The UK CAA recently did some pretty extensive testing, 1.2kg class drones were a significant hazard to helicopters and non-airline airplane windscreens and propulsion surfaces. That means almost the entire firefighting fleet in most cases as only the larger 737 class tankers would be likely to have birdstrike rated windscreens and fanblades, and those are only used on the largest of forest fires.
It's interesting that relatively weak CPU coupled with multiple fast vector processors that could do massive parallel calculations was basically the design of the original Cray supercomputers and we're back to that design coupled with fast interconnects to team up many, many nodes. Kinda cool to see that Seymour had it right =)
The valuation is in line with the 29x current earnings of Facebook. Take the same 27% net profit margin and multiply by the revenue and multiply by 29 and you get $78.3B, a 13% premium for a company with a faster growth rate and better underlying fundamentals (younger audience in this case) is pretty small.
If their revenue estimates are even somewhat accurate then it's hardly tulips, 8x current revenue (say 30x net profits?) is hardly a huge premium for a company that is still growing rapidly.
Except he was 8 months into a 36 month contract, the only reason it expired was a change of ownership on the other parties side. Depending on how the company was acquired it might not have even been legal for the system to do what it did (breech of contract).
He's wrong, I prefer a stable API, unstable APIs and bad vendors is why we have the mess we do with Android and why Treble is the kinda solution, by freezing on a LTSR kernel they've effectively created an artificially stable API which they backport everything else around. In an idealized world, yes I'd love to have every driver in mainline so that everything gets updated together and the kernel devs were free to monkey with things to their hearts content, in the real world I live in a stable API allows vendors to dump a driver once and for it to be usable for a decade or more (see most Windows drivers from Vista still working on 10 today).
With Intel shutting down their commercial support business last year and now LustreFS being removed from the mainline kernel is Lustre dead as a common solution? What is replacing it as a scalable FS in HPC applications?
Why would you buy this over any of the myriad of cheaper or better equipped options? It doesn't even have a killer screen which I could see paying a premium for from Samsung, 12" 1080p is just meh.
I have a clue, I have managed enterprise class routers and firewalls and been using Linux since 1995, I use a Netgear router at home. Their no cost integration with OpenDNS for content filtering and anti-malware protection is better than any opersource solution I have found. They also continue to provide security updates for years after the device is no longer for sale (previous model was ~8 years old when I replaced it for better WiFi performance, it had had a firmware update about 3 months before I retired it for a CVE).
I've only had 2 reviews so far, but both have come with raises closer to 5% than 2.5%, easily covering inflation and actually netting me a few percent. I have no idea if this is typical as the company is notoriously secretive about salaries, but based on the cars and houses my coworkers are buying I don't think anyone is underpayed (a 30 year guy in the next row has an M5, another guy with 15 years has a Golf R as examples)
Actually, I've tried to negotiate additional vacation times multiple times when switching jobs, in every case the best the hiring managers could do is offer me to start off at the company max time. I've received signing bonuses, been repaid for COBRA costs that I incurred when working a temp to hire, even received a salary higher than my boss at one job, but I've never managed to get more than the maximum allowed days/hours of vacation time. It's one reason I'm thinking of doing semi-retirement at 50, work a contract for 6-18 months then take off 6 months before looking for the next contract.
Lasted 9.5 years at my last job, could have stayed longer but an opportunity finally came along that was enough better to get me to move (~40% bump in pay when looking at total compensation package, and I was already well compensated). Been at the new job 2.5 years and I'm still considered one of the new guys, not a month goes by without someone celebrating a 20,25,30 year anniversary. So yes, jobs do still last more than 2 year in 2018. In fact any company not retaining employees for >24 months is seriously messing up, IME it takes 18-48 months for someone to really learn their job and the company culture well enough to be maximally productive within a given environment.
No, I call 911 whenever there is an accident on the freeway that is blocking traffic, because lots and lots of video evidence shows that many drivers are idiots and will drive into a stopped vehicle on the side of the road, let alone one in a travel lane.
No, they ask for what city I'm calling from so that they can direct me to the correct POP, something that should be obvious from the E911 data and which there's a maybe 5% chance that I know the answer to. Even when I'm near home there are 59 municipalities in my county alone, about 200 in the metroplex.
