Nobody Cares! (Science & Tech Edition)

[danielravennest]

Well, Google Skynet will take over the world soon enough!

Google owns 5% of SpaceX who is launching Skynet (Starlink), so you are quite correct.

 

[Brenda Archer]

That’s it, I quit. The Internet has finally defeated us all.

 

[Isabeau]

We're doomed.

These Are the First Living Robots: Machines Made from Frog Stem Cells

Ladies and gentlemen, we have a new kind of life.

www.popularmechanics.com

 

[Argent Stonecutter]

Wot, again?

 

[Sid]

We're doomed.

These Are the First Living Robots: Machines Made from Frog Stem Cells

Ladies and gentlemen, we have a new kind of life.

www.popularmechanics.com

So you mean I no longer should buy lottery tickets?

 

[Bartholomew Gallacher]

Computer scientists do know the halting problem, which in general boils down to this: when looking at a computer program and a given input set the question is if this program runs forever or will come to a halt eventually.

For example this pseudo code runs forever:

while (true) continue

while this program terminates:

print "Hello world!"

For simple programs and input sets the questions can be easiliy solved; the more complex the data though gets the harder it gets to give an answer.

This problem was being first postulated by Kurt Gödel (incompleteness theorems) and Alan Turing proved that it is impossible to create an algorithm on a Turing machine which is able to compute an answer to this problem.

So this is something of a hard border for computer sciences and one of its classic problems/proofs, well... until now maybe. Computer scientists created as thought experiment a hypothetical new proof system called MIP, where a regular computer asks questions to a pair of all-knowing but not necessarily honest “oracles” that cannot communicate with one another.

In reality the problem would be to build such a "all-knowing oracle", which might be impossible to do so. But since this is theory we can already pretend that such things do exist and work with them, also determine their power; even better, we could entangle those oracles on quantum level and call that thing MIP*, which is an improved version of this proof system.

Last year, Wright and Natarajan drafted a proof showing that the spooky connection in the MIP* class gives the verifier interrogating the entangled provers the power to check even more difficult problems (those whose complexity increases double-exponentially with the size of the input). The added entanglement gives more knowledge to the verifier to ask better questions of the provers (the oracles).

And on top of that this construction would be even more powerful than originally thought: They posit that MIP* could efficiently verify every problem in the “recursively enumerable class,” or RE, basically every problem for which it would take a finite amount of time to calculate if the answer was “yes;” a “no” answer could take an infinite amount of time to calculate. MIP*=RE.

The halting problem is part of the RE class. So this means, if this research including its proof is to be found correct, that in theory it is possible to build a machine which is being able to compute the halting problem reliable in fast time. Which by the way would not invalidate Turing's findings, because he worked on a different theoretical construction for his proof, it would just mean that in theory there could be a class of machines which could to this - if we are ever being able to build them is a totally different kind of story. But quite probably we might not be able to do that.

'Remarkable' Mathematical Proof Describes How to Solve Seemingly Impossible Computing Problem

You enter a cave. At the end of a dark corridor, you encounter a pair of sealed chambers. Inside each chamber is an all-knowing wizard. The prophecy says that with these oracles’ help, you can learn the answers to unanswerable problems. But there’s a catch: The oracles don’t always tell the...

gizmodo.com

Link to the paper: [2001.04383] MIP*=RE

 

[Beebo Brink]

Coronavirus news is shifting quickly.
What we know about the coronavirus spreading in China and elsewhere

And:

 

[Beebo Brink]

And just a few hours later:

First Wuhan Coronavirus Patient Identified in the United States

 

[Arilynn]

And just a few hours later:

First Wuhan Coronavirus Patient Identified in the United States

This scares me in a way that a lot of health alert stuff doesn’t.

Despite my denial, I have asthma. It doesn’t bother me at all except I get bronchitis every year and it lasts for 6 - 12 weeks, +/- pneumonia. It’s good that I’m chained to my house while my broken arm heals: My husband cancelled/postponed trips to Asia or the West Coast until I can drive, and I’m not going out pretty much at all. Now I just have to figure out how to dip my kids in boiling bleach....

I hope for everyone’s sake this gets contained, here and abroad.

 

[Beebo Brink]

Virus is continuing to spread, with 9 confirmed fatalities.

The Wuhan virus sweeping China has now killed 9 people and infected 440

 

[Kamilah Hauptmann]

 

[Kamilah Hauptmann]

Spoiler: Can't be unseen

 

[Essence Lumin]

Spoiler: Can't be unseen

That twitter thread lead me to this free 28 page ebook.

Finally, A Practical Guide for Roadside Wildflower Viewing

 

[Isabeau]

Virus is continuing to spread, with 9 confirmed fatalities.

The Wuhan virus sweeping China has now killed 9 people and infected 440

Beijing just canceled New Year's celebrations. I'm glad they're acting faster than they did during the SARS period, and I hope this isn't a repeat.

 

[Argent Stonecutter]

Apparently bats are a reservoir for coronaviruses.

 

[Kamilah Hauptmann]

 

[Kamilah Hauptmann]

Dog nearly put down for weighing 173 lbs. goes on dramatic weight loss journey

The dog had been given up by his previous owners, who nearly had him put down because of his alarming weight — 173 pounds.

edmonton.ctvnews.ca

 

[Argent Stonecutter]

The bloke has obviously only ever used Jira and Trello and I kind of think he has baby duck syndrome with Trello.

I've used all kinds of problem tracking systems, even written a web and GUI front end to an old email-based system that was STILL better than the thing that the new QA department bought to replace it*. I've used Remedy and Service Manager and Bugzilla and I can't remember them all.

I'd put Jira in the top 1%. Anyone who disses Jira that badly is spoiled AF.

* They swore that IT wouldn't need to do anything but keep the servers up, too, and within six months we were running it. It needed half a dozen Windows servers that were flakey as hell, and nobody like the crappy UI. But that's a digression.

 

[Fionalein]

The bloke has obviously only ever used Jira and Trello and I kind of think he has baby duck syndrome with Trello.

I've used all kinds of problem tracking systems, even written a web and GUI front end to an old email-based system that was STILL better than the thing that the new QA department bought to replace it*. I've used Remedy and Service Manager and Bugzilla and I can't remember them all.

I'd put Jira in the top 1%. Anyone who disses Jira that badly is spoiled AF.

* They swore that IT wouldn't need to do anything but keep the servers up, too, and within six months we were running it. It needed half a dozen Windows servers that were flakey as hell, and nobody like the crappy UI. But that's a digression.

It all depends on the effectiveness of the team handling the Jira. Maybe the only time he used Jira was with LL...

 

[Argent Stonecutter]

I admit the givafuckness of the team matters. It was pretty good at first, when they gave a fuck.