Now their procedures require them to put their own lives above all others, including use of lethal force on the slight chance their lives are in danger (such as the shooting of the kid with the toy gun).
....
Soliders are heroes. They put themselves in danger.
Except that soldiers use lethal force to protect themselves far more often than the police do. Soldiers in Iraq shot unhappy civilians because they had no way of knowing they weren't suicide bombers... of course they didn't, but that's always the way. Look at the paranoia in Vietnam, and the impunity that soldiers and officers had to shoot any Vietnamese person on the grounds that they might possibly be Viet Minh soldiers (note, Viet "Cong" was an invented label to make them seem scarier and justify shooting them).
When NATO intervened in the Kosovo crisis, we sat back and threw stones over the wall (and hit many wrong targets) because while we had to do something, "our boys'" lives were too valuable to risk on ground assault. Thousands and thousands of civilians die at the hands of our troops because our military values the lives of soldiers over the lives of civilians (see also the difference in sentence between "unjustified killing" by a soldier and "murder" by a civilian).
Are those extremely intelligent people who've been reviewing and editing professional journals for many years? Because I think these guys have a fairly good idea how journal publishing works by now.
Elsevier have run through the market acquiring the most prestigious journals with the express intention of achieving a near monopoly and the opportunity to milk the universities with ever higher fees. As I said, the guys doing this have years of experience in academic publishing -- I'll trust their judgement, thanks.
While the free mass market can bulldoze through with more marketing, in the realm of journals, reputation is everything. Anyone in the field will understand that Lingua has become a different journal overnight, and the editors will take their well-earned reputations with them to Glossa.
A whole 22 cents per person per year for a subscription. Very expensive.
That's only a fair calculation if every faculty member and every student is going to make use of the subscription. When I did CS at uni, I don't recall once reading a single journal article.
Well, yes. That's precisely what they've said they're going to do, and given that they are all remarkably intelligent people, I think they've already done the sums on the hosting costs. They certainly know how much time is involved in it, seeing as how they've been doing that exact job for years.
So if people don't know about George Boole's disproven theory of human thought which accidentally triggered the information revolution, they won't be able to vote responsibly?!?
No-one's advocating switching from Windows to Linux here. Google is working on a niche here, and trying to expand it. I believe that the niche could be expanded even further if better document editing was available, but I don't think it's going to wipe out Windows and Office.
I'm all for alternatives, but Office compatibility tends to make or break computing devices.
Except that MS Office for Android isn't feature-complete. If LO/OO can reengineer the full suite to trim out the fat and make them fully tablet friendly, there would be something there.
Ah, but MS ended up looking like they were reducing functionality, and they were imposing an unfamiliar workflow on users. The ChromeAndroidOSBook wouldn't be looking to replace Windows, but ChromeOS, so you're not going to have office workers bitching about the ridiculous new save dialog in Word. More importantly, Android is in very common use, so the workflow is already familiar, unlike the barely-used Windows Phone UI that became "Metro". Microsoft's goal was to use ubiquitous desktop Windows to get people familiar with the Phone UI and therefore sell more Windows phones. That's exactly the opposite of what Alphabetti Spagoogley are up to.
Well, if the merged platform has ChromeOS as a base layer and the Dalvik VM and Android widgets only being booted when first invoked, that's not going to be a problem: you'll only need to chew up the cycles when you need an app that isn't available for the ChromeOS layer anyway, and ChromeOS was never really targeted as an app platform to begin with.
The app ecosystem's the thing, and while many can survive with Google Docs, some of us need offline tools. If this spurs a proper overhaul of LibreOffice or Apache OpenOffice to a lightweight, Android-friendly version, we could be on the cusp of something very interesting indeed.
The thing with Scheme's mutability is that it's got to be an explicit choice by the programmer. Mutability-by-default leads to unintended changes of value (eg C if(c=0)) which is one of the commonest sources of error, and often the hardest to debug, and the biggest practical advantage of FP to me is that immutability-by-default vastly reduces the opportunities for such error.
It's not black-and-white thinking, it's what's-the-point thinking. When I was at university, I struggled to see the practical purpose behind FP, and part of the problem was the problem set we were given didn't naturally show the benefits of FP, and part of the problem was the gaps in SML's functional model undermined what should have been the whole point of FP.
As a student I always had difficulties learning to do something a particular way when there seemed to be a better or more appropriate way to get the same results more efficiently -- human learning is all geared to getting results the easiest way. As a student teacher, I'm now spending a lot of time trying to avoid my students suffering the same frustrations as me.
