com/codex/security/

Here's the macOS kernel-enforced sandbox profile that gets applied to processes spawned by the LLM: https://github.com/openai/codex/blob/main/codex-rs/core/src/...

I think skepticism is healthy here, but there's no need to just guess.

Is it asking you permission to run that python command? If so, then that's expected: commands that you approve get to run without the sandbox.

The point is that Codex can (by default) run commands on its own, without approval (e.g., running `make` on the project it's working on), but they're subject to the imposed OS sandbox.

This is controlled by the `--sandbox` and `--ask-for-approval` arguments to `codex`.

> If you look at the security measures in other coding agents, they're mostly security theater. As soon as your agent can write code and run code, it's pretty much game over.

At least for Codex, the agent runs commands inside an OS-provided sandbox (Seatbelt on macOS, and other stuff on other platforms). It does not end up "making the agent mostly useless".

I'm thinking that was just a typeface (looks like Profil) that was available to their printer. It also shows up in the word WARRANTY in the back of the manual:

https://s3data.computerhistory.org/brochures/apple.applei.19...

Doesn't `git merge -s ours` do this?

    This resolves any number of heads, but the resulting tree of the merge is always
    that of the current branch head, effectively ignoring all changes from all other
    branches. It is meant to be used to supersede old development history of side
    branches. Note that this is different from the -Xours option to the ort merge strategy.

It drives me nuts that local governments in the US continue to use Twitter/X to disseminate communications, despite having perfectly good web sites of their own.