Exactly. I learned this with the OrangePi boards. And my 'kodi boxes' are retired in favour of Rpi2's (or pi3's, but pi3's slightly higher power needs are a annoyance so far as what USB sockets you can run them off).
Both religious aversion to science, as we see in some, and also an equally worrying trend of memorizing what is needed for exams only as long as said exams are on the horizon, are symptoms of a common anti-pattern in education.
We should not spoon feed children facts, or purported facts, or disproved 'facts'. What should be taught are the generic skills required to problem-solve, research, fact check, and basically work stuff out for yourself. Importantly, the engineering-like idea that a chain is only as strong as its weakest link, and that weakest link is only as strong as its sternest test. And teach these ideas first in the context of practical engineering and problem solving. Let religious nuts drive 'genesis as literal' ideas all they want. With the above skills well trained, the religious ideas, free from the medieval risks of burning at the stake for heresy, will just seen too silly to too many. And the class time will be better spent than merely spoonfeeding a naive and simplistic picture of how evolution actually works in practice.
Humans are all too easily addicted to the Shiny Thing, all too easily placated, and all too easily rendered docile, like a chickens neck is stroked before being broken. Humans are too content to be made cattle, and look where allowing themselves to be cattle farmed led Cowkind. #Moo
You want my shitty McJob Mr Robot, you can have my shitty McJob. Really, be my guest. Computers replaced the need for a room full of well-caned schoolboys to do sums. We just need to have less fucked up attitudes about business and ownership when it comes to robots.
If you value security, don't run the mission-critical parts of your infrastructure on a general purpose operating system like Windows, but rather run it on a minimalist, locked-down OS that has _only_ the facilities needed to do its job. The update carousel is a nightmare. If you want to ensure your Windows box doesn't sporadically reboot during a long unattended operation in order to update, what do you do? If you want to lock Windows down so it can only do the job to hand, and nothing else, you're screwed. If you run mission-critical stuff on a full-featured general purpose OS (and the same can be said for off-the-shelf Linux distros like Ubuntu and Fedora), you are kinda asking for it.
That this idea is older than me, but is ignored, is laughable.
By heavily marketing Microsoft Windows to the point that it is used, in a capacity where it can run things like Minecraft, in mission-critical IT infrastructure, they have done much to bring the current situation about. Mission critical IT infrastructure should be decomposed as a system of well-defined, hardware-isolated roles, each of which has only the authority necessary to do its job, and nothing more. (This is the principle of least authority.) There is more profit for Microsoft and major IT consultancies in just pushing Windows. Indeed Linux, in its 'desktop' flavour is no better. But Linux, being open-source, is sufficiently customisable that, as in Android or embedded uses, you can remove as much as you like.
For example, there is no need, in a patient records system, for the facility to arbitrarily create, overwrite, and delete files. If you have one machine that stores important details, another that categorises records stored by the first, and another that reads back the result, and can do nothing else (such as run Microsoft Word or Minecraft), then there is simply far less to go wrong. But systems need to be architected around this. The current trend to maximise 'bang for buck' has led to maximising flexibility and agility and, with it, maximising the flexibility and and agility offered to attackers and, thus, maximising vulnerability.
Microsoft and other proprietary software vendors, in pursuing their market positions, have done much to bring this situation about, and only when we learn that a general purpose OS is not a good idea for actually running mission-critical infrastructure (even while they are great for designing and programming them), will we start to get out of this mess of 'cyber insecurity' that we find ourselves.
When will people get it that, with a mission-critical computer system, it should have no more ability or authority to do _anything_ than it needs. If you computer is only there to do your financial stuff, the it doesn't need to be able to run Minecraft, so it should not be able to run Minecraft at all. Having a single all-things-to-all-people OS that, once booted, can do anything and everything, and is so complex that even its manufacturer can't track all the bugs and holes, and nobody else can even tell if it works, just so MS and other vendors have hid-e-holes to put their copy protection stuff in, and you can run Word, Minecraft and watch your Kitten videos on the same machine... that is just plain fing asking for it. (Linux isn't really much better _except_ that you can, and people do, produce tailored versions with extraneous stuff removed, and if you want to see the code for every last bit of software running, you can.)
Principle of least privilege (or least authority). For mission-critical stuff this is a must, and that precludes a general purpose OS like Windows or a typical Linux distro.
Suppose Facebook seek a court order in the US, say, that the Austrian court has no jurisdiction with regards to Facebook posts outside of Austria. What then?