What I find stupid about the E911 mandate is that nearly 20 years after they were first written into the law I STILL get asked where I'm calling from every time I call 911 from a cellphone. Most of the time I have zero clue what jurisdiction I'm in, I just give the highway and mile marker, something they should be able to easily get from a functioning E911 system without wasting time asking me.
Other injustices in the system can (mostly) be corrected at a later date, the death penalty doesn't have an oops, our bad, here's money for your suffering solution. We know absolutely that the death penalty has been carried out in error and yet we still have it for some reason. Once the innocence project started and was able to overturn a non-trivial number of convictions it became obvious that our system is too imperfect to be doing something as irreversible as killing the prisoner.
Yup, was a customer of a major AT&T center in Ashburn, VA, ~400k square feet and no more than 6 people including the security guard there most of the time. That place might have had 20-30 employees total for all 3 shifts. The next center we moved to in Columbus was about 100k square feet and had like 10 employees total onsite for all 3 shifts.
Um, most of the He is reclaimed from Methane wells, the posters point is that currently we only capture a small percentage of all of the He present in all of the wells we tap because it is not economical to do so and so most of it goes through the natural gas distribution system and escapes when the natural gas is burned.
It's not $1B to the local economy, it's a handful of medium pay jobs after the construction is complete, and because datacenters are so specialized the construction is usually handled by a firm that does nothing but plan and build them so you don't even get temporary construction jobs. From a land and resource usage perspective a datacenter is probably one of the worst candidates.
Yeah, looked for OpenBSD and wasn't all that surprised that they weren't affected. Theo is a bit of a douche to work with but he's usually right when it comes to security. OpenBSD also wasn't vulnerable to Meltdown and the OS level variants of Specter because he was already paranoid about cache flushing when thunking between rings due to an earlier Intel bug that he didn't believe was correctly addressed.
Slashdot Mirror
Manna covered this in 2003.
The UK CAA recently did some pretty extensive testing, 1.2kg class drones were a significant hazard to helicopters and non-airline airplane windscreens and propulsion surfaces. That means almost the entire firefighting fleet in most cases as only the larger 737 class tankers would be likely to have birdstrike rated windscreens and fanblades, and those are only used on the largest of forest fires.
MIT has solved industrial production already. Not sure what the cost is but it's definitely far below lab sample $10,000/g levels.
It's interesting that relatively weak CPU coupled with multiple fast vector processors that could do massive parallel calculations was basically the design of the original Cray supercomputers and we're back to that design coupled with fast interconnects to team up many, many nodes. Kinda cool to see that Seymour had it right =)
Facebook's net profit margin is 27%, gross is 47%, I assumed that Instagram would be at least as good.
The valuation is in line with the 29x current earnings of Facebook. Take the same 27% net profit margin and multiply by the revenue and multiply by 29 and you get $78.3B, a 13% premium for a company with a faster growth rate and better underlying fundamentals (younger audience in this case) is pretty small.
If their revenue estimates are even somewhat accurate then it's hardly tulips, 8x current revenue (say 30x net profits?) is hardly a huge premium for a company that is still growing rapidly.
Except he was 8 months into a 36 month contract, the only reason it expired was a change of ownership on the other parties side. Depending on how the company was acquired it might not have even been legal for the system to do what it did (breech of contract).
He's wrong, I prefer a stable API, unstable APIs and bad vendors is why we have the mess we do with Android and why Treble is the kinda solution, by freezing on a LTSR kernel they've effectively created an artificially stable API which they backport everything else around. In an idealized world, yes I'd love to have every driver in mainline so that everything gets updated together and the kernel devs were free to monkey with things to their hearts content, in the real world I live in a stable API allows vendors to dump a driver once and for it to be usable for a decade or more (see most Windows drivers from Vista still working on 10 today).
With Intel shutting down their commercial support business last year and now LustreFS being removed from the mainline kernel is Lustre dead as a common solution? What is replacing it as a scalable FS in HPC applications?
Sure we do, Neil Degrasse Tyson, and before him Sagan.
Why would you buy this over any of the myriad of cheaper or better equipped options? It doesn't even have a killer screen which I could see paying a premium for from Samsung, 12" 1080p is just meh.
