In my area (smallish city), when I call the inspector you have to call before 10am or you get a voicemail that never gets returned. It's at least 3 days before they can schedule a walk-through for emergencies (eg. your power is off in the middle of winter), for regular visits it's usually up to 5-7 business days.
At least they let owners off the hook for revisits though. I had to go through it for both an emergency (breaker box burned out) and a renovation (additional living space in attic), it's not fun dealing with the inspector's office.
Same in the US although home owners are typically exempt from a lot of things, the paperwork alone would cost 3 out of the 9 months the article says it took with plenty of weeks in between construction where you have to wait for an inspector to come before proceeding to the next (unless you're a professional and know your schedule in advance, you have to schedule them when you actually finished a portion or risk having to pay for a second visit).
And that's if you do everything correct the first time around, inspectors will make you fix all the edge cases and you have to know a lot of code to actually get a plan approved in the first place. And there are a few things you may still not be able to do yourself, in most jurisdictions that means you cannot connect your plumbing to a main city plumbing line or sewage to a sewage line without a professional. You can also not connect your house electric to a pole without a lineman (because those things are live, no such thing as breakers on the pole) and gas/electric will not turn on until you pass an official inspection.
Up to code in the US these days for that sort of construction means things like fire sprinklers, wired smoke and CO alarm system, proper electric wiring with outlets every 10ft along the wall, sufficient insulation, sufficient space between junction boxes and insulation etc etc
Both can be helpful in certain amounts. Everything is poisonous to some level, even water, but you don't see the FDA banning water.
The problem with the tablets is that they are placebo's (which is acceptable levels of poison to the FDA) but some of them contained thousands of nanograms of deadly nightshade substance. Hyland repeatedly couldn't fix the issue of not having consistent levels of poison in their placebo's.
Botox injections contain some poisonous substance but that amount is very strictly controlled and the same in every syringe. If the syringes contained lethal amounts of Botox in 0.1% of the cases, but acceptable amounts of it in 99.9% of the rest, the FDA would also require a recall.
No, the FDA didn't fuck up, they had the same problems as before - inconsistent amounts of the chemicals in the medicine. This could be due to bad QA or whatever, but that's Hyland's fault.
If you take a sample, most tablets contain what's considered "safe" - no active substance in the tablets aka sugar-pills or placebo. So Hyland is selling a placebo, however once in a while one of the tablets came back with insane levels (potentially psychoactive to adults) of the deadly nightshade chemicals. That's why it's so hard to 'catch' Hyland because most of the tablets and bottles are indeed safe, it's just once every so often, you got actual poison among your placebos.
According to the report the product had very inconsistent amounts of the deadly nightshade. Most tablets had amounts that couldn't even be measured which is the safe standard, others had over a thousand nanograms of the belladonna atropine (one of it's alkaloids) and several hundreds of nanograms of the scopolamine.
Given that for adults the "prescribed" levels are like 0.05g of the leaves (which only contain ~1% of the atropine by weight) for a psychoactive effect, you could consider that these are potentially lethal to infants.
I'm sure it is easier for people that don't know what the hell they are doing. If you just want to clicky-create-account, again, there are plenty of providers that will charge you a tenth of that number you quoted.
For 50 users, you can probably set up a cheap VPS with Postfix/Dovecot somewhere and never look at it again. If you're looking for complete groupware SaaS, there are similarly various options these days that not only will duplicate all the Google/Microsoft features but run it a lot better.
Millenial snowflakes running for congress... I'm getting old, only a few years ago I was not even eligible to run, now it seems I'm too old to even start.
Perhaps you should review the constitution before you spout such non-sense. The federal court has no constitutional jurisdiction over any President's orders about how to enforce existing legislation at the border..
These laws came and were under effect under Obama, you just never bothered yelling about it until Trump signed an order to uphold these laws. But DHS checking Facebook, Twitter, computers and phones has been repeatedly reported for at least half a decade here on Slashdot.
Python IS single-threaded, multi-threading is a big hack on Python and usually (at least a few years ago) came with big warning letters not to use unless you absolutely knew what you were doing.
