people often seem to forget that mining the stuff that makes the batteries is still a thing. That's a whole different ballgame, and however you want to look at: it's finite. Couple the fact that the raw materials often come from less-than-stable geographical regions, it might become Oil 2.0 (with similar wars fought over it).
They probably just lost the ability to build it, or the source code can't be found. Happens quite often. 17 years is a /long/ time to maintain build systems and remember where you put the files.
I don't get this. Dating apps are incredibly popular, and it would be opt-in. If it's the age bucket you're concerned with, I met some of my best friends online during that time (female and male). Adolescence is just like that. I understand that harassment, intimidation and general annoying behavior are a concern. But, if we reached the phase where "two random people can't have a normal conversation if they both wish to" is the default stance then we all need to take a good hard look at ourselves, technology is not going to solve that one.
We have helicopters though, even quadcopters that can carry people, not to mention airplanes. But general purpose "flying cars" are a ridiculous idea: the failure mode is catastrophic (as you pointed out), the noise will be terrible (especially in cities), not to mention pollution (you might not care about CO2, but breathable air is still nice). And there's regulation, since there are no roads you need centralized control (e.g commercial/military flight control systems) which makes it almost impossible to scale and would mean sacrificing to great extend the benefits of cars: autonomy. And above that it solves a weird problem: congestion and capacity, but those are better tackled with self-driving cars and better public transit.
I took GRE without prep as a non-native speaker. Lets just say I severely underestimated it, and went on to do other things in life … the concept of these tests are quite foreign in Northern Europe, so I was not expecting it /at all/. If I wanted to ace it, it would've probably taken me weeks to prep properly (especially all the silly math things, that I mostly forgot about how to do fast enough without mistakes or stressing out). Never ended up submitting my application, even though I had several first author papers as an undergrad, I felt too defeated by it.
Not to mention that the review cycle can take /years/. I've published a paper that was under various stages of review for a year and a half at JAMIA. And, that's actually kinda short. Be prepared to let go, revisit, refine, redefine, etc for potentially a very long time. Often it's not a deal of submit, revise, accepted. It takes time. A lot of it.
That's fair. You kinda get stuck in a certain "mode" though. You start to self-identify with the things you do "I make data science tools for medical informatics". And it's sort of blinding, in a sense. If I enumerate my skills on a CV, the types of companies that would identify with them are in the same space. Doing something else, say games, apps, whatever it is that people "enjoy" interacting with: I'd need to start at the bottom again. No longer senior engineer/team lead ... but rather that young newb. That's frightening too.
It completely locked my drive on a Macbook Pro mid-2014. With no warning or user information, it simply wouldn't accept my password and could not log-in. I had to drop into recovery mode and use the terminal with some obscure commands from a forum post somewhere to unlock it ... talk about user friendly. Apple is really slipping in terms of QA here.
I'd probably swap the mount for an equatorial mount (Sky-Watcher EQ6 or similar), and use an off-axis guider, with a modern CMOS like the ZWO-ASI or one of the QHYs as imaging camera (much cheaper than the CCDs, and just as good if not better). At prime focus (i.e. without a focal reducer or Hyperstar-like system) getting pinpoint guiding with a secondary guidescope is tricky, but an OAG does wonders...starlight express makes the best in my opinion. Optically the C8 is still fine, you might want to re-collimate it, but that's a bit of an after thought. Calculating the backfocus for your optical train might be tricky. But once you have it set-up on a computerized mount (getting plate solving to work would probably also help, I can recommend Sequence Generator Pro) then you're good to go!
Yep! Hobbies I guess. It's a C8 EdgeHD, with the secondary replaced by the Hyperstar [1] which transforms the Schmidt-Cassegrain into a Schmidt camera like system (in my case 425mm f/2.1), in that configuration it's easier to trade sky noise for read noise since the sub exposures can be very short. I might one day trade it for the RASA (Rowe-Ackermann Schmidt Astrograph), but I already have a 12" Newton to maintain in Nerpio!
Living in the Netherlands, the light pollution saddens me [1]. There are no dark places anymore. I tried astrophotograpy with a modest telescope from an inner city. With some success [2], but it requires a stunning amount of technology [3]. Still, being able to know that the night is still out there ... even if it's invisible to the eye, is simply amazing.