The trouble here is that the study hasn't controlled for the _quality_ of code written, and the knock-on consequences. If, for some reason, the female part of the workforce that facebook has happens to be producing lower quality code compared to the average (and _why_ this is the case then needs to be looked into, not simply put down to gender), then naturally the knock-on effect of that is, assuming rejection of code is fair, both women rising more slowly through software engineering ranks (assuming the ranks are done solely on merit), and also women's code being rejected more. Each of these possibilities leads to more questions, and the trouble is that you have an awkward choice of either publishing a study where significant factors haven't even been identified, let alone controlled for, or else trying to do it properly, and either coming up with something inclusive, non-headline-grabbing, and basically forgettable, or else failing to produce something to publish at all. This is one problem that plagues much of the modern world.
I do think there is still a significant gender bias against women, but this kind of 'study' does little other than wave the flag for that. When the contribution of variables you haven't controlled for dwarfs the trend your analysis has picked out, and these 'not controlled for' variables are not statistically independent of your experimental variables, then for all practical purposes, whatever 'signal' you have picked up, is basically swamped by noise. Only when the 'can be explained by noise' explanation can be eliminated can you really confidently claim a result.
So I can believe there is a bias against women, but this sort of study is not going to convince me of anything.
It came on the Amstrad CPC, and you needed to type commands like run"monty just to run your games. Magazines had type-ins and 'pokes' and so on.Back then there was nowhere to hide from the need to at least know how to type 10 print "hello"... 20 goto 10... run.
Terms I use here are 'cognitive load' and 'learning fatigue'. But as an example, you don't ditch a car or PC and buy another one simply for small marginal improvements. Our capacity to learn and adapt is a limited, albeit vast resource. The modern world, however, has expanded like a gas to fill the available learning capacity, and too few care about this, myopically focussing on one 'pet trademarked feature' after another.
At warp 5 we can get there in about 3 months, and at high warp perhaps a few hours. It'll be just like how it was with transatlantic travel in the 19th and 20th centuries!
Whoops, my bad, _Ivy_ Bridge. FTFM. But still, a pre-Haswell CPU, and a GPU that benchmarks roughly similar to a GTX 660 Ti or a Radeon HD 7870. Fine for photoshop and DTP people, I imagine, for for those with serious 3d and/or video workloads...
"Tasks that previously would have required the Mac Pros of old are now being well addressed by today’s iMac."
And creative tasks that require a high-end machine, where once creative pros would turn to Mac Pros, are now being well addressed by high-end Windows workstations, that, you know, allow newer CPUs than Sandy Bridge.
Your mistake is thinking that this process hasn't already begun, and that by blocking DRM from being a W3C standard, we either stop or slow down this path. If the W3C blocks DRM, it will just end up throwing itself off an out-of-control train of corporate greed.
DRM is not the disease. Greed, selfishness and neoliberal delusions that giving human greed a free reign will magically make the world better: those are the disease. It is those that took a mechanism which, in its original form, was essential to avoiding a disastrous free-for-all in the nascent publishing industry of a few centuries ago and, via systematic strategic lobbying, twisted it into modern copyright laws. Remove DRM and the disease remains.
Not making DRM a W3C standard will turn it into another Flash/Silverlight type thing. Look at how hard it is for them to die. If DRM is going to be done, and it is going to be done, a W3C standard is better than nothing. As for where effort is expended, it must be in cultivating DRM-free content platforms, and DRM-free content.
Nearly all, if not all great scientists love their subject. Many of those who find maths or science a turn-off do not choose STEM careers. The emotional connection of a student to their discipline must not be neglected: we are humans, not programmable machines. Only if you engender a positive interest and desire in people will they be inspired to take up STEM careers, or indeed have a casual interest, whilst pursuing other careers.
Stick it on a spare machine, try your mission critical software with test data, see what's different, and what breaks. In any case, you should be asking why you're investigating a non-LTS release for serious stuff, and indeed why you're bothering with Ubuntu at all for serious stuff.
Names are less vulnerable to single-char typos. 17.04 is one char (and physically one key) away from 18.04. Also, it is easier to search for things related to your distro using google if you use the name (since google will struggle to grasp the context for each number you use).
There are many who make a tidy income out of social media, and reaching people through Twitter is part of that. I imagine that is a sensible and reasonably sized target market for one level of 'premium'. Heavy users who can spare a few dollars/pounds per month and are basically Twitter junkies would make another sensible target market. But since Twitter surely can't keep losing money every quarter, it's gotta start thinking of this stuff.
Slashdot Mirror
Exactly. I learned this with the OrangePi boards. And my 'kodi boxes' are retired in favour of Rpi2's (or pi3's, but pi3's slightly higher power needs are a annoyance so far as what USB sockets you can run them off).
