They used to have Antigravity and Gemini CLI.

Now they have Antigravity IDE, Antigravity 2, and Antigravity CLI.

As luck would have it, I tried Antigravity for the first time a few days ago.

It was a complete buggy mess - at one point I asked Gemini why it could not use the network despite having network access enabled in the sandbox settings, and it told me that although it had network access, it couldn't use mdnsresponder while running with the built-in sandbox. Like, how well thought out, network access without DNS.

After burning through about 80% of my 5-hour window of credits, I finally just went sandboxless to get the thing running. It hit the limit pretty quickly. I waited until the 5 hour limit was up, and found the 5 hour window had morphed into a one week window, still drained of credits.

I thought at least I can keep on using Gemini CLI until Google figures out this Antigravity thing. Oh well.

My experience with any built in sandboxing for these command line tools has been awful.

What I've done instead is built a script to create a disposable virtual machine (using incus to manage it).

And then I just run the CLI inside the virtual machine and delete the vm at the end of each day.

Same, this has been a challenge since my development machine also has access to banking/personal sensitive data. I would really like to run with `--dangerously-skip-permissions` (or equivalents) without too much worry.

Local VMs are heavyweight but useful if you are sandboxing an entire IDE/GUI app like Cursor. With containers it's somewhat annoying to share local files - Distrobox helps with GUI apps and mounting the home directory but loses sandboxing. I have been curious about Flatpak/bubblewrap, but haven't had time to try it.

For now I've settled on containers, but I would like to shift to a remote VM like I have at work.

I built a pi extension. Pi repo has an example extension that uses anthropics sandbox which is a total buggy mess. (To be clear, that's anthropics sandbox itself, not the pi extension wrapper which is fine)

I dug into it a little bit to see about improving things there, but decided to write a minimal version that better suited my needs instead.

I'm curious why seemingly none of those projects tried using browsers JS/wasm execution as a sandbox instead

You have not tried Antigravity yet?

> or for people under a certain age

I've been waiting for the "Google decides kids shouldn't vibe code" headline over the Antigravity age verification shenanigans.

> Blocking the right car lane for a drop off is perfectly legal outside of No Stopping zones.

In which municipality? In most cities and states in the US, it is illegal to obstruct a roadway. Taxis may get some carve-outs for loading and unloading disabled people, but usually, even taxis are supposed to pull over before stopping for a passenger.

One of the often-overlooked terms of use that people ignore when accessing a payroll processing / human resources website is that the company providing those services can "share" your data with "trusted partners" which essentially allows them to move your income and other personal information to other entities which, in turn, sell that information to anyone who is willing to pay.

It's one of the more abusive uses of click-through agreements - in order to get paid, you have to login and setup your payroll information, and in order to login, you have to click through and agree to these terms, and there is no way to opt out during or after the process.

Whenever you see a "high-fat diet" study, you can usually translate that to "hypercaloric diet" study, and this one is no exception.

One fly food was 8% glucose (.308 kcal/g) and the other was 8% glucose + 20% coconut oil (.308 kcal/g + .7136 kcal/g = 1.0216 kcal/g) leaving the high-fat arm with 332% of the caloric density of the standard fly food diet. The discrepancy clouds how much of the observed effect is due to the coconut oil, and how much is due to the difference in caloric density.

I looked into how well fruit flies self-regulate caloric consumption, but all that I could find was related to carbohydrates and protein. It would seem that fruit flies don't typically eat much in the way of fats, presenting another confounder in the results.

I’m curious if you consulted with an attorney about that? I’ve heard the opposite from people looking to move from Chicago to CA. Does your former employer have a nexus in CA?

Using Florida as an example, if your contract was signed in Florida, your former employer is in Florida, and your case is tried in Florida, the courts aren’t going to pay any regard to California law, and you can be found liable for breach of contract and damages. Correct me if I’m wrong.

SB 699 (passed in 2024) made noncompetes illegal to enforce against against California workers, regardless of where the noncompete was signed. It's very unambiguous on the topic. However, as you correctly pointed out, that involves getting your case in front of a California judge, which you best bet the company suing you for violation of your noncompete is going to do their damndest to avoid.

However, under the wording of SB 699, even if you get sued for violating your noncompete in, say, Michigan, and lose, you can then sue that company, in California courts, and as long as California has jurisdiction (which they probably do, given most businesses don't completely avoid doing business in the largest economy in the country), you can successfully sue them for suing you over the noncompete.

The agreements don’t have damages (other than the loss of deferred compensation as I noted).

The normal way to use the agreement is to seek injunctive relief. That would have the be in CA, where no judge will allow it.

And they did informally confirm they could not prevent me of taking a competitive job, which my employment lawyer confirmed.

I have been wanting to get my genome sequenced for years, and had been thinking 23andme might be one of the better options because of the possibility of invoking the CCPA to get my data deleted after sequencing. Never did it because I wonder if they sell your info to some third party the second it comes off the sequencer, and also because I'm skeptical that they would fully comply with a deletion request.

For people who would like to get their DNA sequenced but are actually concerned about privacy, are there any better options?

I guess it's just my programming instincts, but I just immediately think of the possible worst case scenarios and how strong are the guarantees they're prevented.

Dividing by X? I immediately think what if X is zero. Dereferencing X.Y ? I think what if X is null / nil. And so on.

So when it came to the DNA, I was hesitant to do it, since your DNA can wind up in all kinds of databases. And it turns out I was probably right.

What you could have done is sent in the information anonymously, or under a fictitious name. You can still use an email address and log in and see the results. Or you could use someone else's name from another country (with their permission), but then if that person ever gets in trouble, the DNA evidence might somehow implicate you (such as the guy with the last name NULL who got a lot of parking tickets LOL). A couple months ago I actually did submit with heritage.com and 23andme for a friend, so I think there was no place where you had to provide ID or something.

"Sequencing" make it sound like you want 100% of your SNPs scanned, that is not something that 23andme does afaik.

This is a "shotgun" chip approach which sparsely samples your genes in an economic way. They will concentrate on areas of more interest and so on, in a way that is useful for genealogy for example.

Until you can do it with a kit at home, probably none

Shared hosting providers have had such installers, typically as part of cPanel, for decades. An example would be Fantastico https://www.wikipedia.org/wiki/Fantastico_(web_hosting)

You’re not going to get per-request billing, but shared hosting is typically as cheap as you’re going to get for “always-on” web hosting.