Because it's not a matter of hacking together a patch, running the unit tests, uploading to production and waiting to see if it crashes.
This stuff has to run the gauntlet of companies, regulators, and customers who have NFI about infosec, but do have some idea of the consequences of rushing untested changes into devices which quite literally keep people alive from minute to minute.
To indulge in some gross stereotyping here, they have huge egos that exceed their (very considerable) talents, and little appreciation that anything that doesn't involve medicine, or indeed surgery, is important.
They also tend to end up running hospitals.
If you tell a surgeon running a hospital that you need to inconvenience him (and it's usually a him) and his fellow surgeons to solve a "problem with the computers", they will ignore you. They are also right - anything that interferes with their ability to do surgery is a huge waste of resources.
An infosec person implementing the "principle of least privilege" is almost certainly going to grossly inconvenience surgeons in the process, to ends that are not at all obvious to most of them. Along the way they will, at the very least, inconvenience patients. Therefore, the infosec person will get told precisely where to stick their principle of least privilege.
Notably, there were no graphics and sound primitives whatsoever in C64 BASIC. If you wanted to take advantage of the (actually quite impressive, for the day) graphics and sound, you had to directly manipulate memory.
The insides of a bike frame are extremely simple - they're just tubes - and the mechanical components in the frame (the "bottom bracket") comes in only a few standard designs. Any plausible motor and battery is going to be big enough to stick out like a sore thumb. So X-Raying would work, as would pointing an IR camera at the bike detect the motor in operation. You can't hide that much waste heat in that small an area.
As for stock bikes, nice idea, but not practical. At the elite level (and even at the serious recreational level) riders often spend a lot of time and money customizing the fit of their bikes. Furthermore, much of the sport's funding comes from equipment manufacturers who would be more than a little peeved if athletes weren't using their expensive gear.
Longer version: 100 watts for 10 minutes in the context of an hour-long cyclocross race is enough to turn an also-ran into a winner. It would be decisive in most road races other than out-and-out bunch sprints as well.
As far as drag goes, that's negligible by all reports. Avoiding drag when a power source is not providing propulsion is a very well-studied problem.
There are things that are net wins for society that can't be built effectively by the private sector on a user-pays basis. Roads and sewers are classic examples. National power grids are probably another.
You're assuming that movie studios invest hundreds of millions of dollars because they're collectively dumb.
They aren't.
Studios fund sequels, franchises, and certain actors and directors because they more reliably attract audiences than films that don't have these properties.
From what I've read, a significant fraction of kiddie porn is produced on a commercial basis. Some Russian spammers back in the golden age of pharma spam, for instance, had other businesses in kiddie porn production.
So your premise is, as far as I can tell, wrong - kiddie porn viewers directly fund kiddie porn production.
With the possible exception of people who go on to be academic mathematicians (or closely related disciplines) I've not met anyone who finds declarative programming an easier introduction to programming than imperative programming. My alma mater tried this with functional languages in the 1990s and gave it away.
If you look at the languages that non-programmers actually take to, they tend to be dynamically typed imperative languages with a minimum of boilerplate and very forgiving syntax. BASIC, Perl, JavaScript. Terrible languages for actually doing real work, but newbies like them.
Yes, we're collecting a bunch of data on student learning, and you can do data mining on it.
Doesn't mean that academics have any time or incentive to do anything based on it.
Academics are not rated on their actual teaching performance, they are rated on a) grant money brought in, c) research money brought in... y) pass rates, z) student satisfaction surveys. Note the complete absence of whether students actually learn anything as an evaluation criteria.
But, then, the universities that employ them aren't really rated on teaching quality either. While there is good teaching happening in the Ivy League and other top-rated institutions, there's also a hell of a lot of coasting on the smarts and work ethic of the students they select.
OK, it turns out you can get GPIO pins for an extra 4 quid in the UK. Then you need either a USB Ethernet interface or the WiFi dongle. So we're up to 16 quid in the UK, which will probably end up being 30 bucks in Australia. The cost difference with full Pi or BBB isn't as dramatic as it was.
Connecting via SSH over a conventional network isn't as easy as it sounds. For one, university WiFi networks are flaky, and there are no spare Ethernet ports that students can plug things into in our labs. And configuring the WiFi will require students to have access to an SD card reader on their PC (which isn't standard on lab PCs, so we'll need a bunch of USB ones). Then students will have to find out what the IP address of *their* pi is (the dhcp servers are run by University central IT and we have no access to it) without the ability to look at a screen connected to the pi...
So, for my application, being able to plug a device into the USB port and SSH to it is a big advantage.
I'm looking into teaching introductory embedded programming to CS undergrads. The thought of a $5 computer that they could simply all buy at the start of semester was very appealing.
