espes

I'm probably one of the people you've talked to. For all his flaws, and there are many, I would give a lot for any team I've been part of since to have someone even half as good as him.

You can self-host zoom: https://support.zoom.us/hc/en-us/articles/201363093-Deployin...

Comcast advertises very clearly that unlimited is an additional $50/m: https://dataplan.xfinity.com/unlimited/

In the UK there are several 2-3 hour train routes that are routinely standing room only. It's not a big deal.

You can choose to enable e2e on Messenger

If you're worried about that Messenger supports end-to-end encryption, just hit the "Go to Secret Conversation" button.

What are you talking about? The United States is one of the only countries in the world where it's possible to walk into a store and walk out with an unlimited prepaid sim, and the only others have a much higher population density.

> I personally do not believe that depth generated purely from deep learning can be used as input to photogrammetry anytime soon.

6d.ai uses depthnets in its mobile photogrammetry pipeline. demo: https://twitter.com/mattmiesnieks/status/1106722396889702406

Intel chips don't support LPDDR4

> I've seen no statements from Facebook about how the data is used, how it is stored, and how long it is retained.

https://www.facebook.com/help/1723057061256937

self-driving minivans with human safety drivers

Why do you think that? There's been way more bugs in V8 over the years than the NaCl core sandbox.

It's ActiveX, but completely secure and open source. Sounds like a good idea.

Regardless, it's pretty much already been killed. Too bad it's taken 8 years for a web standard (WebAssembly) to match it.

Because it uses PNaCl

Planes and trains still have human pilots because they're very cheap in comparison to operational costs.

It links agains the Adreno-GSL, part of the GPU driver, not the GNU Scientific library.

> but if you're going down the highway and the AI decides to suddenly swing the wheel far right as quickly as possible, no human being would have reaction time good enough to recover without crashing

"The EPS controller in the car limits the torque to a very small amount, so regardless of the message, the controller cannot jerk the wheel" [1]

[1] https://github.com/commaai/openpilot/blob/master/SAFETY.md

> It seems incredible that this is the source code for the $999 product originally destined to be available this year.

It's not

demostration: https://www.youtube.com/watch?v=S4Ns5wla9DY

paper: https://www.usenix.org/system/files/conference/usenixsecurit...

You are exactly describing Apple CarPlay and Android Auto

Heh, this 'feature' was the intended solution of a CTF challenge 3 months ago: https://github.com/ctfs/write-ups-2015/tree/master/9447-ctf-...

First release was ~2006

It's a tragedy that it has been almost a decade and it's still just as effective -_-

It's been on textfiles.com since 2001

But it sounds like your model isn't a general graph? DAG and tree isomorphism is in P (and pretty straightforward), and so is tree edit distance...

Why is general graph isomorphism needed for that?

Amusingly, this complete hack using the aesenc instruction claims 0.08 cycles per byte on haswell: https://github.com/gamozolabs/falkhash

> As a related note, the best predictor of university physics scores is high school English scores.

That's pretty counter-intuitive. Can you give a source?

Shout out to the Internet Archive, a non-profit close to what the author describes. Unfortunately it turns out book scanning is expensive and rights-holders still don't like it.

https://archive.org/details/texts&tab=about

Is there any incentive to host nodes?

The actual graphics core has only elementary differences from the retail chips [1]. The differences were mainly in integration (merging in a northbridge, tweaks for UMA, ...)

1. https://github.com/espes/xqemu/blob/xbox/hw/xbox/nv2a.c#L708