Time slicing is not multi-threading, can it allocate a process to a particular CPU or guarantee it's still running on the same CPU when I call back? I think that's the same issue Rust has.
Especially when it comes down to commercial panels, you need to clean them, clear snow in many places, clear bird poop (which can quickly erode rubber gaskets, wiring and plastics). As I said, it doesn't matter much to lose 1% in generation per year on a residential installation, on a commercial installation there are many more 'moving parts' that have to be kept in order.
They're expensive to produce, more expensive than their actual cost. The main reason we don't see that cost is because everything along the way is heavily subsidized and we have exported the pollution (companies are simply dumping toxic waste from solar panel production in nearby rivers). They're also substantially more expensive when you consider everything than other solar power generation technologies.
They don't recycle well, the stats you have are from PV promoters. In reality, Europe had to exempt them from the toxic waste list even though they are on the list of toxic waste. Many in the industry know about the looming problem of lead, cadmium and other heavy metals used in the systems. Sure you can 'technically' recycle most if not all heavy metals, yet the first set of commercial PV installations are being replaced already and we're not sure what to do with them, they are made of the same materials and substrates most computer/technological waste is made off (they are LED's after all), yet we don't see mass recycling of portable phones or computers which have been around in substantial quantities for over 50 years now because it's (still) not economical nor ecologically effective to do so.
Don't last nearly as long as advertised because a) they degrade much faster than advertised (some degrade at 3-4% per year) and b) new technology comes along necessitating replacement just to keep up with energy usage per area installed and also to lower maintenance costs. Again, nobody cares when this happens to a home which make out a very minimal contribution to the overall power system, these are commercial installations, in those cases we're talking about 10-15 years before they are replaced. The problem here is similar to SSD's, the first generation SLC SSD's lasted very long compared to these days SSD's last a lot less long. So in PV, the first generation was sturdy mono-crystalline silicon, these days very thin Cadmium-Telluride panels are very lightweight, very cheap to produce but they degrade 6 times as fast as first generation silicon.
I'm not advocating against PV for solar home installations, I just think they're a big waste of time and resources for actual solar plants which would, in the long run, be much better off with solar thermal if governments all across the world didn't just blindly subsidize "green" technology.
I know people have been asked to volunteer Facebook and other social network information upon entry into the US, this has been the case for at least 2 if not 4 years and started under the Obama administration. Even Twitter held back their outrage about the DHS using their information for border controls until AFTER it was clear that Trump had won but even that story is now several months old.
All these 'new things' including the stopping of people at the border from certain nations has been done and legislated under the Obama administration, to quote another user here on Slashdot when Obama seized public lands from use by Indians and other members of the public: "Obama is just implementing what was legislated before, these orders are just telling particular agencies how they should implement the legislation".
Well, at least now you know why such laws should never be allowed, most of you here on Slashdot didn't care or realize the gravity of the issues when Obama, Bush or Clinton signed off on these exact laws. Next up: how Trump uses the DMCA, Patriot Act, NAFTA, TPP etc. to do whatever the fuck he wants - and you (the American public) gave it to them because you either didn't feel safe, didn't care or didn't believe that someone could rise up that would abuse those legislations. Too late now.
The problem is that Windows 10 overrides those settings lately. I have a singular machine that runs Windows 10 in our network which is set to automatically update outside of business hours, enforced by group policies, no user is an admin, users are set not to be notified and per Microsoft, we also needed the latest WSUS incarnation to schedule updates so yes, we are running a Windows Server to make sure the only Windows machine on our network is not auto-updating.
Just last week: Pop-up over a full-screen window: "Updates are ready to install now, your machine will reboot in 60.. 59.. 58.."
It's your fault to run critical applications on Windows, yes. Windows is for gaming and running viruses, that's all. If you can use Windows to do your job, your job is trivial and will be automated.
Sun Servers did as well, they were one of the first machines besides mainframes that even had hot-swappable CPU and RAM, Solaris kernels could be upgraded without a reboot.