If it took you 1 day in.BAT programming, then it would take about 20 seconds in BASH/sh/csh... If you seriously did text processing in a bat file, I commend you, but the power of the *box command line *will* blow your fucking mind. I used to do bat programming too when I was like 12, but once I moved to Linux (bsd actually) I couldn't believe I ever wasted my time w errorlevel goto set blah blah...seriously tho, go look at what sed and awk can do for text processing...you'll never want to use a BAT file again
Do you know what might blow your fucking mind? The idea that when working in a corporate Windows SMS environment you are kind of likely to encounter Windows machines, and getting Cygwin approved for install on your domain controller is going to take a lot longer than simply using the tools available to hand. Now excuse me, I'm off to do test some theories in computational linguistics in Prolog. So don't start getting all smart about "real" tools with me, sonny.
First-class functions is only a tiny part of it. Functional programming was a movement to make computer programming better match classical maths, and determinism is a huge part of that. If you throw away immutability, you've lost practically everything that made FP a worthwhile endeavour (like provability of code).
Academic: does this theoretical problem exist?
Market: download the free app.
Oooohh... if only. I've been looking for various things, free OR paid, on my iPad, and many aren't there. Or if they exist, the basic function is trapped in some webservice somewhere.
Case in point: pencasts. I have an iPad, and I have a stylus for touchscreen. I have a microphone. I want to make little video clips that I can mix with camera videos to create little educational shorts. I've found pencast software that works with an expensive proprietary digital pen that requires expensive custom-printed digital paper. I've found a free package that allows me to dynamically resize, scroll and write on images while simultaneously drawing over them -- but the output is only available via a webplayer on the company's website. So I'm now just looking at going low-tech and mounting my ipad over a desk.
The app market is really good at generating millions of stupid games, but very few niches are being genuinely filled.
Javascript has horrible scoping rules that result in really weird behaviour if you forget to var something (particularly a problem for loop indexes) -- I wouldn't put a beginner anywhere near it.
You cannot teach functional programming in JS, because JS is not functional. Not only does it not have immutability, but the scoping problems mentioned in the last paragraph mean you've also got global variables, which is the polar opposite of immutability, if such a concept makes any sense.
Any job that involves sitting at a computer leads to days when knowing how to automate a simple task is beneficial. My big brother's doing data entry, and he's been battering together VBA scripts (VBA because Excel is the nearest thing to a programming environment on his client's computers) to generate search strings that account for common mistypings of address formats etc, making him one of the quickest on his team. I did a 3-day data filtering job in 1 day because I scripted in in Windows.BAT rather than manually searching through the data. I'm now training to be a school teacher, and I'm programming my own web-pages to make use of the interactive whiteboard in a way that's natural to me and keeps me in control of everything happening in the class. Everyone can use some programming now and again.
The thing that BASIC got right was simple IO: PRINT and INPUT. The reason "real" languages are usually unfriendly to beginners is the amount of hoops you have to jump through to get that input and output. Back when C was the language of choice for CS classes, I think the mistake was to start with "hello world" in C -- they should have started with BASH scripting and then introduced C in its original natural context -- writing commands to extend the shell. Input and output handling could have come later.
Now, the problem with saying "we should go back to BASIC" is that people's expected I/O channel is not a scrolling text shell window, so PRINT and INPUT are no longer adequate. As they are the only specifically meritorious features of standard BASIC dialects, suddenly, BASIC is valueless. We can fit any language with a dialog "This is equivalent to print." and variable = inputDialog "Enter number of whatever:" and we're just as well off as we would be doing that to BASIC.
Making a "simple" language is hardly even a question of language architecture (although JS's variable scoping rules are a right pain), because most are largely straightforward, if you leave OO models out of it. It's the general purpose nature of modern languages, and thus the reliance on libraries, that makes languages unfriendly to beginners (which also drags OO models back into the equation). Scratch isn't considered good for learners because of its programming model (it's just a standard C-alike, really) but because it has a specific purpose and therefore a very limited set of actions.
It's a difficult one, because the contract says "reasonably available". What is a reasonable availability to do unpaid work for a company that made you redundant? Some would say "zero". I would be one of them. The former employer would have to stand up in court and define "reasonable" before they could do anything else. No-one's going to Streisand themselves on this one -- it's an unenforceable term, and they'd be mad to chase anyone legally over it, but it'll be a good opening gambit if and when they need someone to come in.
Thinking about it, it's quite possible that the point of the exercise is to torpedo attempts by middle managers to bring back experienced (previously redundant) hands as contract consultants, which is kind of cutting off your nose to spite your face.
Slashdot Mirror
Now their procedures require them to put their own lives above all others, including use of lethal force on the slight chance their lives are in danger (such as the shooting of the kid with the toy gun).
....
Soliders are heroes. They put themselves in danger.
Except that soldiers use lethal force to protect themselves far more often than the police do. Soldiers in Iraq shot unhappy civilians because they had no way of knowing they weren't suicide bombers... of course they didn't, but that's always the way. Look at the paranoia in Vietnam, and the impunity that soldiers and officers had to shoot any Vietnamese person on the grounds that they might possibly be Viet Minh soldiers (note, Viet "Cong" was an invented label to make them seem scarier and justify shooting them).
When NATO intervened in the Kosovo crisis, we sat back and threw stones over the wall (and hit many wrong targets) because while we had to do something, "our boys'" lives were too valuable to risk on ground assault. Thousands and thousands of civilians die at the hands of our troops because our military values the lives of soldiers over the lives of civilians (see also the difference in sentence between "unjustified killing" by a soldier and "murder" by a civilian).
I think our anonymous friend was being sarcastic.
Are those extremely intelligent people who've been reviewing and editing professional journals for many years? Because I think these guys have a fairly good idea how journal publishing works by now.
Elsevier have run through the market acquiring the most prestigious journals with the express intention of achieving a near monopoly and the opportunity to milk the universities with ever higher fees. As I said, the guys doing this have years of experience in academic publishing -- I'll trust their judgement, thanks.
While the free mass market can bulldoze through with more marketing, in the realm of journals, reputation is everything. Anyone in the field will understand that Lingua has become a different journal overnight, and the editors will take their well-earned reputations with them to Glossa.
A whole 22 cents per person per year for a subscription. Very expensive.
That's only a fair calculation if every faculty member and every student is going to make use of the subscription. When I did CS at uni, I don't recall once reading a single journal article.
"let them start their own"
Well, yes. That's precisely what they've said they're going to do, and given that they are all remarkably intelligent people, I think they've already done the sums on the hosting costs. They certainly know how much time is involved in it, seeing as how they've been doing that exact job for years.
Syllogisms are all about "all Xs are Ys" and "some Xs are Ys", aren't they? That's just implication ("->" in Boolean algebra) -- no ands or ors.
So if people don't know about George Boole's disproven theory of human thought which accidentally triggered the information revolution, they won't be able to vote responsibly?!?
No-one's advocating switching from Windows to Linux here. Google is working on a niche here, and trying to expand it. I believe that the niche could be expanded even further if better document editing was available, but I don't think it's going to wipe out Windows and Office.
Or, you know, Office
I'm all for alternatives, but Office compatibility tends to make or break computing devices.
Except that MS Office for Android isn't feature-complete. If LO/OO can reengineer the full suite to trim out the fat and make them fully tablet friendly, there would be something there.
Ah, but MS ended up looking like they were reducing functionality, and they were imposing an unfamiliar workflow on users. The ChromeAndroidOSBook wouldn't be looking to replace Windows, but ChromeOS, so you're not going to have office workers bitching about the ridiculous new save dialog in Word. More importantly, Android is in very common use, so the workflow is already familiar, unlike the barely-used Windows Phone UI that became "Metro". Microsoft's goal was to use ubiquitous desktop Windows to get people familiar with the Phone UI and therefore sell more Windows phones. That's exactly the opposite of what Alphabetti Spagoogley are up to.
Well, if the merged platform has ChromeOS as a base layer and the Dalvik VM and Android widgets only being booted when first invoked, that's not going to be a problem: you'll only need to chew up the cycles when you need an app that isn't available for the ChromeOS layer anyway, and ChromeOS was never really targeted as an app platform to begin with.
The app ecosystem's the thing, and while many can survive with Google Docs, some of us need offline tools. If this spurs a proper overhaul of LibreOffice or Apache OpenOffice to a lightweight, Android-friendly version, we could be on the cusp of something very interesting indeed.
The thing with Scheme's mutability is that it's got to be an explicit choice by the programmer. Mutability-by-default leads to unintended changes of value (eg C if(c=0)) which is one of the commonest sources of error, and often the hardest to debug, and the biggest practical advantage of FP to me is that immutability-by-default vastly reduces the opportunities for such error.
It's not black-and-white thinking, it's what's-the-point thinking. When I was at university, I struggled to see the practical purpose behind FP, and part of the problem was the problem set we were given didn't naturally show the benefits of FP, and part of the problem was the gaps in SML's functional model undermined what should have been the whole point of FP.