These days however, I remote control a telescope in Southern Spain. The difference is well, night and day [4]. But sometimes, when the clouds move ... I still setup the scope at home. Just to remind myself that the stars are still there, hiding behind the curtain of industrialization.
Note that it does seem to work reasonably well for the "adult" industry. People are often reluctant to pay for recurring fees on those sites (social stigma plays a role there, probably). But "pay-as-you-go" or pay per view schemes seem to fare far better. And, it makes sense; why pay for a monthly thing when the moment of enjoyment is relatively short. A DRM free download for that thing you like is mentally easier to justify. The big hurdle is still sign-up and credit cards I think, I'd personally be much more inclined to use systems with more privacy conscious approaches: since micropayments give an opportunity to track /exactly/ what you're looking at. This might even be more relevant for news outlets that cover a wide range of topics and writing styles.
Except it isn't really. Yes, Python is incredibly slow for day to day stuff, but the sheer amount of easy to use numerical libraries (numpy, scipy, scikit-learn, tensorflow, keras, opencv, just to name a few) make it one of the fastest out there for numerical computation. I tried doing some numerical heavy stuff on the JVM (with Java and Clojure) and it fights you every step of the way ... and that has static typing and all the things the article mentions. Of course, Python derives that functionality from C and Fortran … but just having that interop at your fingertips is magical in terms of productivity. I still get nightmares from working with the JNI.
Yep. I agree. Although maybe XML schema (https://www.w3.org/XML/Schema) -> JSON Schema. And then there's also XQuery, DTDs, namespaces, entities,… which are also in need of reinventing I guess.
What exactly is Watson except a marketing term? I mean, as far as I understood it, it's Apache UIMA with some bare bones NLP bolted on, using RDF ontologies to power some inference. But I've never seen it being used in the wild except as a tool to win over people to the IBM-side, raking in large contracts ultimately doomed to fail due to lack of domain knowledge. It all seems so ... familiar. Now everyone is rushing into Deep Learning for NLP, but apart from things like sentiment analysis on synthetic data sets ... it's still /hard/ and needs constant attention. I really don't see how you can "IBM Watson" stuff unless you're already heavily invested on integrating NLP/ML into your workflow ... and that's tricker than it looks. The stuff on BlueMix seems nice and all, but far from a silver bullet or golden hammer when it gets down to the nitty gritty of real world data.
It's very impressive. But, I'm curious about some of the technical details. For each of these representations you basically end up with a dense vector on a song level. Which I assume you would then kNN with a user specific vector. But I've never come across a nice kNN data structure that supports high dimensional vectors in a larger than memory setting whilst supporting updates. Spotifys own Annoy is cool https://github.com/spotify/annoy, but changing or adding a song requires rebuilding the whole structure ... surely that's prohibitive at scale?
Just tried it with the Nightly by setting gfx.webrender.enabled to true in about:config. Wow, that thing flies. It's seriously amazing. And so far no bugs or visual inconsistencies I could detect. Firefox is really making great progress on this front!
This is so exciting! The field of synthetic biology is advancing rapidly, and apart from the social stigma at times: it's a really promising technology. When I was still an undergrad I was lucky enough to participate in the International Genetically Engineered Machine (iGEM) competition. But that was back in 2010, a lot has changed since then!
Another interesting tidbit is that Evidence Based Medicine, what we've come to known as randomized controlled trials with systematic reviews, is /much/ younger than people think. The movement properly started in the 1960s, but only came to fruition after Archie Cochrane (1970s). Still Evidence Based Practice is a contested thing, and very few systematic reviews eventually end up informing policy.
better video chat with screen share abilities and decent noise cancellation. We tried Skype, Zoom.us, Slack video, Google Hangouts and WebEx … they all have issues and none of them work reliable enough even with decent internet connections across all three major operating systems. If we would need custom hardware for that: we'd even buy it, a good high quality webcam and headset would go a long way.
I honestly can't believe that's the biggest problem we've had so far, other than that it has actually been a smooth ride.
I just wish they would sell mono cameras more (e.g. without the Bayer mask). Heck, I don't even need the camera. Just sell the chip and let companies like Atik, StarlightExpress and QHY pick them up.
That's not true. Old mercury and sodium lights are much easier to filter since their emission lines are narrow. LED often has a very broad and spiky spectrum, which makes it almost impossible to filter. (Source: astrophotographer). You can buy reliable light suppression filters for sodium/mercury (google CLS or UHC filter), but not LED