You could inflate the numbers a bit so you get a better offer. Another reason would be to justify the H1B's they're planning on hiring. But if you work for a company that only wants to give you the lowest number possible, do you really want to work for them? There are plenty of job openings for skilled people and plenty of numbnuts that are desperate enough that you don't have to grovel for a company like that.
With the license attached to it, the software maker has given no guarantee that this will work for any purpose. Most open source licenses don't guarantee the product to work. This is where proper engineering comes into play (including mechanical etc), if there is someone (a company perhaps) that wants to commercialize it, then they'll be on the hook for any faults, problems, testing, licensing.
Obviously at some point, failures will happen just like current cars fail, who is on the hook depends on whose fault it is that the failure happened. Was it lack of maintenance or a manufacturing defect and if it was a manufacturing defect, was it negligible of them to create or ignore the problem or was it just because 'shit happens'. If it was a cost saving measure or problem the company internally was aware of but didn't fix because it cost too much, they could be held liable (civilly and criminally), if it was because because things sometimes go wrong and they couldn't have foreseen it, that's why we have car insurance in the first place.
Autonomous cars don't change too much to the liability question, they do shift some of the liability to the manufacturer who will calculate that cost and include it into the price, some of it to the owner who should do the same (you can be held accountable (negligence) if you didn't bring your car to the shop in a reasonable time after you got the recall notice) and then your insurance will carry some of the cost. In the end if the cars reduce accidents or reduce damage by even a few percentage points costs for everyone will go down, the insurance will have less to pay out so they have to charge the owner less, the car company will be less impacted by liability claims (eg. if a serious mechanical issue only occurs in event of a car crash (such as Tesla's batteries or VW's airbag deployment issues), having less car crashes reduces the liability exposure). A deer/child will still end up jumping in front of a car once in a while, if the car does what it's supposed to do or better in those instances (even now, your car has to have a certain amount of braking power based on it's weight and speeds), then why does it matter to the software developer when someone dies?
Not necessarily the person that wanted it, the person that ended up installing it in contravention of rules and safety. I meant the installer that mechanically connects the drive shaft and the brakes to this machine. The maintainer of the software has nothing to do with it in this instance he's just putting his science project on the net, unless he sells it (and/or gives it away) with some sort of guarantee that it will work.
Once you have a company and you sell it to do some sort of function and it ends up being a scam or without the proper engineering, inspection, tests etc then that company ends up with the liability. It could obviously be the owner of the car if they are a DIYer but just because you change your own brakes, doesn't mean you can't be liable for the workmanship if they don't work, at this point no serious mechanic would install this in a car though for everyday use.
I don't know much about Python since I don't use it. I know it's single threaded because I tried multithreading some existing code once and gave up.
A well-designed single threaded program should be predictable after compilation, what the hell is it doing that makes it non-deterministic, doing a "sleep rand(1...10)" after every line?
I meant a processor that you can predict it's state after every line, for most small microcontrollers you can easily (even manually) trace every code interaction and predict how much time it takes before you get an answer. Modern x86 practically have their own OS that not only translates the instruction set but also 'does things' like cleanup and optimization which makes it very hard, if not impossible, to predict what will happen and when it will return the next instruction you give.
Not quite the same as 24 year uptime. In the same vein, I have a Sun server that is still running since the mid-90's, part of a medical device and used to compile very particular software code for an old small-bore MRI system. We shut it down when the power goes out (very rare), but it's SCSI drives are still good.
The installer/maintainer (whoever you paid to 'make it work'), the source code license doesn't matter really in these kinds of things. Just because I make the code to a pacemaker open source and it fails, doesn't mean that I'm not liable. That would be a very easy way to avoid liability though.
Python is single-threaded and thus it's speed and results should be predictable. Not sure whether Python would still run on a decent real-time processor though (x86 has not been in a long time) because it's overhead would be enormous.
Slashdot Mirror
Well, in my area the fuse turns off the entire block, the poles are strung together through people's backyards.
