close
  • Ar chevron_right

    Missing mass? Not on our watch—Dr. Paul Sutter explains dark matter

    news.movim.eu / ArsTechnica · 3 days ago - 14:10

Produced and directed by Corey Eisenstein. Click here for transcript . (video link)


Greetings, Arsians! We have something special for you today: the premiere of a new science series we're creating, called Edge of Knowledge . We've recruited physicist and author Dr. Paul Sutter ( Google Scholar link ) to be our host and guide on an eight-episode romp through the mysteries of the cosmos, touching on topics that we at Ars find fascinating. This means we'll have episodes on black holes, the future of climate change, the origins of life, and, one of my favorite topics for our premiere: dark matter .

Dark matter: The universal majority

As Ars readers, you're all probably familiar with Douglas Adams' " Space is big " opening to The Hitchhiker's Guide to the Galaxy , but "big" only tells part of the story. You might assume that, as a corollary to all that bigness, space should also be generally vast and empty, with just an occasional stray hydrogen atom whipping its way through an otherwise perfect vacuum of nothingness—but nothing could be further from the truth.

Read 6 remaining paragraphs | Comments

index?i=pWToxvlj5n4:D6QRZsmI8C8:V_sGLiPBpWUindex?i=pWToxvlj5n4:D6QRZsmI8C8:F7zBnMyn0Loindex?d=qj6IDK7rITsindex?d=yIl2AUoC8zA
  • Ar chevron_right

    The best Mario games ever made

    news.movim.eu / ArsTechnica · Wednesday, 24 November - 13:00 · 1 minute

Mario loves us. We love Mario.

Enlarge / Mario loves us. We love Mario. (credit: BEHROUZ MEHRI / Getty Images)

Ars Technica Editor-in-Chief Ken Fisher has a rule: If you have a dumb, fun conversation in the Ars Slack that lasts for more than 10 minutes, it's probably worth turning that conversation into some kind of article. And that's how a weekday water-cooler-style discussion about Platonic idealism and Mario became what you're reading now!

For people of a certain age—which, dear readers, most of us are—"video games" and "Nintendo" meant practically the same thing. (There are even a few of us who are older than a certain age, who came from the Great Long Long Ago time when "video games" meant "Atari," and even those few acknowledge Nintendo's culture-changing dominance in the mid-to-late 1980s.) So all of us have played at least a few different games featuring the world's most famous plumber, Mario Mario. (Yeah, his last name is also Mario . Which means his brother's name is Luigi Mario. Which means that calling Luigi "Green Mario" is actually correct! Vindication !)

A few Ars staffers volunteered to brave the inevitable slings and arrows of the comments section to put down their thoughts on a simple question: out of every video game in which Mario made an appearance, which one is your absolute top-shelf favorite, and why?

Read 44 remaining paragraphs | Comments

index?i=3LtF193FfcA:CGnMG76ee5U:V_sGLiPBpWUindex?i=3LtF193FfcA:CGnMG76ee5U:F7zBnMyn0Loindex?d=qj6IDK7rITsindex?d=yIl2AUoC8zA
  • Ar chevron_right

    Securing your digital life, the finale: Debunking worthless “security” practices

    news.movim.eu / ArsTechnica · Wednesday, 17 November - 14:00 · 1 minute

Extreme close-up photograph of jar of pills labeled

Enlarge / Take one daily to keep Evil Hackerman away! (credit: Aurich Lawson | Getty Images )

Information security and privacy suffer from the same phenomenon we see in fighting COVID-19: " I've done my own research " syndrome. Many security and privacy practices are things learned second- or third-hand, based on ancient tomes or stuff we've seen on TV—or they are the result of learning the wrong lessons from a personal experience.

I call these things "cyber folk medicine." And over the past few years, I've found myself trying to undo these habits in friends, family, and random members of the public. Some cyber folkways are harmless or may even provide a small amount of incidental protection. Others give you a false sense of protection while actively weakening your privacy and security. Yet some of these beliefs have become so widespread that they've actually become company policy.

I brought this question to some friends on InfoSec Twitter: " What's the dumbest security advice you've ever heard ?" Many of the replies were already on my substantial list of mythological countermeasures, but there were others that I had forgotten or not even considered. And apparently, some people (or companies... or even vendors!) have decided these bad ideas are canon.

