This might be a pretty naive question, but: in a hypothetical world where the vast majority of systems programming is done in "memory safe" langs, what would most vulnerabilities look like? How much safer would networked systems be, in broad strokes?
I struggle to think of a piece of software I've been as happy with paying for as ST. It's just a reliable and snappy text editor that _feels good to use_. Sure, it doesn't have all the fancy new things Atom and VSCode often do, but that doesn't really matter to me. I work on fairly small codebases (scientific data analysis, mostly) and ST is more than good enough for it. Whenever I feel the need for something more advanced, I'll switch to neovim. But ST + neovintageous usually get the job done. I'll happily continue to support ST!
In my day job I'm constantly reading PDFs. Being able to zoom in/out, scroll, and move the cursor with such low effort can even bring me a little bit of joy. It's not really about what is most efficient. Computing should feel good. That's why it's hard for me to move away from Apple's trackpads.
Our technology is finally getting into the realm of things where something like this might be made possible, for small brains such as those of fruit flies or zebrafish. Already we can perform near-whole-brain recordings of these animals using 2-photon technology. And with EM reconstruction methods advancing at such a rapid pace, very soon we'll be able to acquire a picture of what an entire brain's structure (down to the synapse) and activity across all these structures looks like.
Anyone have a good resource on understanding the policies that govern the ownership and handling of data collected by an Apple Watch? I quite like that they can be used for tracking sleep, heart rate, and many other health-related items. Ideally, I'd have full control over how these valuable, longitudinal datasets I generate are handled and used. But I'm skeptical that's anywhere near the case.
Wow, I just tried it out. This is really great. And it solves my interactive plot saving requirement. Easy as doing `Plotly.savehtml(fig, "test_fig.html")` :). Thanks!
I have, though I've mostly stuck with the plain PyPlot.jl package due to the familiar syntax and interactivity. Perhaps things have changed, but I just recall being frustrated at the inability to zoom in/out, and again save interactive figs. Perhaps it was just due to the particular backend I was using. I'll give the VS Code experience another try!
I'm a graduate student that's switched almost completely over to Julia. Prior to it I worked in both MATLAB (the IDE is so nice, and writing out matrix computations is just great) and Python (for ML). Julia is absolutely nicer to write in than either of the two. MATLAB is slow and at times feels less like a programming language and more like an incomplete and brittle interface with the JVM. Python is also slow, and it feels awkward to use given that it was not explicitly designed for scientific workflows. With Julia I get proper typing, incredible speed, easy parallelization, and a kick ass REPL.
The only thing I truly miss in using Julia is the plotting capacities of MATLAB. I haven't found an environment that can match it in terms of interactivity. Give me the ability to (easily) save interactive figures for later use and Julia would be perfect.
Are these things so important that we should give up the easier maintainability (and potentially, security) that comes from centralization and the standardization of user clients? I absolutely think projects like Matrix are worthwhile, but it seems foolish to me to argue that there are no benefits that come from doing things the way Signal does. Besides, as sexy as decentralization is, in the wild it is not really practiced. Mastodon is decentralized, but last I heard ~50% of users were on 3 nodes [0]. I'm assuming it's a similar situation for Matrix.
I never use the DDG results when looking for something. People on here will sometimes say that the results are as good as Google, but it's just not true in my experience. I would switch completely to Startpage, but I value the !bang feature too much. So for general searches, everything is prefaced with !sp. It's just muscle memory now.
The game is great on PC, and sure there are some bugs, but nothing catastrophic. I've been playing the game on GeForce Now and I was pretty amazed with how much game streaming tech has progressed since it was first introduced. There's something amazing about playing Cyberpunk on my mac mini with ultra settings and ray tracing on. Latency's not even noticeable if you have a good connection!
It might rather be due to the brain simply requiring a baseline level of excitability to be present in order for it to properly function. Many unconscious states are characterized by reduced activity throughout the brain. The thalamus, being connected to practically every area of cortex, is well positioned to push the cortex into a more active state. And note that there are loops everywhere! The thalamus connects to the cortex, the cortex connects to the basal ganglia, which connects back to the thalamus. And there are many more feedback loops between these structures. So it could be that you simply need to “kickstart” the system, as this article is suggesting, and once a baseline level of excitability is reached, all of these loops do the job of keeping the system at a functioning, stable state of excitability. To what extent this leads to the repair of damaged brain areas is less clear, but we know that lack of activity causes atrophy, so pushing the system back into an excited state could in principle prevent further damage.
There is work in anesthetized monkeys (the neural and behavioral signatures of anesthesia are similar to those of coma) demonstrating that high frequency thalamic stimulation is sufficient to restore some markers of wakefulness, even while the monkey is continuously receiving anesthetic. It’s wild.
One way in which I've found the two to be incredibly similar: they're both very good at throwing me into the flow zone. And I love being there, which is why my full time job is basically programming all day long.
They're different because you're solving pretty different problems when you're doing them. Drawing well requires honing the particular set of fine motor skills which allow you to set down lines smoothly. It also requires you to be able to visualize shapes well, and to develop a good aesthetic sense. These things are for the most part nonverbal, and require a long time to master. When I'm drawing, I'm not really thinking as explicitly about what I'm trying to do as when I'm programming. It's more seat of the pants, because you can't put a line down 100% in the way you'd like to. And you're not really trying to anyway. You adjust here and there, but ultimately "right" is just a subjective feeling. Programming isn't fuzzy like that -- there are particular bounds to what the the compiler is going to let you do, and there is often a well defined "correct" behavior. The problems you're solving in programming are much more easily verbalized.