In my area (smallish city), when I call the inspector you have to call before 10am or you get a voicemail that never gets returned. It's at least 3 days before they can schedule a walk-through for emergencies (eg. your power is off in the middle of winter), for regular visits it's usually up to 5-7 business days.
At least they let owners off the hook for revisits though. I had to go through it for both an emergency (breaker box burned out) and a renovation (additional living space in attic), it's not fun dealing with the inspector's office.
Same in the US although home owners are typically exempt from a lot of things, the paperwork alone would cost 3 out of the 9 months the article says it took with plenty of weeks in between construction where you have to wait for an inspector to come before proceeding to the next (unless you're a professional and know your schedule in advance, you have to schedule them when you actually finished a portion or risk having to pay for a second visit).
And that's if you do everything correct the first time around, inspectors will make you fix all the edge cases and you have to know a lot of code to actually get a plan approved in the first place. And there are a few things you may still not be able to do yourself, in most jurisdictions that means you cannot connect your plumbing to a main city plumbing line or sewage to a sewage line without a professional. You can also not connect your house electric to a pole without a lineman (because those things are live, no such thing as breakers on the pole) and gas/electric will not turn on until you pass an official inspection.
Up to code in the US these days for that sort of construction means things like fire sprinklers, wired smoke and CO alarm system, proper electric wiring with outlets every 10ft along the wall, sufficient insulation, sufficient space between junction boxes and insulation etc etc
Both can be helpful in certain amounts. Everything is poisonous to some level, even water, but you don't see the FDA banning water.
The problem with the tablets is that they are placebo's (which is acceptable levels of poison to the FDA) but some of them contained thousands of nanograms of deadly nightshade substance. Hyland repeatedly couldn't fix the issue of not having consistent levels of poison in their placebo's.
Botox injections contain some poisonous substance but that amount is very strictly controlled and the same in every syringe. If the syringes contained lethal amounts of Botox in 0.1% of the cases, but acceptable amounts of it in 99.9% of the rest, the FDA would also require a recall.
No, the FDA didn't fuck up, they had the same problems as before - inconsistent amounts of the chemicals in the medicine. This could be due to bad QA or whatever, but that's Hyland's fault.
If you take a sample, most tablets contain what's considered "safe" - no active substance in the tablets aka sugar-pills or placebo. So Hyland is selling a placebo, however once in a while one of the tablets came back with insane levels (potentially psychoactive to adults) of the deadly nightshade chemicals. That's why it's so hard to 'catch' Hyland because most of the tablets and bottles are indeed safe, it's just once every so often, you got actual poison among your placebos.
According to the report the product had very inconsistent amounts of the deadly nightshade. Most tablets had amounts that couldn't even be measured which is the safe standard, others had over a thousand nanograms of the belladonna atropine (one of it's alkaloids) and several hundreds of nanograms of the scopolamine.
Given that for adults the "prescribed" levels are like 0.05g of the leaves (which only contain ~1% of the atropine by weight) for a psychoactive effect, you could consider that these are potentially lethal to infants.
I'm sure it is easier for people that don't know what the hell they are doing. If you just want to clicky-create-account, again, there are plenty of providers that will charge you a tenth of that number you quoted.
It's Microsoft Office, you pay for it 356 days of the year, you use it perhaps 50 and half of the time it's down or extremely slow.
For 50 users, you can probably set up a cheap VPS with Postfix/Dovecot somewhere and never look at it again. If you're looking for complete groupware SaaS, there are similarly various options these days that not only will duplicate all the Google/Microsoft features but run it a lot better.
No you can't, Google Drive syncs links to Google Docs to your hard drive, not the actual documents.
Millenial snowflakes running for congress... I'm getting old, only a few years ago I was not even eligible to run, now it seems I'm too old to even start.
Perhaps you should review the constitution before you spout such non-sense. The federal court has no constitutional jurisdiction over any President's orders about how to enforce existing legislation at the border..
These laws came and were under effect under Obama, you just never bothered yelling about it until Trump signed an order to uphold these laws. But DHS checking Facebook, Twitter, computers and phones has been repeatedly reported for at least half a decade here on Slashdot.