Both religious aversion to science, as we see in some, and also an equally worrying trend of memorizing what is needed for exams only as long as said exams are on the horizon, are symptoms of a common anti-pattern in education.
We should not spoon feed children facts, or purported facts, or disproved 'facts'. What should be taught are the generic skills required to problem-solve, research, fact check, and basically work stuff out for yourself. Importantly, the engineering-like idea that a chain is only as strong as its weakest link, and that weakest link is only as strong as its sternest test. And teach these ideas first in the context of practical engineering and problem solving. Let religious nuts drive 'genesis as literal' ideas all they want. With the above skills well trained, the religious ideas, free from the medieval risks of burning at the stake for heresy, will just seen too silly to too many. And the class time will be better spent than merely spoonfeeding a naive and simplistic picture of how evolution actually works in practice.
Humans are all too easily addicted to the Shiny Thing, all too easily placated, and all too easily rendered docile, like a chickens neck is stroked before being broken. Humans are too content to be made cattle, and look where allowing themselves to be cattle farmed led Cowkind. #Moo
You want my shitty McJob Mr Robot, you can have my shitty McJob. Really, be my guest. Computers replaced the need for a room full of well-caned schoolboys to do sums. We just need to have less fucked up attitudes about business and ownership when it comes to robots.
If you value security, don't run the mission-critical parts of your infrastructure on a general purpose operating system like Windows, but rather run it on a minimalist, locked-down OS that has _only_ the facilities needed to do its job. The update carousel is a nightmare. If you want to ensure your Windows box doesn't sporadically reboot during a long unattended operation in order to update, what do you do? If you want to lock Windows down so it can only do the job to hand, and nothing else, you're screwed. If you run mission-critical stuff on a full-featured general purpose OS (and the same can be said for off-the-shelf Linux distros like Ubuntu and Fedora), you are kinda asking for it.
That this idea is older than me, but is ignored, is laughable.
By heavily marketing Microsoft Windows to the point that it is used, in a capacity where it can run things like Minecraft, in mission-critical IT infrastructure, they have done much to bring the current situation about. Mission critical IT infrastructure should be decomposed as a system of well-defined, hardware-isolated roles, each of which has only the authority necessary to do its job, and nothing more. (This is the principle of least authority.) There is more profit for Microsoft and major IT consultancies in just pushing Windows. Indeed Linux, in its 'desktop' flavour is no better. But Linux, being open-source, is sufficiently customisable that, as in Android or embedded uses, you can remove as much as you like.
For example, there is no need, in a patient records system, for the facility to arbitrarily create, overwrite, and delete files. If you have one machine that stores important details, another that categorises records stored by the first, and another that reads back the result, and can do nothing else (such as run Microsoft Word or Minecraft), then there is simply far less to go wrong. But systems need to be architected around this. The current trend to maximise 'bang for buck' has led to maximising flexibility and agility and, with it, maximising the flexibility and and agility offered to attackers and, thus, maximising vulnerability.
Microsoft and other proprietary software vendors, in pursuing their market positions, have done much to bring this situation about, and only when we learn that a general purpose OS is not a good idea for actually running mission-critical infrastructure (even while they are great for designing and programming them), will we start to get out of this mess of 'cyber insecurity' that we find ourselves.
When will people get it that, with a mission-critical computer system, it should have no more ability or authority to do _anything_ than it needs. If you computer is only there to do your financial stuff, the it doesn't need to be able to run Minecraft, so it should not be able to run Minecraft at all. Having a single all-things-to-all-people OS that, once booted, can do anything and everything, and is so complex that even its manufacturer can't track all the bugs and holes, and nobody else can even tell if it works, just so MS and other vendors have hid-e-holes to put their copy protection stuff in, and you can run Word, Minecraft and watch your Kitten videos on the same machine... that is just plain fing asking for it. (Linux isn't really much better _except_ that you can, and people do, produce tailored versions with extraneous stuff removed, and if you want to see the code for every last bit of software running, you can.)
Principle of least privilege (or least authority). For mission-critical stuff this is a must, and that precludes a general purpose OS like Windows or a typical Linux distro.
Suppose Facebook seek a court order in the US, say, that the Austrian court has no jurisdiction with regards to Facebook posts outside of Austria. What then?
