Hacker News new | past | comments | ask | show | jobs | submit | bhassel's comments login

A few quotes from that document (from 2018):

Each time a phone connects to a cell site, it generates a time-stamped record known as cell-site location information (CSLI). Wireless carriers collect and store this information for their own business purposes.

[...]

With just the click of a button, the Government can access each carrier’s deep repository of historical location information at practically no expense.

[...]

With access to CSLI, the Government can now travel back in time to retrace a person’s whereabouts, subject only to the retention polices of the wireless carriers, which currently maintain records for up to five years. Critically, because location information is continually logged for all of the 400 million devices in the United States — not just those belonging to persons who might happen to come under investigation — this newfound tracking capacity runs against everyone.

[...]

The Government and JUSTICE KENNEDY contend, however, that the collection of CSLI should be permitted because the data is less precise than GPS information. [...] The location records [...] placed [Carpenter] within a wedge-shaped sector ranging from one-eighth to four square miles.

[...]

While the records in this case reflect the state of technology at the start of the decade, the accuracy of CSLI is rapidly approaching GPS-level precision. As the number of cell sites has proliferated, the geographic area covered by each cell sector has shrunk, particularly in urban areas. In addition, with new technology measuring the time and angle of signals hitting their towers, wireless carriers already have the capability to pinpoint a phone’s location within 50 meters.


The google TV chromecast does support wired ethernet, but you need to buy the separate adapter for it.


ArsTechnica article discussing this paper (and one other), with a bit of background information: https://arstechnica.com/science/2021/01/two-ways-of-performi...


According to their visitor stats [1], April is busier than the five months preceding it, so I suppose it's relative.

1: https://www.nps.gov/yose/planyourvisit/visitation.htm


> In another break with industry practice, the chip won’t be sold on its own, but will be packaged into a computer “appliance” that Cerebras has designed. One reason is the need for a complex system of water-cooling, a kind of irrigation network to counteract the extreme heat generated by a chip running at 15 kilowatts of power.

15 kW, yikes.


So, Azul Systems 2.0? clever(ish) hardware with good(ish) results, for too much $$$ for anyone to actually buy?


The main difference is that what Azul built had more limits - once you run your workload well enough, there is little incentive to have more compute power.

When it comes to ML, the more compute power you throw at it, the better.


That was briefly touched on in the article:

> One approach that has been discussed recently is to create a pointcloud using stereo cameras (similar to how our eyes use parallax to judge distance). So far this hasn’t proved to be a great alternative since you would need unrealistically high-resolution cameras to measure objects at any significant distance.

Doing some very rough math, assuming a pair of 4K cameras with 50 degree FOV on opposite sides of the vehicle (for maximum stereo separation) and assuming you could perfectly align the pixels from both cameras, it seems you could theoretically measure depth with a precision of +/-75 cm for an object 70 meters away (a typical braking distance at highway speeds.) In practice, I imagine most of the difficulty is in matching up the pixels from both cameras precisely enough.


From https://www.bostondynamics.com/handle :

Handle is a robot that combines the rough-terrain capability of legs with the efficiency of wheels. It uses many of the same principles for dynamics, balance, and mobile manipulation found in the quadruped and biped robots we build, but with only 10 actuated joints, it is significantly less complex. Wheels are fast and efficient on flat surfaces while legs can go almost anywhere: by combining wheels and legs, Handle has the best of both worlds.

Handle can pick up heavy loads while occupying a small footprint, allowing it to maneuver in tight spaces. All of Handle’s joints are coordinated to deliver high-performance mobile manipulation.


It sounds like 170,000 is every possible combination of actions that might ever be valid. They stated that usually around 1000 are valid at any point in time.

Based on the examples under the "Model structure" section, I'm guessing they are counting all combinations of spell and target location, including locations on the ground for ground-targetable spells? That could add up quick... e.g. 10 spells * 20 target units * 9x9 grid of locations around each = around 16,000 possibilities.


One that was certainly cool at the time, but has since burned up and was superseded by a collaborative effort. :p


I'm curious how? This FAQ page on their site claims you cannot search email contents, although they plan to support that in their upcoming Protonmail Bridge (which works with a desktop client to search locally): https://protonmail.com/support/knowledge-base/search/

That is what I'd expect, since obviously they cannot build a search index over text they cannot read. Transferring the entire contents of your emails to your web browser to search locally would be slow and impractical on anything but a very fast connection.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: