> "cops are disproportionately killing black kids", then that was disproven
Was that disproven? The studies that I've read have shown disproportionate police shooting fatality rates among African American men, especially in the unarmed case when adjusted for population. [1]
The unarmed case is the one I pay attention to the most as I think that it (mostly?) filters out cases where shootings were justified.
I'd like to keep an open mind so I welcome any reading that has disproven the disproportionate argument.
When I looked at that data source last year, I read through every case of unarmed black males killed by police. Many of them were actually armed, but with a tree branch or some other blunt object which weren't considered to be weapons. Some were truly unarmed but clearly attacking police (think Michael Brown). A few others were bystanders hit by a stray bullet as police shot at armed suspects (clearly not racism). If memory serves, there were only 12 national police shootings of African Americans where the police shooting was likely malicious. I think the number for whites was comparably low, and I think these are both too low to be statistically useful.
There have been a few others as well. I should also clarify that my use of "disproven" is too strong. I should have said "the bulk of the evidence supports a contradictory conclusion" or some such.
> The unarmed case is the one I pay attention to the most as I think that it (mostly?) filters out cases where shootings were justified.
An armed person can be shot and it can still be unjustified. The presence of a gun does not mean there was any danger or even the appearance of danger.
An unarmed person can give the appearance of having a weapon or may pose a clear threat regardless and such a killing could be justified.
I think mostly paying attention to the unarmed case allows the armed case to mask any potential problems, while also includes the implicit assumption that unarmed cases are mostly unjustified will exaggerate the appearance of problems.
You really can't, because black people are arrested, tried and convicted WAY disproportionately to the actual rate of crimes. eg white people smoke weed as much or more than black people, and are arrested at a far lower rate for it.
These statistics don't involve a trial or conviction. It's based entirely on the police report when the crime was reported and only includes the race of the offender when that information is known.
A minority of police shootings involve violent crime. You will also notice that nonviolent crimes are the biggest issue, with BLM most concerned about things like Eric Garner or Tamir Rice, who was twelve years old.
"Perusal of Table 4 reveals that suspects were an average age of 36, with whites somewhat older
than blacks. Thirty-nine percent of the suspects were involved in a violent crime, 17% in a property
crime, and 5% in a drug crime. Hispanics were less likely to be involved in a violent crime,
while blacks were more likely to be involved in a property crime than whites or Hispanics.
Blacks are least likely to be armed."
Per my sibling comment, laypeople often assume the UCR cannot be trustworthy because it is collected, compiled, and analyzed by policing agencies; however, it's data are supported by the national crime victimization survey, whose data comes from victims and is collected and compiled by the justice department. So your point is even stronger than you made it out to appear. :)
The national crime victimization survey largely agreed with the FBI's Unified Crime Report (on matters of race and violent crimes, anyway). The latter is data collected by policing agencies and the former comes from victims. If law enforcement discriminated against Blacks, we would expect a discrepancy between these reports (it's possible that the victims are simply "equally racist", but given that the victims of black offenders are themselves overwhelmingly black, this is unlikely). Also, there have been a few other studies like this one[1] which don't find anti-black bias in sentencing, conviction, etc. This doesn't mean it doesn't exist, just that the conclusion of a racist criminal justice system is not entirely supported by the evidence.
One anecdote about the weed thing, since that comes up so often--I live in Chicago and apart from festivals, I never see white people strolling around downtown smoking weed, but I do see this probably once a month, and the user is consistently black. The white people I know who use marijuana do so at home and out of public view. This is just an anecdote, but cultural differences in usage could better explain the arrest discrepancy than police racism. At least I would like to discontinue the trend of assuming that every gap is necessarily and automatically attributable to discrimination.
"The Times analysis found that among the largest police departments in each of those four states, black drivers were between 1.5 and 5.2 times more likely to have their cars searched than white drivers. These searches occur with the consent of the driver, so the officer doesn't need to meet any legal standard, like probable cause, to initiate one. [...] More recent figures from 2014 and 2015 published by the New York Times show blacks who are searched are around 20 percent less likely to be carrying contraband than whites who are searched. "
Studies have also shown that disproportionate police shooting fatality rates among men when adjusted for population (50% of the population are men but 96% of those shot by police are men).
Violent crime rates are the more useful base rate than the general population. This is the best study I've seen to date that attempts to take into account rates of violent crime: https://www.nber.org/papers/w22399
For me, the full-measure solution would be to provide a native port of Visual Studio for Linux. I understand that this is not practical on many, many levels, but if Microsoft were to ever take on that herculean effort they would win me over in terms of my choice for dev tools under Linux.
For now I'll have my Windows stuff in Windows and my Linux stuff in my Linux partition. I do look forward to checking out the linux-in-windows (dynamic syscall translation) feature soon though.