However, I'm leaning towards using the Beaglebone Black rather than the Pi Zero or even the standard Pi, on the following grounds:
If you want to hook anything to the Pi Zero's GPIO pins, you need to break out the soldering iron. In a university context with OH&S laws, that means getting access to the electrical engineering labs - and, funnily enough, they tend to have just enough labs to cater for electrical engineering classes and little surplus.
To write your first program on the BBB, you get a USB cable, plug one end into the BBB, the other end into a PC, and fire up a browser. Add putty and you've got full shell access. By contrast, with the Pi you need to plug monitors and keyboards into it, or figure out how to put it on the network in an accessible location. All of this requires a pile of cooperation from university IT, which is not dissimilar to the accounting department in Dilbert cartoons.
All in all, the BBB looks to have much lower barriers to entry, despite the higher cost of the units - and I don't think my usage scenario is that unique.
I've just been investigating this very question.
I'm develop a tertiary course in "software engineering process". Small teams will need to work together to build *something*. For a variety of reasons, we think that building an embedded system would be a good thing for them to have exposure to, so I'm trying to find a suitable platform to develop on.
The current Raspberry Pi and Beaglebone Black both have their strengths and weaknesses, but both would do for the job. But they both cost at least 50 AUD, which is affordable but not equivalent to zero for planning purposes.
By contrast, both the C.H.I.P. and the Pi zero are so cheap that the cost can be ignored.
However, both platforms require you to break out the soldering iron if you want to attach things to the GPIO ports. By the time you have something you can hook things up to without soldering, you're back up to the cost of a standard Pi anyway.
...it ain't going to achieve much.
China has an official English-language TV channel, which screens in at least some parts of China, is carried on satellites and is streamed online. It is unwatchably dull.
...in which distinguished elderly physicist becomes convinced that the entirety of a "lesser" field is wrong and its practitioners fools. See Penrose, Roger.
Do you want a security system or a hobby? Find a professional and let them worry about whether it's been designed and more importantly installed properly.
That said, before you turn your home into a fortress might I suggest you buy a copy of Ground Control a book about British urban planning which has a lot to say about the downsides of "secure" homes.
The Surface Pro 3 is 800 USD and a 12 inch form factor. A Nexus 7 was~200 USD and is small enough to hold in one hand.
All jokes aside, there are plenty of tasks out there for which the Surface Pro 3 or other full-size tablets are oo big. We have built a web app for in-class marking in tertiary software engineering lab classes, and a 7 or 8 inch tablet is the perfect device for carrying around with you and entering marks. A 5 inch phone is too fiddly to press on radiobuttons and can't fit readable descriptions for more than a couple of marking criteria on the page; a full-size tablet requires you to put it down on a desk to use and gets heavy to carry round - and the Pro 3, I suspect, would be particularly bad.
I might get a Pro 3 or its successor one day, but it will be an adjunct rather than a replacement for my own Nexus 7.
Slashdot Mirror
OK, then, what about this study? Good enough for you?
This stuff has to run the gauntlet of companies, regulators, and customers who have NFI about infosec, but do have some idea of the consequences of rushing untested changes into devices which quite literally keep people alive from minute to minute.
To indulge in some gross stereotyping here, they have huge egos that exceed their (very considerable) talents, and little appreciation that anything that doesn't involve medicine, or indeed surgery, is important.
They also tend to end up running hospitals.
If you tell a surgeon running a hospital that you need to inconvenience him (and it's usually a him) and his fellow surgeons to solve a "problem with the computers", they will ignore you. They are also right - anything that interferes with their ability to do surgery is a huge waste of resources.
An infosec person implementing the "principle of least privilege" is almost certainly going to grossly inconvenience surgeons in the process, to ends that are not at all obvious to most of them. Along the way they will, at the very least, inconvenience patients. Therefore, the infosec person will get told precisely where to stick their principle of least privilege.
Notably, there were no graphics and sound primitives whatsoever in C64 BASIC. If you wanted to take advantage of the (actually quite impressive, for the day) graphics and sound, you had to directly manipulate memory.
I propose an increase in taxation to support a Libertarian Emigration Fund.
The insides of a bike frame are extremely simple - they're just tubes - and the mechanical components in the frame (the "bottom bracket") comes in only a few standard designs. Any plausible motor and battery is going to be big enough to stick out like a sore thumb. So X-Raying would work, as would pointing an IR camera at the bike detect the motor in operation. You can't hide that much waste heat in that small an area. As for stock bikes, nice idea, but not practical. At the elite level (and even at the serious recreational level) riders often spend a lot of time and money customizing the fit of their bikes. Furthermore, much of the sport's funding comes from equipment manufacturers who would be more than a little peeved if athletes weren't using their expensive gear.