I have a clue, I have managed enterprise class routers and firewalls and been using Linux since 1995, I use a Netgear router at home. Their no cost integration with OpenDNS for content filtering and anti-malware protection is better than any opersource solution I have found. They also continue to provide security updates for years after the device is no longer for sale (previous model was ~8 years old when I replaced it for better WiFi performance, it had had a firmware update about 3 months before I retired it for a CVE).
I've only had 2 reviews so far, but both have come with raises closer to 5% than 2.5%, easily covering inflation and actually netting me a few percent. I have no idea if this is typical as the company is notoriously secretive about salaries, but based on the cars and houses my coworkers are buying I don't think anyone is underpayed (a 30 year guy in the next row has an M5, another guy with 15 years has a Golf R as examples)
Actually, I've tried to negotiate additional vacation times multiple times when switching jobs, in every case the best the hiring managers could do is offer me to start off at the company max time. I've received signing bonuses, been repaid for COBRA costs that I incurred when working a temp to hire, even received a salary higher than my boss at one job, but I've never managed to get more than the maximum allowed days/hours of vacation time. It's one reason I'm thinking of doing semi-retirement at 50, work a contract for 6-18 months then take off 6 months before looking for the next contract.
Lasted 9.5 years at my last job, could have stayed longer but an opportunity finally came along that was enough better to get me to move (~40% bump in pay when looking at total compensation package, and I was already well compensated). Been at the new job 2.5 years and I'm still considered one of the new guys, not a month goes by without someone celebrating a 20,25,30 year anniversary. So yes, jobs do still last more than 2 year in 2018. In fact any company not retaining employees for >24 months is seriously messing up, IME it takes 18-48 months for someone to really learn their job and the company culture well enough to be maximally productive within a given environment.
No, I call 911 whenever there is an accident on the freeway that is blocking traffic, because lots and lots of video evidence shows that many drivers are idiots and will drive into a stopped vehicle on the side of the road, let alone one in a travel lane.
No, they ask for what city I'm calling from so that they can direct me to the correct POP, something that should be obvious from the E911 data and which there's a maybe 5% chance that I know the answer to. Even when I'm near home there are 59 municipalities in my county alone, about 200 in the metroplex.
What I find stupid about the E911 mandate is that nearly 20 years after they were first written into the law I STILL get asked where I'm calling from every time I call 911 from a cellphone. Most of the time I have zero clue what jurisdiction I'm in, I just give the highway and mile marker, something they should be able to easily get from a functioning E911 system without wasting time asking me.
Lethal injections are pretty much a non-option due to EU laws and the fact that almost all pharmaceutical companies are multinational.
Other injustices in the system can (mostly) be corrected at a later date, the death penalty doesn't have an oops, our bad, here's money for your suffering solution. We know absolutely that the death penalty has been carried out in error and yet we still have it for some reason. Once the innocence project started and was able to overturn a non-trivial number of convictions it became obvious that our system is too imperfect to be doing something as irreversible as killing the prisoner.
Yup, was a customer of a major AT&T center in Ashburn, VA, ~400k square feet and no more than 6 people including the security guard there most of the time. That place might have had 20-30 employees total for all 3 shifts. The next center we moved to in Columbus was about 100k square feet and had like 10 employees total onsite for all 3 shifts.
Um, most of the He is reclaimed from Methane wells, the posters point is that currently we only capture a small percentage of all of the He present in all of the wells we tap because it is not economical to do so and so most of it goes through the natural gas distribution system and escapes when the natural gas is burned.
It's not $1B to the local economy, it's a handful of medium pay jobs after the construction is complete, and because datacenters are so specialized the construction is usually handled by a firm that does nothing but plan and build them so you don't even get temporary construction jobs. From a land and resource usage perspective a datacenter is probably one of the worst candidates.
Yeah, looked for OpenBSD and wasn't all that surprised that they weren't affected. Theo is a bit of a douche to work with but he's usually right when it comes to security. OpenBSD also wasn't vulnerable to Meltdown and the OS level variants of Specter because he was already paranoid about cache flushing when thunking between rings due to an earlier Intel bug that he didn't believe was correctly addressed.