neetdeth

Ha! Guess it was nice while it lasted.

Absolutely, you need to be dealing with a specialty vendor that understands their own supply chain. I’ve had good experiences with batteryspace.com (AA Portable Power). Not affiliated.

Amazon’s logistics advantage is huge, but unless I absolutely need it tomorrow they’re rarely the best option. Oddly enough, there are a few resellers of marked up Chinese electronics that have established a certain amount of brand trust with me - “DROK” and “SMAKN” come to mind. Brands that provide a reasonable amount of curation to help you avoid the absolute worst crap, but are small enough not to attract counterfieters themselves. It’s an odd niche that Amazon’s failure of inventory control has created.

The former can be just as disingenuous as the latter. Jon Stewart wielded immense political influence amongst the under-30 crowd, but whenever seriously challenged, hid behind the defense of comedy.

Yiannopoulos consistently billed himself as a provocateur, little different from a Coulter or Maher.

People have been blending news, entertainment, and polemic together for centuries and it's not going to stop because some people want to codify rules of Serious Journalism and impose them universally across the Internet.

Although fraud is no doubt rampant in Chinese industry, the penalties are... rather more severe. You had better make sure your scheme doesn't piss off the wrong people. The PRC's military and aerospace supply chains are probably in pretty good shape.

I'm actually happy to see this. Strongly feel that the current generation of headsets are not good enough and VR needs at least one more cycle of pushing the envelope at the high end price point before finding mass market success.

Unfortunately, GPU performance per watt/$ seems to be capping out just short of the goal.

Are you telling me he shares his personal opinions on his personal Twitter account? The monster!

Now clearly, Mr. Persson has not been completely un-personed. Yet. But the process is rolling.

Fun idea: On-orbit CDN endpoints with predictive caching to warm up the next satellite before it passes overhead.

Surely it would not be a problem to have Markus stay on topic and avoid any controversial territory in his remarks.

However, is it possible to ask others to refrain from "calling him out" in such a public setting? No, not at all. Notch is a hated figure, and he is hated with an intensity that only Progressive religious fervor can bring about.

This is how you silence people. By your own actions, make association with the target so troublesome that the institutions - reliably feckless as they are - will simply prefer to exclude them.

Brings to mind this article, a history of the rise and fall of Common Lisp at JPL, which was almost selected as the control language for their robotic space missions.

http://www.flownet.com/gat/jpl-lisp.html

And of course, JavaScript having its origins as a more marketable syntactic overhaul of Lisp for in-browser scripting, it's not hard to see how we end up with JS on the JWST for many of the same reasons.

Like others, I tend to idealize space tech from a distance and imagine it using technologies that are across the board better, cooler, more pure than anything that stays on the ground. While the need for extreme reliability does change things a lot, it turns out most of the same forces that guide software selection anywhere else are at play there as well.

Assuming every one of those overnight stays means two days spent in Berlin, and if they are spread out over the year, that's 175 thousand tourists at any given time, about 5% of the population. Of course they'll congregate in certain locations and certain times of year, but you're overstating the case.

Instead of shaking your fist at airplanes flying overhead, maybe try increasing the hotel tax and cracking down on AirBnBs.

Of course, you'll probably encounter opposition to this because some locals are making a great deal of profit.

House sharing: It's like podsharing but you get an entire room with a door that closes.

deGooglifying by moving from a Google Google phone to an LG Google phone.

Of course, the only viable alternative at this point is Apple. That's the choice I made and I feel they've earned higher consumer trust... for the moment.

I really think that, with smartphone technological improvements flattening out and the oligopoly starting to squeeze their customer base, there's never been a better time for open smartphone projects to get a foothold.

The false dichotomy persists in your rephrasing. You could maintain excellent fitness with a more moderate exercise regime.

Sorry if it comes across as an attack on your lifestyle. It isn't meant to be. If you find extrinsic motivations morally suspect, then let me assure you I believe your motivations are entirely intrinsic and pristine.

But regardless, there are such things as trends. It isn't presumptuous to assume that people do things for the social benefits. That's just being realistic. If it builds their social network, increases their status, widens their career prospects, makes it easier to find a better mate... perhaps they should be doing that thing, whether it's running an ultramarathon or doing a 6-hour headstand in a park. What does it matter?

