Article note: Computing has become such a rent-seeking shithole in such a short time.
I suspect that Wacom doesn’t really think that it’s acceptable to record the name of every application I open on my personal laptop. I suspect that this is why their privacy policy doesn’t really admit that this is what that they do. I imagine that if pressed they would argue that the name of every application I open on my personal laptop falls into one of their broad buckets like “aggregate data” or “technical session information”, although it’s not immediately obvious to me which bucket.
Does Wacom have any competitors? Can you even vote with your wallet, or is this yet another market that isn’t really a market at all?
Article note: That's a _bad_ look.
I've been wondering for some time how the Democratic party apparatus was going to fuck it up, and here they are, already working on it.
Enlarge/ Volunteers tally votes during the first-in-the-nation Iowa caucus at the Southridge Mall in Des Moines, Iowa, US, on Monday, Feb. 3, 2020. The app used to submit the results turned out not to be seamless, scalable or robust. (credit: Al Drago/Bloomberg via Getty Images)
Iowa's Democratic Party turned to an untested software platform tied to a mobile application to streamline reporting from its presidential caucuses last night. What could possibly go wrong?
In a collapse that echoed the failure of a canvassing application used by Sen. Mitt Romney's 2012 presidential bid, the caucus reporting app repeatedly hung as precinct leaders attempted to submit returns. A backup hotline was jammed for hours. And as of the morning after the caucuses, the full results are still not tallied. The Iowa Democratic Party has promised at least 50 percent of results by the end of the day.
The application was built on technology provided by Shadow Inc.—a technology company that received seed funding from the nonprofit ACRONYM.
Article note: Well, knowing that YouTube is making $15 Billion a year makes the arguments about "No competition because it's a money pit that Google supports for data and mindshare" seem even less credible.
It also makes the amount of petty fuckery that Youtube monetization is famous for a lot harder to justify.
Illustration by William Joel / The Verge
YouTube generated nearly $5 billion in ad revenue in the last three months, Google revealed today as part of parent company Alphabet’s fourth quarter earnings report. This is the first report under newly instated Alphabet CEO Sundar Pichai, who took over as the chief executive of the entire company late last year after co-founders Larry Page and Sergey Brin stepped back from day-to-day duties and promoted Pichai, formerly Google CEO, to the top spot.
The announcement marks the first time in YouTube’s nearly 15 years as a Google-owned platform, since Google bought the website in 2006 for $1.65 billion, that the company has revealed how much money YouTube-hosted ads contribute to the search giant’s bottom line.
Article note: A variation of this trick will be used in some show for a 2020s version of the "Fiddling the stoplights" scene from Hackers.
It is pretty clever.
Article note: Those fuckers. I figured I just needed to update Widevine on my TV computer, but no, they intentionally broke it.
I don't really want to pay $6/mo to Amazon for my junk TV which is largely covered by the "Free" CBS streaming offerings just because of my platform, now I have to pick user-hostile ripoff or piracy again.
Update: Hey! They responded and fixed the issue! I can go back to watching some garbage TV as the provider prefers without excessively shitting up my experience!
Update, January 31: After this story went live earlier in the week, an Ars reader reached out to speculate that the problem was most likely due to enabling VMP (Verified Media Path) on CBSi's Widevine server. Verified Media Path, similarly to UEFI Secure Boot, makes certain that content will only be delivered to browsers with sanctioned, verifiable "authentic" framework; this is a configurable behavior, and by default, unverified platforms are allowed to receive licenses.
This morning we asked CBSi executives to check with their engineers and see if this was the problem. While we never received a response, two hours later, CBS All Access was playing successfully on Google Chrome on multiple Linux distributions. (Firefox still crashes.) For now, we have verified that the fix—which, again, may or may not actually have involved VMP—covers all of CBS' content and not merely the first episode of Picard, which CBS released yesterday on YouTube for a limited time. If we hear official word from CBS regarding what happened behind the scenes, we'll update this post accordingly. The original story appears unchanged below.