No, but Visual Studio Code is off to a very good start, and it runs well on Windows, Mac, and Linux, all free and open source. It's not a "rewrite" per se, but a new product with the same name but targeted at developers on all platforms. They are talking about adding enough serious IDE power over time (and the original VS shows that they know what it takes) that they could become a serious power in the open, cross-platform dev world.
They would not need to port all the Windows-specific functionality that has accreted on Visual Studio over the years to become the "new Visual Studio" for developers on all platforms.
Not necessarily. Getting it to work with Wine is probably easier.
I've spent many hours running Publisher in CrossOver Office and never ran into any problems, so this sort of thing is certainly possible for complex apps.
Has anyone spent a lot of time with VS Code? I tried it a while back when it was first announced and have not found a reason to re-visit it yet. At the time it felt like a sublime-text alternative instead of an IDE (was it always positioned to be just an editor?) Always great to see more options though.
It is far from perfect, but it proved to me you can build a modern editor on top of chromium without killing my battery and cpu, unlike, you know, atom.
I never thought I'd ditch Sublime Text after using it almost exclusively since TextMate (so for the past few years). I especially never thought I'd ditch it for a Microsoft product. And perhaps most importantly, I never thought I'd use an editor that didn't have Vim keybindings (yet! it's at the top of their list, apparently).
But after using VS Code for a while now I'm pretty sure I'll keep using it, at least for Node.js development. The IntelliSense support through TypeScript definitions, the out-of-the-box support for what I consider essentials (linting, task running, emmett, etc.), and perhaps most importantly the built-in debugger make my life significantly easier.
I enjoyed working with node-inspector, but the debugging in VS Code is a lot more convenient (mostly).
To be fair, for non-node development I'd say VS Code is not yet good enough to replace Sublime Text, but that might change when enough plugins and features have been added.
What it boils down to for me is that VS Code somehow seems to be optimized exactly for my workflow. It doesn't have tons of buttons and interface elements that I don't need (as many IDE's do), and it isn't slow. But on the other hand it offers very sensible defaults that other editors such as Atom and Sublime don't have. On both those editors I've had some issues with setting up linting, debugging, IntelliSense, and a few misc. other things. The main reason for that, I suppose, is that they're all plugins developed by different people. I like how these just come with VS Code and work right away.
I'm rather baffled that I'm saying all this, considering my general aversion of using Microsoft products (even if this is based on the past). But it is what it is. VS Code is a huge timesaver for me, and (mostly) hits the sweet spot between plain, fast editor and IDE.
It is my main IDE for Typescript since v0.1. VSCode in combination with TypeScript provides a superior Frontend-Dev experience to any other Language - IDE combination in my opinion.
Apart from an occasional restart required it has been pretty solid. Performance(autocompletion etc) is very good and for TypeScript development, VSCode is pretty feature complete...
I think that's exactly what it is trying to be: a simple editor with some capabilities.
A lot of people expect a "real IDE" to have a lot of fancy, specific, semi-proprietary features, but what I believe VSC is providing great editing capabilities and offloading everything else to good task managers (and now, extensions). A lot that you can do in "real IDEs" (like building your project) can be handled by Gulp, Grunt, or other runners in a more or less independent way. I actually prefer it this way now, regardless of the IDE I use - that allows my projects to be independent of the IDE.
To me, where VSC shines is really in editing comfort, be it with the super fast typing response (really!), or with the Git integration, or with how it handles work space for opened files. It looks like it is supposed to be uncomplicated.
This is one that I admit is fairly subjective, and I apologize in advance if this is not scientific enough, but: it just feels right.
I think they optimized the way code drawing is done in such a way that it just happens with the least amount of time possible between typing and drawing. The end result is the same as if you typed anywhere else, but because the response is cut by, say, 16ms, it just feels fast. I remember the first time I tried VSC - I almost felt like the letters where coming out from under the cursor ahead of time! It was a weird moment.
The counterpoint is IntelliJ. It's not a slow IDE (I used Eclipse before so that'd be unfair to say), but maybe it has a couple of frames too many before drawing what's typed, so it feels like it's lagging in comparison. Maybe why they're making it a point of future releases[1] to reduce any typing lag.
I've been having some good discussions about this exact topic lately. I'm glad you brought it up.
After mulling it over with some friends, I think it would be awesome to have a community gathering where the principles of its "teachings" are based on respect for the earth, each other, and critical thinking.
Instead of typical weekly worship, it could be replaced by something like "Science Sundays" - where folks can bring their kids for a lecture and experiment that the family can do. If one can find enough local individuals who could serve as advocates then I think something like this is possible. Although, I don't what spark it would need for it to reach critical mass.
The closest thing that we've identified as such a group is the Maker community. One problem that we've had in these discussions is that any time we try to add structure to the idea (ie, a common reference book, hierarchy) we immediately realize that it sounds like it would be mistaken for a cult. :( http://abstrusegoose.com/31
What they're saying is that their underlying hardware has been selected such that it can be used with the Arduino programming environment to target it as if it were a regular Arduino. So it is compatible with Arduino software and tools.