Python IS single-threaded, multi-threading is a big hack on Python and usually (at least a few years ago) came with big warning letters not to use unless you absolutely knew what you were doing.
Time slicing is not multi-threading, can it allocate a process to a particular CPU or guarantee it's still running on the same CPU when I call back? I think that's the same issue Rust has.
Especially when it comes down to commercial panels, you need to clean them, clear snow in many places, clear bird poop (which can quickly erode rubber gaskets, wiring and plastics). As I said, it doesn't matter much to lose 1% in generation per year on a residential installation, on a commercial installation there are many more 'moving parts' that have to be kept in order.
They're expensive to produce, more expensive than their actual cost. The main reason we don't see that cost is because everything along the way is heavily subsidized and we have exported the pollution (companies are simply dumping toxic waste from solar panel production in nearby rivers). They're also substantially more expensive when you consider everything than other solar power generation technologies.
They don't recycle well, the stats you have are from PV promoters. In reality, Europe had to exempt them from the toxic waste list even though they are on the list of toxic waste. Many in the industry know about the looming problem of lead, cadmium and other heavy metals used in the systems. Sure you can 'technically' recycle most if not all heavy metals, yet the first set of commercial PV installations are being replaced already and we're not sure what to do with them, they are made of the same materials and substrates most computer/technological waste is made off (they are LED's after all), yet we don't see mass recycling of portable phones or computers which have been around in substantial quantities for over 50 years now because it's (still) not economical nor ecologically effective to do so.
Don't last nearly as long as advertised because a) they degrade much faster than advertised (some degrade at 3-4% per year) and b) new technology comes along necessitating replacement just to keep up with energy usage per area installed and also to lower maintenance costs. Again, nobody cares when this happens to a home which make out a very minimal contribution to the overall power system, these are commercial installations, in those cases we're talking about 10-15 years before they are replaced. The problem here is similar to SSD's, the first generation SLC SSD's lasted very long compared to these days SSD's last a lot less long. So in PV, the first generation was sturdy mono-crystalline silicon, these days very thin Cadmium-Telluride panels are very lightweight, very cheap to produce but they degrade 6 times as fast as first generation silicon.
I'm not advocating against PV for solar home installations, I just think they're a big waste of time and resources for actual solar plants which would, in the long run, be much better off with solar thermal if governments all across the world didn't just blindly subsidize "green" technology.
I know people have been asked to volunteer Facebook and other social network information upon entry into the US, this has been the case for at least 2 if not 4 years and started under the Obama administration. Even Twitter held back their outrage about the DHS using their information for border controls until AFTER it was clear that Trump had won but even that story is now several months old.
All these 'new things' including the stopping of people at the border from certain nations has been done and legislated under the Obama administration, to quote another user here on Slashdot when Obama seized public lands from use by Indians and other members of the public: "Obama is just implementing what was legislated before, these orders are just telling particular agencies how they should implement the legislation".
Well, at least now you know why such laws should never be allowed, most of you here on Slashdot didn't care or realize the gravity of the issues when Obama, Bush or Clinton signed off on these exact laws. Next up: how Trump uses the DMCA, Patriot Act, NAFTA, TPP etc. to do whatever the fuck he wants - and you (the American public) gave it to them because you either didn't feel safe, didn't care or didn't believe that someone could rise up that would abuse those legislations. Too late now.
The problem is that Windows 10 overrides those settings lately. I have a singular machine that runs Windows 10 in our network which is set to automatically update outside of business hours, enforced by group policies, no user is an admin, users are set not to be notified and per Microsoft, we also needed the latest WSUS incarnation to schedule updates so yes, we are running a Windows Server to make sure the only Windows machine on our network is not auto-updating.
Just last week: Pop-up over a full-screen window: "Updates are ready to install now, your machine will reboot in 60.. 59.. 58.."
It's your fault to run critical applications on Windows, yes. Windows is for gaming and running viruses, that's all. If you can use Windows to do your job, your job is trivial and will be automated.