Except that if 6-hour headstands become the hot new thing that every young professional seems to aspire to, I'm going to ask about whether "society" should be glorifying something that's obviously going to lead to neck damage versus, say, a nice 6-hour foot race. That's all.

> Which society? What glorification?

The high-functioning, wealthy society of working age professionals. I think of the United States - and I assume most countries are the same, but perhaps to a lesser degree - as a cluster of societies that sometimes intermingle and share space with each other but are otherwise distinct. Obesity is a disease of the poor.

In certain circles there is a glorification of extreme forms of exercise. Partially as an indicator of status, to demonstrate that you have the both the personal agency and excess resources to pour into a training program to achieve impressive feats. But also as a sacrifice to the gods of success. All the respected leaders in business are doing ultra endurance sports, so it would stand to reason that there is something, some X-factor it's contributing to their ability to achieve.

It's worth asking whether there's any benefit to this kind of thing past a certain point. Are people their wasting time and causing long term damage to their bodies just to impress their friends?

Your last sentence, by the way, gave me a laugh. It could be the dictionary example for false dichotomy. Not sure whether this was intentional.

As long as you're admitting that there's no general principle being evenly applied here, and it is entirely discretionary on Facebook's part, we have an understanding.

e: But it's not an arbitrary exercise of power when I do it!

> other countries have banned hate speech without ... banning legitimate political discourse

This is, of course, entirely a matter of opinion.

But they don't say that, and if they did, you certainly wouldn't need any new policies to ban them. There are the 8chan lunatics, but outside that there are a wide range of views on what the white nationalist end-goal is and how to achieve it.

I'm reminded of the Sargon v Richard Spencer debate here. Sargon's line of attack was essentially this: You advocate for an ethnostate, people will not react well to the non-violent policies you publicly advocate to bring this about, thus there will be violence, therefore your rhetoric is violent. Even if you approve this line of reasoning, you have to admit that it can be applied to any separatist movement anywhere in the world.

The hype around machine learning has reached a point where every time someone sees a moderately challenging problem they leap to some nebulous "ML" solution.

No, we've long had the technology to solve most of this. Clip correspondence is content based image retrieval across a database of keyframes. Matching framing and position is an image registration problem (feature correspondence). Color balance seems like an almost trivial problem if you've solved the other ones, because the information _is_ there -- just modify each channel to match the histogram in the lower resolution image [1]

The challenge is that earlier stages in the pipeline need to be robust to inexact matches, and we don't want to rely on absolute color or pixel position. But I don't think that should be insurmountable for a slightly creative implementor - use local variation in color, gradient descriptors, pull in the motion vectors for an additional channel, etc...

Sure, you could go down the deep learning path by trying to reduce scenes to bags of labeled objects and semantic actions but that's bringing a water cannon to a squirt gun fight, which only makes sense if Google is giving you a free water cannon.

I'd probably try it if I really thought there was a fortune to be made here, but it's such a niche application. When something hasn't been done yet, there's usually a good economic reason, unfortunately.

1. First Google result for "histogram matching color balance": https://blog.mapbox.com/color-balancing-imagery-with-histogr...

Any attempt to scale is going to have to come up with serious barriers to hold back the tsunami of lonely and undesirable males. I don't believe it's actually possible.

The clock source is typically generated by a phase locked loop from the external crystal oscillator. There's a PLL multiplier and divider value that you can set obtain the desired frequency.

I think the MCU starts running at a low speed from an internal RC oscillator, and then setting up the high speed clock source happens early in the initialization code. This code can be generated by ST's STM32Cube software, which is useful as a reference implementation even if you don't use the generated code in your project.

There’s a pretty vast gulf between 70s era threshold, edge detect, morphology type methods and deep learning.

Robust feature detectors, voting methods, etc. can fill that gap for object detection and pose estimation. Also simpler machine learning models for classification like SVMs.

It increasingly feels like the slice of computer vision techniques I learned in the early 2000s is receding in relevance. But at the same time, there ought to be a more modern grab bag of robust tools to build engineered solutions.

In the end any algorithm is only as good as the dataset it’s developed/validated against. Ad-hoc methods may fail in bright sunshine, but so too would a neural network if that condition never arises in its training set.

