apawloski

This project is also using aircraft: https://avirisng.jpl.nasa.gov/index.html

Nobody is denying that the quantum system performs the task faster. The question is whether it performs a task that can't feasibly be performed by a classical system. That is what the researchers are specifically referring to when they say "quantum supremacy."

2.5 days is feasible. 10,000 years is not feasible.

Ah. Thanks for the link.

They're saying explicitly:

>We argue that an ideal simulation of the same task can be performed on a classical system in 2.5 days and with far greater fidelity.

and therefore

>Because the original meaning of the term “quantum supremacy,” as proposed by John Preskill in 2012, was to describe the point where quantum computers can do things that classical computers can’t, this threshold has not been met.

Where do you see anything like that in the article?

The argument is that the interface is too complicated for non-experts to parameterize securely.

Consider the libsodium example: https://libsodium.gitbook.io/doc/public-key_cryptography/aut.... The crypto implementation parameters are handled under the hood.

I am reasonably confident that the author is aware of the distinction between correlation and causation.

If you're interested or skeptical, I would highly recommend this book. It's from the "Sleep Loss and the Cardiovascular System" section of Chapter 8. He includes chapter endnotes discussing sources.

My colleague recently recommended the book "Why We Sleep" by Matthew Walker. While I was expecting it to tout the benefits of good sleep, I was surprised at the evidence around the inverse: not getting a full night's sleep is alarmingly bad for you.

The evidence suggests that this extends far beyond cognitive and psychological health impacts. For example, the author notes that there is a 24% increase in heart attacks the day after we lose an hour of sleep during daylight savings. This is just the tip of the iceberg, in my opinion.

Given these mental and physical risks, it is insane that we allow physicians to routinely work shifts that endanger themselves and lead to measurable drops in their performance.

This is a client on top of NASA's Global Image Browse Service (GIBS), like most of NASA EOSDIS' software and Earth observation data, it is free to use and build with. [1] [2]

Both Worldview and GIBS are open source. [3] [4]

Finally, if you're interested in this, also check out https://search.earthdata.nasa.gov, which is a search client on top of NASA's metadata holdings index (CMR). It's not quite as beginner friendly -- it expects that you're somewhat familiar with remote sensing products and geospatial searching -- but it's a good front end into EOSDIS' data.

[1] https://earthdata.nasa.gov/about/science-system-description/...

[2] https://wiki.earthdata.nasa.gov/display/GIBS/GIBS+API+for+De...

[3] https://github.com/nasa-gibs/worldview

[4] https://github.com/nasa-gibs/onearth

James Mickens wrote an interesting and very funny essay on this problem: https://scholar.harvard.edu/files/mickens/files/thesaddestmo...

Because you could have 50 subdomains, for which you'd need either a wildcard cert or 50 separate certs.

That might seem like a ton per week, but consider a PaaS (example-123.herokuapp.com) or a blog platform (example-diary.someblogapp.com).

Personally I'd prefer a wildcard cert there, but at organizations where certificate inventory is a requirement (where they need to track, procure, and invalidate on a per-subdomain basis) Let's Encrypt is a solid option.

They're referring to this lawsuit: https://www.nytimes.com/2018/07/29/us/harvard-admissions-asi...

>> ...thieves can intercept that one-time code by tricking your mobile provider into either swapping your mobile device’s SIM card or “porting” your mobile number to a different device.

> Has this ever actually happened?

Yes. For example: https://krebsonsecurity.com/2018/05/t-mobile-employee-made-u...

How are you managing your SSH keys to make <1HR rotation feasible?

You can encrypt any subdomain at which you can serve an answer to Let's Encrypt's ACME Challenge [1][2].

That being said, I know they have a blacklist of certain domains. I've seen it once with amazonaws.com [3], and it's possible they have similar entries for azure.com, heroku, etc. They don't publicly release their blacklist.

[1] https://letsencrypt.org/how-it-works/

[2] https://ietf-wg-acme.github.io/acme/draft-ietf-acme-acme.txt

[3] As in "my-loadbalancer-1234567890.us-west-2.elb.amazonaws.com"

Where is the Sentinel-2 data hosted on GCE and AWS?

Slightly related, here is an L1 satellite's view of a solar eclipse that I stitched together from Japan's Himawari-8: http://earth.apawl.com/eclipse/

Samy Kamkar had a great talk about radio attacks at DEF CON 23: https://www.youtube.com/watch?v=UNgvShN4USU

Is $15k a weirdly low number for this, given the importance the FBI has attributed to it? Anyone with similar contracting experience who can comment?

"Live" views of the Earth from Himawari-8 (a geostationary Japanese satellite).

Latest image: http://earth.apawl.com

Video of the previous 24 hours at: http://earth.apawl.com/recently

Bonus: Video of the eclipse earlier this week. http://earth.apawl.com/eclipse

>I read this and I think of the courage that it must have taken to write, however, I quickly realize that my thoughts are based on today's mindset where questioning the government gets you labeled a conspiracy nut or traitor.

It's interesting you mention that, as Smedley Butler was the alleger of the Business Plot -- what he claimed to be a planned coup d'etat on FDR by wealthy businessmen. [1]

[1] https://en.wikipedia.org/wiki/Business_Plot

These side-channel attacks are so cool. In 2012, Zhang et al demonstrated a similar cache-based side channel attack across the Xen hypervisor to extract private keys [1]. That is, one virtual machine could extract the keys of a separate virtual machine living on the same hardware.

As cool as these are, most papers indicate the difficulty of pulling these attacks off -- especially given the easier vectors out there.

[1] https://www.cs.unc.edu/~reiter/papers/2012/CCS.pdf

This is very much untrue. 1Password syncs an encrypted vault through separate channel (e.g. Dropbox, iCloud) -- it has zero-knowledge of your passwords. It just picks up a big encrypted blob from wherever you store it.

"The easiest way for us to protect your data and data about you is to not have that data in the first place. You may be noticing a theme by now: we can’t reveal or abuse data that we don’t have.

We do not have your 1Password data. We do not know your 1Password Master Password. We don’t even know if you use 1Password. We do not know how many items you have in your vault or their type."

https://support.1password.com/private-by-design/#what-we-cou...

Just about all of these are visible with about 30 seconds of Googling. Some of them are even the first image in the Google Knowledge Graph summary when you search.

I've had Quinnipiac call me on my cell phone. Assuming the polling uses random-digit dialing, it shouldn't matter if it's a landline or cell phone, right?

Oh, you're talking about a policy limitation, not a technological one. (And if you're talking about the DOE or NSF/Teragrid/XSEDE clusters, then you're probably right. Haven't touched those in years -- and even when I did, I wasn't doing anything crazy.)

I'm somewhat confused -- I thought we were arguing similar points?

I don't want to drag this out, but where do you see the language constraint? You need an MPI binding, sure, but what else?

Write in any language you want; that's irrelevant to the nature of the computations being done here.

By codes they mean -- at the minimum -- pretty much anything that requires frequent communication between any or all nodes as a necessary part of computation. (For example, simulations across a large 3D space, where the changing states of particles on node A directly impacts the states of particles on adjacent nodes.)

Maybe for embarrassingly parallel tasks, but if you require nontrivial interprocess communication, a server farm can't compete with the interconnect of a modern supercomputer.

Search-wise? You're ending up at Lucene either way.

Scaling-wise? Distributed Elasticsearch doesn't have a Zookeeper dependency, which is nice. But Solr has more sharding flexibility and partition tolerance.

Like everything else, depends on your needs.