Longer version: 100 watts for 10 minutes in the context of an hour-long cyclocross race is enough to turn an also-ran into a winner. It would be decisive in most road races other than out-and-out bunch sprints as well.
As far as drag goes, that's negligible by all reports. Avoiding drag when a power source is not providing propulsion is a very well-studied problem.
There are things that are net wins for society that can't be built effectively by the private sector on a user-pays basis. Roads and sewers are classic examples. National power grids are probably another.
You're assuming that movie studios invest hundreds of millions of dollars because they're collectively dumb. They aren't. Studios fund sequels, franchises, and certain actors and directors because they more reliably attract audiences than films that don't have these properties.
From what I've read, a significant fraction of kiddie porn is produced on a commercial basis. Some Russian spammers back in the golden age of pharma spam, for instance, had other businesses in kiddie porn production. So your premise is, as far as I can tell, wrong - kiddie porn viewers directly fund kiddie porn production.
If you look at the languages that non-programmers actually take to, they tend to be dynamically typed imperative languages with a minimum of boilerplate and very forgiving syntax. BASIC, Perl, JavaScript. Terrible languages for actually doing real work, but newbies like them.
Doesn't mean that academics have any time or incentive to do anything based on it.
Academics are not rated on their actual teaching performance, they are rated on a) grant money brought in, c) research money brought in... y) pass rates, z) student satisfaction surveys. Note the complete absence of whether students actually learn anything as an evaluation criteria.
But, then, the universities that employ them aren't really rated on teaching quality either. While there is good teaching happening in the Ivy League and other top-rated institutions, there's also a hell of a lot of coasting on the smarts and work ethic of the students they select.
Lead. Mercury. Asbestos.
Connecting via SSH over a conventional network isn't as easy as it sounds. For one, university WiFi networks are flaky, and there are no spare Ethernet ports that students can plug things into in our labs. And configuring the WiFi will require students to have access to an SD card reader on their PC (which isn't standard on lab PCs, so we'll need a bunch of USB ones). Then students will have to find out what the IP address of *their* pi is (the dhcp servers are run by University central IT and we have no access to it) without the ability to look at a screen connected to the pi...
So, for my application, being able to plug a device into the USB port and SSH to it is a big advantage.
However, I'm leaning towards using the Beaglebone Black rather than the Pi Zero or even the standard Pi, on the following grounds:
All in all, the BBB looks to have much lower barriers to entry, despite the higher cost of the units - and I don't think my usage scenario is that unique.
You will note an oversupply of:
The most interesting thing from this video is that Amazon is playing with tiltrotor concepts for their delivery drone design.
I've just been investigating this very question. I'm develop a tertiary course in "software engineering process". Small teams will need to work together to build *something*. For a variety of reasons, we think that building an embedded system would be a good thing for them to have exposure to, so I'm trying to find a suitable platform to develop on. The current Raspberry Pi and Beaglebone Black both have their strengths and weaknesses, but both would do for the job. But they both cost at least 50 AUD, which is affordable but not equivalent to zero for planning purposes. By contrast, both the C.H.I.P. and the Pi zero are so cheap that the cost can be ignored. However, both platforms require you to break out the soldering iron if you want to attach things to the GPIO ports. By the time you have something you can hook things up to without soldering, you're back up to the cost of a standard Pi anyway.
Even if you take the view that Blair deserves what's coming to him, these dicks are going after his parents and siblings. That's reprehensible.
...it ain't going to achieve much. China has an official English-language TV channel, which screens in at least some parts of China, is carried on satellites and is streamed online. It is unwatchably dull.
...in which distinguished elderly physicist becomes convinced that the entirety of a "lesser" field is wrong and its practitioners fools. See Penrose, Roger.
Well, they are "embedded systems" in that they are computers embedded in a mechanical system, but:
As such, throwing hardware at the problem to make the programming easier seems like an entirely rational approach.
Do you want a security system or a hobby? Find a professional and let them worry about whether it's been designed and more importantly installed properly.
That said, before you turn your home into a fortress might I suggest you buy a copy of Ground Control a book about British urban planning which has a lot to say about the downsides of "secure" homes.
Every Australian politician and science bureaucrat knows that Australian radio astronomers invented Wi-Fi...
</sarcasm>
All jokes aside, there are plenty of tasks out there for which the Surface Pro 3 or other full-size tablets are oo big. We have built a web app for in-class marking in tertiary software engineering lab classes, and a 7 or 8 inch tablet is the perfect device for carrying around with you and entering marks. A 5 inch phone is too fiddly to press on radiobuttons and can't fit readable descriptions for more than a couple of marking criteria on the page; a full-size tablet requires you to put it down on a desk to use and gets heavy to carry round - and the Pro 3, I suspect, would be particularly bad.
I might get a Pro 3 or its successor one day, but it will be an adjunct rather than a replacement for my own Nexus 7.