The trouble here is that the study hasn't controlled for the _quality_ of code written, and the knock-on consequences. If, for some reason, the female part of the workforce that facebook has happens to be producing lower quality code compared to the average (and _why_ this is the case then needs to be looked into, not simply put down to gender), then naturally the knock-on effect of that is, assuming rejection of code is fair, both women rising more slowly through software engineering ranks (assuming the ranks are done solely on merit), and also women's code being rejected more. Each of these possibilities leads to more questions, and the trouble is that you have an awkward choice of either publishing a study where significant factors haven't even been identified, let alone controlled for, or else trying to do it properly, and either coming up with something inclusive, non-headline-grabbing, and basically forgettable, or else failing to produce something to publish at all. This is one problem that plagues much of the modern world.
I do think there is still a significant gender bias against women, but this kind of 'study' does little other than wave the flag for that. When the contribution of variables you haven't controlled for dwarfs the trend your analysis has picked out, and these 'not controlled for' variables are not statistically independent of your experimental variables, then for all practical purposes, whatever 'signal' you have picked up, is basically swamped by noise. Only when the 'can be explained by noise' explanation can be eliminated can you really confidently claim a result.
So I can believe there is a bias against women, but this sort of study is not going to convince me of anything.
It came on the Amstrad CPC, and you needed to type commands like run"monty just to run your games. Magazines had type-ins and 'pokes' and so on.Back then there was nowhere to hide from the need to at least know how to type 10 print "hello"... 20 goto 10... run.
It simply refuses to die.. even when it's dead.
Terms I use here are 'cognitive load' and 'learning fatigue'. But as an example, you don't ditch a car or PC and buy another one simply for small marginal improvements. Our capacity to learn and adapt is a limited, albeit vast resource. The modern world, however, has expanded like a gas to fill the available learning capacity, and too few care about this, myopically focussing on one 'pet trademarked feature' after another.
Many may not listen to Shuttleworth, but perhaps they should listen to Yoda:
“Fear is the path to the dark side. Fear leads to anger. Anger leads to hate. Hate leads to suffering.”
At warp 5 we can get there in about 3 months, and at high warp perhaps a few hours. It'll be just like how it was with transatlantic travel in the 19th and 20th centuries!
Whoops, my bad, _Ivy_ Bridge. FTFM. But still, a pre-Haswell CPU, and a GPU that benchmarks roughly similar to a GTX 660 Ti or a Radeon HD 7870. Fine for photoshop and DTP people, I imagine, for for those with serious 3d and/or video workloads...
"Tasks that previously would have required the Mac Pros of old are now being well addressed by today’s iMac."
And creative tasks that require a high-end machine, where once creative pros would turn to Mac Pros, are now being well addressed by high-end Windows workstations, that, you know, allow newer CPUs than Sandy Bridge.
Your mistake is thinking that this process hasn't already begun, and that by blocking DRM from being a W3C standard, we either stop or slow down this path. If the W3C blocks DRM, it will just end up throwing itself off an out-of-control train of corporate greed.
DRM is not the disease. Greed, selfishness and neoliberal delusions that giving human greed a free reign will magically make the world better: those are the disease. It is those that took a mechanism which, in its original form, was essential to avoiding a disastrous free-for-all in the nascent publishing industry of a few centuries ago and, via systematic strategic lobbying, twisted it into modern copyright laws. Remove DRM and the disease remains.
Not making DRM a W3C standard will turn it into another Flash/Silverlight type thing. Look at how hard it is for them to die. If DRM is going to be done, and it is going to be done, a W3C standard is better than nothing. As for where effort is expended, it must be in cultivating DRM-free content platforms, and DRM-free content.
Nearly all, if not all great scientists love their subject. Many of those who find maths or science a turn-off do not choose STEM careers. The emotional connection of a student to their discipline must not be neglected: we are humans, not programmable machines. Only if you engender a positive interest and desire in people will they be inspired to take up STEM careers, or indeed have a casual interest, whilst pursuing other careers.
There is _no_ idiot proof way of manufacturing clever thinking. Who'd have thought that?
And once it has moved to Mir, all those other Linux distros won't even be properly compatible with the industry standard Ubuntu setup. [ ducks ]
Stick it on a spare machine, try your mission critical software with test data, see what's different, and what breaks. In any case, you should be asking why you're investigating a non-LTS release for serious stuff, and indeed why you're bothering with Ubuntu at all for serious stuff.
Names are less vulnerable to single-char typos. 17.04 is one char (and physically one key) away from 18.04. Also, it is easier to search for things related to your distro using google if you use the name (since google will struggle to grasp the context for each number you use).
There are many who make a tidy income out of social media, and reaching people through Twitter is part of that. I imagine that is a sensible and reasonably sized target market for one level of 'premium'. Heavy users who can spare a few dollars/pounds per month and are basically Twitter junkies would make another sensible target market. But since Twitter surely can't keep losing money every quarter, it's gotta start thinking of this stuff.