Read 30 remaining paragraphs | Comments

index?i=bXOF12jF69Q:5K0sxRvnri0:V_sGLiPBpWUindex?i=bXOF12jF69Q:5K0sxRvnri0:F7zBnMyn0Loindex?d=qj6IDK7rITsindex?d=yIl2AUoC8zA
  • Ar chevron_right

    What a “converged battlefield” means to the future of air… I mean, joint warfare

    news.movim.eu / ArsTechnica · Wednesday, 3 November - 13:00 · 1 minute

A three-ship flight of F-22 Raptors. As one of the US Air Force

Enlarge / A three-ship flight of F-22 Raptors. As one of the US Air Force's most sophisticated fighters, it might see combat in the future not against insurgencies, but against technologically sophisticated adversaries like China or Russia. (credit: Stocktrek Images / Getty Images)

With the war in Afghanistan in the rearview mirror, US military planners are trying to pivot toward a very different set of challenges than fighting the Taliban. Competition with what Department of Defense officials have commonly called "near peers" presents a set of new challenges to a military that has been focused for the past two decades on counterinsurgencies and terrorism. The United States wants to maintain its ability to respond to nonstate, "asymmetric" adversaries while also figuring out how to fight countries that are America's technological equals.

In the synopsis of the 2018 National Defense Strategy, DOD officials laid out the problems: an ascending China looking for "Indo-Pacific regional hegemony in the near term and displacement of the United States to achieve global preeminence in the future" and an ambitious Russia seeking "to shatter the North Atlantic Treaty Organization and change European and Middle East security and economic structures to its favor"—with both nations using information warfare and modernized conventional and nuclear forces to flex their muscles. This makes for a vastly different adversary than the kind the United States has been fighting for the last 20 years.

Responding to China's rapidly growing air and naval forces will require more than just better ships and aircraft, too; the People's Liberation Army (PLA) air force and navy are in many ways newer than their US counterparts and increasingly similar in capability. There's also the matter of the US being able to manage forces in an electromagnetically hostile environment—full of jamming, active countermeasures, and potential intrusion and disruption of military networks by other means. The Taliban didn't have the capability to engage in sophisticated electronic warfare against the US—but China and Russia absolutely do.

Read 24 remaining paragraphs | Comments

index?i=UuJxaefVUTw:oylw_2cseV8:V_sGLiPBpWUindex?i=UuJxaefVUTw:oylw_2cseV8:F7zBnMyn0Loindex?d=qj6IDK7rITsindex?d=yIl2AUoC8zA
  • Ar chevron_right

    iPhone 13 and 13 Pro review: If you could have three wishes

    news.movim.eu / ArsTechnica · Friday, 24 September - 14:01

The iPhone 13 Pro Max, photographed by the iPhone 13 Pro in low light.

Enlarge / The iPhone 13 Pro Max, photographed by the iPhone 13 Pro in low light.

Imagine you were visited by a genie who would grant you three wishes, but they all had to be about what you want from your next smartphone. As market research and surveys tell it, almost everyone would make the same three wishes: great battery life, excellent cameras, and big, beautiful screens.

This year, Apple is that technology genie, because that’s exactly what the iPhone 13, iPhone 13 mini, iPhone 13 Pro, and iPhone 13 Pro Max deliver when they hit store shelves today.

Cupertino’s flagship phone lineup might seem like an iterative “S”-style update, given that the phones look almost the same as last year's models and that there are no major new features apart from screens with higher refresh rates in the priciest models. But since Apple zeroed in on most people's highest priorities, this seemingly iterative update ends up being a noteworthy one.

Read 83 remaining paragraphs | Comments

index?i=azffJflPxIw:nqKb3SrroNw:V_sGLiPBpWUindex?i=azffJflPxIw:nqKb3SrroNw:F7zBnMyn0Loindex?d=qj6IDK7rITsindex?d=yIl2AUoC8zA
  • Ar chevron_right

    War Stories: How Crash Bandicoot hacked the original PlayStation

    news.movim.eu / ArsTechnica · Sunday, 5 September - 12:45 · 1 minute

Shot by Sean Dacanay, edited by Jeremy Smolik. Click here for transcript .