As of this month, the CBS All Access streaming-video platform—home of popular shows including The Late Show with Stephen Colbert and now Star Trek: Picard—stopped working on Linux PCs, regardless of the choice of browser. Ten years ago, this would have been just another day in the life of a Linux user, but it's a little surprising in 2020. We were originally tipped off to the issue by a few irate readers but quickly found it echoed in multiple threads on Reddit, Stack Exchange, and anywhere else you'd expect to find Linux users congregating.
Trailers and ads all work fine on CBS All Access, in any browser. The problem isn't the streaming—it's apparently something to do with the DRM itself. [credit:
Jim Salter ]
I'm both a Linux user and a CBS All Access subscriber myself, but I had been unaware of the problem since I do all my own watching on a Roku. Technically, the Roku is a Linux PC in its own right—but CBS has its own app in the Roku store, which works perfectly.
Article note: Ooh, a horrible consequence of "expert [support] systems" I hadn't thought much about because it's so transparently shitty; intentional manipulation by interest groups.
ERIC BARADAT/AFP via Getty Images
San Francisco-based medical records startup Practice Fusion allegedly developed software for pharmaceutical companies to help increase the number of prescriptions doctors wrote for pain medications, according to a settlement with the US Department of Justice, Bloombergreports. The company, which supplied its software to tens of thousands of doctors offices nationwide, admitted doing so as part of a $145 million federal settlement this week to resolve civil and criminal penalties, including $113 million to be paid to the federal government and more than $5 million to states.
Here’s how the software worked: When a health care provider accessed a patient’s electronic health records (EHR) on Fusion’s software, a pop-up window would appear...
Article note: Interesting. It's not a "RedHat bought IBM with IBM's money" situation, but has a lean in that direction.... because I think IBM is looking for any kind of credible direction.
IBM named Arvind Krishna as chief executive officer, replacing longtime CEO Virginia Rometty. Krishna is currently the head of IBM’s cloud and cognitive software unit and was a principal architect of the company’s purchase of Red Hat, which was completed last year. Rometty, 62, will continue as executive chairman and serve through the end of the year, when she will retire after almost 40 years with the company, IBM said in a statement Thursday. Good luck to the man, I guess. IBM isn’t exactly the most exciting company in the world.
Article note: I feel like almost all of those compromises went in the sensible direction. If your code is being used, almost every compromise should favor improving run-time behavior over compiler speed.
Article note: It's generally a good tale of responsible academic scrutiny and appropriate reaction.
And now the grim, I'm not suggesting the victim/author here should be reprimanded because they more or less did right, but it does demonstrate the "Academic misconduct pays off as long as you get tenure before you get caught" principle that has made academia extra bullshit lately, both in terms of the research people are choosing to do (Machine Learning and Quantum are both "safe" topics because they give results that aren't near-term falsifiable, doing to computing what string theory &co. did to physics years ago) and in terms of the credibility of the promotion process.
Article note: Sweet. From my limited playing, WireGuard is good tech that straightforwardly solves real problems, plus it triggered some good cleanup of the kernel crypto tools.
Enlarge/ It's not likely to be an accident that "add WireGuard" is number one on this list. (credit: Jim Salter)
Yesterday, Linux creator Linus Torvalds merged David Miller's net-next into his source tree for the Linux 5.6 kernel. This merger added plenty of new network-related drivers and features to the upcoming 5.6 kernel, with No.1 on the list being simply "Add WireGuard."
As previously reported, WireGuard was pulled into net-next in December—so its inclusion into Linus' 5.6 source tree isn't exactly a surprise. It does represent clearing another potential hurdle for the project; there is undoubtedly more refinement work to be done before the kernel is finalized, but with Linus having pulled it in-tree, the likelihood that it will disappear between now and 5.6's final release (expected sometime in May or early June) is vanishingly small.
WireGuard's Jason Donenfeld is also contributing AVX crypto optimizations to the kernel outside the WireGuard project itself. Specifically, Donenfeld has optimized the Poly1305 cipher to take advantage of instruction sets present in modern CPUs.