It was incredibly worth it for me to turn my gaming PC into a hackintosh, if only for the experience of running macOS on strong hardware without needing to spend thousands of dollars. Unfortunately, I can't upgrade higher than High Sierra, since I'm using an Nvidia GPU. So I'll have to switch to AMD GPUs at some point down the road. I suppose that's what I get for using proprietary software.
Completely agree. I've become increasingly convinced that we'll need a fundamental restructuring of the academy if we want to fix any of these issues. The fact that places like MIT or Stanford will hire professors simply because they output Nature papers every year, regardless of the fact that they are totally abusive to their trainees, is a perfect demonstration of it.
Is it just me, or does probability theory in general have fairly terrible notation? Ambiguity between random variables and their distributions because of them simply being distinguished by being upper-case or lower-case, writing likelihood functions alternatively with an L() or p(), and using p() (with different arguments) to refer to different probability distributions. Perhaps I'm just having such a difficult time grokking probability theory because it's just difficult stuff, but I often find myself immensely frustrated with the notation.
Haven't used paperlike, but I use a much cheaper matte protector from TechArmor for drawing / taking notes on my iPad Pro and it's great. Would definitely recommend getting something like it. Bare screen's always felt too 'slippery' for me.
It's funny how industries whose entire basis lies in the racialized colonization of the global south become reconstituted in a way that maintains their existence, but further obscures their origin. Perhaps true justice would mean the complete abolition of this industry (and the economic/political structures that brought it about). But instead we now force it to put on more and more masks to justify its continued existence, and in so doing, make the system harder to understand. In my mind this is how liberalism, i.e. the attempt at making capitalism more progressive, can be a double-edged sword.
Past the rebranding, the system is fundamentally the same. We still live under racial capitalism (in which, as many a Marxist have noted, uneven development is a feature, not a bug), and the underlying conditions that cause people to go buy skin lightening creams remain unchanged.
DF is wondrous not just because of what it is (I think it's one of the greatest living examples of the artistic potential of computers as a medium), but also because of how it came about. In a society which does about nil to allow for this kind of deep creative work to flourish (I mean, was Toady basically living on ramen for most of the 2000s?), the Adams brothers set out and did it anyway. And we're getting to see it happen in real time!
Right, this is why the US GDP has grown by a factor of 20 in the past half century while adjusted average hourly wages haven't budged. What a frightfully naive point of view.
I find it incredibly unlikely that we are going to be able to avoid the brunt of climate change. Mass flooding is going to happen and countries like Bangladesh will be swallowed up by the sea. Refugee crises are going to happen at unprecedented scales and there will likely be wars. Even the paltry GND, which is almost certainly not going to be instituted, would not prevent this.
This was inevitable under capitalism. As a system that requires constant growth, it is incredibly brittle and does not lend itself well to long-term, sustainable planning. The only question for me now is: how do we best prepare ourselves for the upcoming crisis, such that we might be able to construct an equitable, sustainable society out of it? Crisis doesn’t guarantee revolutionary change — but it can if we're organized properly.
> Economists use the word "capital" without making reference to Marx. Marx didn't come up with that word, it had a meaning in classical economics since Adam Smith (and maybe before).
How does this refute OP's point? They're really just pointing out how absurd it is that a study on wealth inequality and class fails to even consider the Marxist perspective. Genuine scholarly negligence.
> This is also bizarre. Piketty does have a theory about "why inequality is so rampant in the first place". He just thinks we can change the laws and fix it!
OP isn't arguing that Piketty has no theory -- just that it's a pretty naive one. A Marxist perspective understands that "changing the laws to fix it" is not enough. The state is dependent on the bourgeoisie for its survival (see [0]), and will thus bend to their demands. So long as we have no organized and militant labor movements (and labor parties to represent them), the working class will have neither leverage over the state nor the means to enact their demands.
Unrelated, but the MacRumors thread that's linked to has someone posting about neighbors harassing them by doing things such as "moving my cursor all over the place and speeding up the temperature, causing the fan" and that their "neighbors also learned to slow my internet (ATT DSL) or to disconnect it." [0] Hence their need change their MAC address.
They're now in the process of collecting evidence, and post: "This legal evidence will benefit many, as they boast of being Freemasons, and this being part of their "craft" and their "process". Therefore, when I go public in legal proceeds about their "craft" and "process", while actually forcing those individuals which I have substantial evidence on to appear for the courtroom cameras, the rest of humanity they plague will see the low caliper, socially outcast, inadequates they are. No one need fear them. It's fear they try to attain, and before I'm finished, no one on earth ever will."
> Why not both? Left neocortex hemisphere handles discrete, right handles continuous. Iain McGilchrist makes a convincing argument for that in his book The Master and His Emissary.
I haven't read the book, but this sounds totally absurd. What kind of evidence is there for this?
There's no point in your life where you'll cross some barrier and think "well, I'm successful now." Stop caring about it so much and focus efforts on better things: finding fulfillment in doing good work, nurturing relationships with other people, and enjoying life.
How can you be very privacy conscious yet lack ownership of the phone number that your Telegram account's registered on? Whoever takes ownership of that number can easily lock you out of your account while retaining access to everything you've ever posted.