To add another data point: 4,000 IU (two Kirkland capsules) daily keeps my Vitamin D levels at the high end of the reference range. I also take six Kirkland fish oil capsules which happens to hit the 1,500 mg DHA+EPA target suggested in the article.
Honestly, Costco supplements are hard to beat since they're both USP certified and are usually the cheapest.
I can't speak for the parent, some things that stand out to me
1. k8s and bare metal, when you make a bunch of threads things get slower. with the FIFO case, you can have pending requests in the queue that don't get their connection canceled event, and the same user puts another request in the queue.
2. Silently dropping is bad, you want an alert - really you want an alert when you get close, so you can add more capacity
3. bounded queue with unbounded threads is really just an unbounded queue - a short line with a mob pushing to get in line
Then, you know, memory on k8s, pod gets OOM killed. that sucks cause you have to reschedule and restart. all the pending requests are dropped.
It's very easy to make something that works, but is actually quite detrimental when things are on fire. little extra gasoline helps get over the hills, but when things are on fire, gasoline makes a bigger fire.
Unreal, I recall seeing this a while back but I forgot to bookmark it and couldn't find it again. I really think that the actor model and functional pipelines are the correct way to do concurrency in most software these days. The actor model can be used to encapsulate mutable state when needed and functional pipelines otherwise.
I really think that Mathematica and Matlab failed to capitalize on the rise of AI and ML in the last decade. Seems like Python is the only go-to language these fields.
It's a winner-take-it-all situation, though. R is already struggling as a strong second-placer.
What's more: the numerics stack in JavaScript is being modeled after the Python style, so when we're all dragged kicking and screaming to node (UMAP is already implemented in js and that came out when?), those who have cut their teeth on numpy and pandas will be a little less stranded.
Yes and No. If you ignore Python being a powerhouse in the scripting world and just focus on numerical stuff like matrix and symbolic math, both Python and Mathematica have that. Some people like myself that love Python and know how to use it and Numpy bought Mathematica licenses. I still use Python for scripting, but I've moved most exploratory work to Mathematica. I miss Python's Spyder IDE, but with Mathematica I can enter equations into Notebooks much more elegantly than having to do OO code which looks significantly different than the math.
Well another OO way would be to add a state member to the Heroine class and create a state class for every possible state (X,Y,Z). Each state can then be responsible for interpreting the input and transitioning to the next appropriate state (if any)
The proposed state table is hardly more scalable. It's obviously not infinitely scalable but In my proposed way, you only have to specify the cases where a state transition actually occurs.
one possibility is to use hierarchical-state-machines (HSM), which allow (encourage even ?) the decomposition of state-machine into common behavior for reuse across states.
HSM have been there for a while, being proposed in '84 (!) by David Harel under the name 'statecharts'
Some like to pretend that the economy is a zero-sum game, that in order for some people to get rich, they need to climb on the backs of the poor. In some cases this is true, but in most, wealth isn't transferred, it is created. This is certainly one of those examples.
Except Facebook and Google aren't operating mass incarceration camps that target people solely based on ethnic/religious affiliation. The moral dubiousness of working for Google isn't the same as working as an informant for the Chinese secret police.
Facebook has paramilitary factions genociding the Rohingya? I'm not denying that Facebook played a role in the spread of misinformation and conspiracy theories in Myanmar but to place the blame directly on them is absurd. Facebook is just an ill-moderated medium.
EDIT: This whole idea that Facebook's only sin was not moderating enough, and that they didn't give direct support to the government to help the genocide is just their PR and legal departments going into overdrive because they know they screwed up hard.
I used to think that before I tried using Qt Creator. Your project doesn't need to be a Qt project for you to use it. Make sure you enable the clang code model plugin. It's a bit slow but otherwise flawless.
Certainly willing to give it a try. I can say that I can't imagine any tagging system to be any better than CDT, so at best it may be comparable. Though if the system is comparable with an interface that's faster than Eclipse, it would be quite nice.
Having used both, Qt Creator is so much better than eclipse that it's not even funny. Of course, you first have to forget your eclipse habits and learn the equivalent QtC ones - in particular using the locator (http://doc.qt.io/qtcreator/creator-editor-locator.html), and the available refactorings (http://doc.qt.io/qtcreator/creator-editor-refactoring.html). Recent versions also ship with a clang backend for completion ; it also supports clang-format and clang-tidy natively, with inline fix-its that are shown automatically.
Opinions vary on QT Creator. I personally found it to be hostile to my desires. The little thing that made it unfit for my purpose was that, when I code for Cygwin use, I need my Windows editor to be configurable for Unix conventions, but QT Creator was not configurable to Unix conventions. Eclipse is configurable.
I believe if I had been making programs targeting QT, I might not have run into this issue. But I don't make that kind of program.
Honestly, Costco supplements are hard to beat since they're both USP certified and are usually the cheapest.