Sun Servers did as well, they were one of the first machines besides mainframes that even had hot-swappable CPU and RAM, Solaris kernels could be upgraded without a reboot.
You could inflate the numbers a bit so you get a better offer. Another reason would be to justify the H1B's they're planning on hiring. But if you work for a company that only wants to give you the lowest number possible, do you really want to work for them? There are plenty of job openings for skilled people and plenty of numbnuts that are desperate enough that you don't have to grovel for a company like that.
With the license attached to it, the software maker has given no guarantee that this will work for any purpose. Most open source licenses don't guarantee the product to work. This is where proper engineering comes into play (including mechanical etc), if there is someone (a company perhaps) that wants to commercialize it, then they'll be on the hook for any faults, problems, testing, licensing.
Obviously at some point, failures will happen just like current cars fail, who is on the hook depends on whose fault it is that the failure happened. Was it lack of maintenance or a manufacturing defect and if it was a manufacturing defect, was it negligible of them to create or ignore the problem or was it just because 'shit happens'. If it was a cost saving measure or problem the company internally was aware of but didn't fix because it cost too much, they could be held liable (civilly and criminally), if it was because because things sometimes go wrong and they couldn't have foreseen it, that's why we have car insurance in the first place.
Autonomous cars don't change too much to the liability question, they do shift some of the liability to the manufacturer who will calculate that cost and include it into the price, some of it to the owner who should do the same (you can be held accountable (negligence) if you didn't bring your car to the shop in a reasonable time after you got the recall notice) and then your insurance will carry some of the cost. In the end if the cars reduce accidents or reduce damage by even a few percentage points costs for everyone will go down, the insurance will have less to pay out so they have to charge the owner less, the car company will be less impacted by liability claims (eg. if a serious mechanical issue only occurs in event of a car crash (such as Tesla's batteries or VW's airbag deployment issues), having less car crashes reduces the liability exposure). A deer/child will still end up jumping in front of a car once in a while, if the car does what it's supposed to do or better in those instances (even now, your car has to have a certain amount of braking power based on it's weight and speeds), then why does it matter to the software developer when someone dies?
Not necessarily the person that wanted it, the person that ended up installing it in contravention of rules and safety. I meant the installer that mechanically connects the drive shaft and the brakes to this machine. The maintainer of the software has nothing to do with it in this instance he's just putting his science project on the net, unless he sells it (and/or gives it away) with some sort of guarantee that it will work.
Once you have a company and you sell it to do some sort of function and it ends up being a scam or without the proper engineering, inspection, tests etc then that company ends up with the liability. It could obviously be the owner of the car if they are a DIYer but just because you change your own brakes, doesn't mean you can't be liable for the workmanship if they don't work, at this point no serious mechanic would install this in a car though for everyday use.
I don't know much about Python since I don't use it. I know it's single threaded because I tried multithreading some existing code once and gave up.
A well-designed single threaded program should be predictable after compilation, what the hell is it doing that makes it non-deterministic, doing a "sleep rand(1...10)" after every line?
I meant a processor that you can predict it's state after every line, for most small microcontrollers you can easily (even manually) trace every code interaction and predict how much time it takes before you get an answer. Modern x86 practically have their own OS that not only translates the instruction set but also 'does things' like cleanup and optimization which makes it very hard, if not impossible, to predict what will happen and when it will return the next instruction you give.
Not quite the same as 24 year uptime. In the same vein, I have a Sun server that is still running since the mid-90's, part of a medical device and used to compile very particular software code for an old small-bore MRI system. We shut it down when the power goes out (very rare), but it's SCSI drives are still good.
The installer/maintainer (whoever you paid to 'make it work'), the source code license doesn't matter really in these kinds of things. Just because I make the code to a pacemaker open source and it fails, doesn't mean that I'm not liable. That would be a very easy way to avoid liability though.
Python is single-threaded and thus it's speed and results should be predictable. Not sure whether Python would still run on a decent real-time processor though (x86 has not been in a long time) because it's overhead would be enormous.