As a student I always had difficulties learning to do something a particular way when there seemed to be a better or more appropriate way to get the same results more efficiently -- human learning is all geared to getting results the easiest way. As a student teacher, I'm now spending a lot of time trying to avoid my students suffering the same frustrations as me.
If it took you 1 day in .BAT programming, then it would take about 20 seconds in BASH/sh/csh ... If you seriously did text processing in a bat file, I commend you, but the power of the *box command line *will* blow your fucking mind. I used to do bat programming too when I was like 12, but once I moved to Linux (bsd actually) I couldn't believe I ever wasted my time w errorlevel goto set blah blah...seriously tho, go look at what sed and awk can do for text processing...you'll never want to use a BAT file again
Do you know what might blow your fucking mind? The idea that when working in a corporate Windows SMS environment you are kind of likely to encounter Windows machines, and getting Cygwin approved for install on your domain controller is going to take a lot longer than simply using the tools available to hand. Now excuse me, I'm off to do test some theories in computational linguistics in Prolog. So don't start getting all smart about "real" tools with me, sonny.
First-class functions is only a tiny part of it. Functional programming was a movement to make computer programming better match classical maths, and determinism is a huge part of that. If you throw away immutability, you've lost practically everything that made FP a worthwhile endeavour (like provability of code).
Academic: does this theoretical problem exist?
Market: download the free app.
Oooohh... if only. I've been looking for various things, free OR paid, on my iPad, and many aren't there. Or if they exist, the basic function is trapped in some webservice somewhere.
Case in point: pencasts. I have an iPad, and I have a stylus for touchscreen. I have a microphone. I want to make little video clips that I can mix with camera videos to create little educational shorts. I've found pencast software that works with an expensive proprietary digital pen that requires expensive custom-printed digital paper. I've found a free package that allows me to dynamically resize, scroll and write on images while simultaneously drawing over them -- but the output is only available via a webplayer on the company's website. So I'm now just looking at going low-tech and mounting my ipad over a desk.
The app market is really good at generating millions of stupid games, but very few niches are being genuinely filled.
Javascript has horrible scoping rules that result in really weird behaviour if you forget to var something (particularly a problem for loop indexes) -- I wouldn't put a beginner anywhere near it.
You cannot teach functional programming in JS, because JS is not functional. Not only does it not have immutability, but the scoping problems mentioned in the last paragraph mean you've also got global variables, which is the polar opposite of immutability, if such a concept makes any sense.
Any job that involves sitting at a computer leads to days when knowing how to automate a simple task is beneficial. My big brother's doing data entry, and he's been battering together VBA scripts (VBA because Excel is the nearest thing to a programming environment on his client's computers) to generate search strings that account for common mistypings of address formats etc, making him one of the quickest on his team. I did a 3-day data filtering job in 1 day because I scripted in in Windows .BAT rather than manually searching through the data. I'm now training to be a school teacher, and I'm programming my own web-pages to make use of the interactive whiteboard in a way that's natural to me and keeps me in control of everything happening in the class. Everyone can use some programming now and again.
The thing that BASIC got right was simple IO: PRINT and INPUT. The reason "real" languages are usually unfriendly to beginners is the amount of hoops you have to jump through to get that input and output. Back when C was the language of choice for CS classes, I think the mistake was to start with "hello world" in C -- they should have started with BASH scripting and then introduced C in its original natural context -- writing commands to extend the shell. Input and output handling could have come later.
Now, the problem with saying "we should go back to BASIC" is that people's expected I/O channel is not a scrolling text shell window, so PRINT and INPUT are no longer adequate. As they are the only specifically meritorious features of standard BASIC dialects, suddenly, BASIC is valueless. We can fit any language with a dialog "This is equivalent to print." and variable = inputDialog "Enter number of whatever:" and we're just as well off as we would be doing that to BASIC.
Making a "simple" language is hardly even a question of language architecture (although JS's variable scoping rules are a right pain), because most are largely straightforward, if you leave OO models out of it. It's the general purpose nature of modern languages, and thus the reliance on libraries, that makes languages unfriendly to beginners (which also drags OO models back into the equation). Scratch isn't considered good for learners because of its programming model (it's just a standard C-alike, really) but because it has a specific purpose and therefore a very limited set of actions.
Just don't grow tobacco - getting pu239 traces in your lungs could give you cancer.
Most people smoking tobacco are already disregarding a well known cancer risk....
Do I get to sue if the stress causes me to comfort-eat, leading to late-onset diabetes...?
Thinking about it, it's quite possible that the point of the exercise is to torpedo attempts by middle managers to bring back experienced (previously redundant) hands as contract consultants, which is kind of cutting off your nose to spite your face.