There's now a glut of overproduced, samey content trying to compete as timewasters for a general audience. It feels like browsing through mass market paperbacks at the supermarket.

I think we're in a golden age if you're interested in fringe politics. Otherwise, we're well past it and on the decline.

Because Amazon is no longer a simple bookseller. It's a diversified mega-corporation that is big enough that the average individual will want to think twice before getting on its bad side.

What if you want to host something on AWS? Stream on Twitch? Shop at an Amazon Go? Keep listening to your Audible library?

What if you're using home automation equipment that's tethered to a cloud service and Amazon acquires the company?

Sure, a simple chargeback probably won't trigger broad spectrum retaliation though all their subsidiaries, but the fact is that Amazon has accrued a lot of power in recent years.

“Doctor Pinker, you’re killing the patient!”

“Well that’s a fairly narrow and bigoted way of looking at it. Look at how many people are benefiting from her blood and organs.”

This study was peer reviewed, and neither the researchers nor the reviewers are drooling morons as you seem to imply. From the article:

"Based on prior research, we included a rich set of commonly observed predictors of well-being as control variables (see Dolan et al., 2008 for a review of this literature). These include socio-demographic variables such as age, household income, gender, relationship status, number of children and education. Such controls are important as it is possible that fruit and vegetable consumption would co-vary with, for example, income – someone becoming richer and as a result enjoying higher well-being scores and changing their food related behaviours. We also controlled for the presence of long-standing health conditions as well as health and lifestyle related behaviours such as walking frequency and smoking behaviour – again potential confounders that could bias any estimates of the relationship between fruit and vegetable consumption and well-being."

While I do think a lot of research in the social sciences is of questionable value, there's no need to shoot from the hip just for the snarky HN comment.

Possible that the hype around deep learning will dissipate as highly anticipated applications turn out to be more challenging than expected, technologists settle in for a long trek, and investors flee the space in search of easier returns. We've seen this happen before.

The real growth opportunity seems to be with Tegra. If we see a boom in autonomous vehicles and robotics, they're well positioned to profit. But we may not.

It also appears AMD has finally arrived at the correct strategy to challenge NVIDIA's datacenter moat with HIP (basically API compatibility with CUDA modulo trademarks). Does not mean they have mindshare or a desirable product line, but when you have a completely dominant market position that's baked into the stock price you have to consider every threat.

NVIDIA is a great company with great products that will do well regardless, just positing some reasons one might be short the stock apart from Chinese market considerations.

Regarding background sources, the fact that the target of interest is moving could actually be a great benefit for background rejection. Place two detectors on the path separated by some distance, knowing how long it takes for a typical vehicle to travel between them, and you can filter out simultaneous spikes as background fluctuation. Modern photon counting detectors are pretty amazing, and it would take only a small number of temporally correlated discrete events at each detector to reach the probability threshold of alarm.

I don't work on these systems, obviously - those who do aren't talking - and this is just idle speculation. But I do know that lots of DHS money is shoveled into R&D on this kind of thing every year and I would not be surprised if some of the tech does work.

> Presuming you meant radiation treatments, those do not leave the patient radioactive at all.

Not categorically true, even with external beam radiation therapy.

https://www.ncbi.nlm.nih.gov/pubmed/25271934

Other kinds of treatment can leave the patient significantly radioactive, for example brachytherapy, which involves direct implantation of radiation sources in tissue.

Well, I tried to reproduce it with a private browsing window.

That Jordan Peterson video was in the #4 position in the sidebar from the Musk interview. However, it was the first video that did not include Joe Rogan. So far so good.

Thereafter, clicking the first recommended brought me to another Peterson video, then Gordon Ramsay, and then into an endless loop of Kitchen Nightmares clips.

Certainly, there's some stochastic element here, but I'm still not convinced.

Ok, I'll bite. What exactly did Google do wrong here?

I'd be persuaded if you could show that a random walk through the recommendations ended up in far-right opinion some disproportionate percentage of the time. What you did was a goal directed search. At every point you selected the most right leaning perspective, and in the end you got where you wanted to go.

The only way to stop this from happening is to ban objectionable opinions from the platform entirely, or alternatively to disconnect them from the graph of recommendations. The latter is what YouTube's "limited state" already does.