When you hear the name Crash Bandicoot , you probably think of it as Sony's platformy, mascoty answer to Mario and Sonic. Before getting the full Sony marketing treatment, though, the game was developer Naughty Dog's first attempt at programming a 3D platform game for Sony's brand-new PlayStation. And developing the game in 1994 and 1995—well before the release of Super Mario 64 —involved some real technical and game design challenges.

In our latest War Stories video, coder Andy Gavin walks us through a number of the tricks he used to overcome some of those challenges. Those include an advanced virtual memory swapping technique that divided massive (for the time) levels into 64KB chunks. Those chunks could be loaded independently from the slow (but high-capacity) CD drive into the scant 2MB of fast system RAM only when they were needed for Crash's immediate, on-screen environment.

The result allowed for "20 to 30 times" the level of detail of a contemporary game like Tomb Raider , which really shows when you look at the game's environments. Similar dynamic memory management techniques are now pretty standard in open-world video games, and they all owe a debt of gratitude to Gavin's work on Crash Bandicoot as a proof of concept.

Read 3 remaining paragraphs | Comments

index?i=TXlBDHNvAbA:ETYYcQahb-0:V_sGLiPBpWUindex?i=TXlBDHNvAbA:ETYYcQahb-0:F7zBnMyn0Loindex?d=qj6IDK7rITsindex?d=yIl2AUoC8zA
  • Ar chevron_right

    Our AI headline experiment continues: Did we break the machine?

    news.movim.eu / ArsTechnica · Tuesday, 20 July, 2021 - 15:00 · 1 minute

Our AI headline experiment continues: Did we break the machine?

Enlarge (credit: Aurich Lawson | Getty Images)

We're in phase three of our machine-learning project now—that is, we've gotten past denial and anger, and we're now sliding into bargaining and depression. I've been tasked with using Ars Technica's trove of data from five years of headline tests, which pair two ideas against each other in an "A/B" test to let readers determine which one to use for an article. The goal is to try to build a machine-learning algorithm that can predict the success of any given headline. And as of my last check-in , it was… not going according to plan.

I had also spent a few dollars on Amazon Web Services compute time to discover this. Experimentation can be a little pricey. ( Hint : If you're on a budget, don't use the "AutoPilot" mode.)

We'd tried a few approaches to parsing our collection of 11,000 headlines from 5,500 headline tests—half winners, half losers. First, we had taken the whole corpus in comma-separated value form and tried a "Hail Mary" (or, as I see it in retrospect, a " Leeroy Jenkins ") with the Autopilot tool in AWS' SageMaker Studio. This came back with an accuracy result in validation of 53 percent. This turns out to be not that bad, in retrospect, because when I used a model specifically built for natural-language processing—AWS' BlazingText —the result was 49 percent accuracy, or even worse than a coin-toss. (If much of this sounds like nonsense, by the way, I recommend revisiting Part 2 , where I go over these tools in much more detail.)

Read 29 remaining paragraphs | Comments

index?i=CNcItAMFBjg:vM98vJY7pXQ:V_sGLiPBpWUindex?i=CNcItAMFBjg:vM98vJY7pXQ:F7zBnMyn0Loindex?d=qj6IDK7rITsindex?d=yIl2AUoC8zA
  • Ar chevron_right

    Feeding the machine: We give an AI some headlines and see what it does

    news.movim.eu / ArsTechnica · Thursday, 15 July, 2021 - 15:00

Turning the lens on ourselves, as it were.

Enlarge / Turning the lens on ourselves, as it were.

There's a moment in any foray into new technological territory when you realize you may have embarked on a Sisyphean task. Staring at the multitude of options available to take on the project, you research your options, read the documentation, and start to work—only to find that actually just defining the problem may be more work than finding the actual solution.

Reader, this is where I found myself two weeks into this adventure in machine learning. I familiarized myself with the data, the tools, and the known approaches to problems with this kind of data, and I tried several approaches to solving what on the surface seemed to be a simple machine-learning problem: based on past performance, could we predict whether any given Ars headline will be a winner in an A/B test ?

Things have not been going particularly well. In fact, as I finished this piece, my most recent attempt showed that our algorithm was about as accurate as a coin flip.

Read 33 remaining paragraphs | Comments

index?i=X_hkc5ytrUw:JLp3_Dc6clI:V_sGLiPBpWUindex?i=X_hkc5ytrUw:JLp3_Dc6clI:F7zBnMyn0Loindex?d=qj6IDK7rITsindex?d=yIl2AUoC8zA