Ignoring the quality difference and only focusing on the tech difference for the purpose of argument. Thanks for the reference though!
Not MU-MIMO, only MIMO. Big difference. Sub 900Mb is pretty slow.
I'm willing to pay upwards of $400 for an AP, and I really want one because I want to do VLANs, but I have yet to see a 6 antenna MU-MIMO 1700+ AP.
Another issue that I realized is that I have a better chance of a home wireless router letting me install OpenWRT than an AP. One of the eventual changes to OpenWRT is per device wifi queues. Wifi is only fast because it supports batching, but the network stacks are so dumb that they only batch when there are 2 or more packets in a row destined for the same device. If you get two devices using a lot of bandwidth, instead of batching, you get the wifi sending out one packet at a time since it send out packets FIFO and only batches sequential packets. the proposed changes for OpenWRT have been tested to no do strict FIFO, but instead FIFO per device queue, but round robin the queues instead of the packets. This dramatically increased effective bandwidth while reducing latency for gaming and VoIP.
One of the biggest benefits of MU-MIMO is that they have N number of queues because they need to support transmitting to multiple devices concurrently. This effectively gives your per device queues so long as your number of devices do not exceed the number of concurrent transmissions and the beams do not overlap.
We've increase the precision of what defines a kilogram, but made sure it stayed within the accuracy of modern day measurements. As we get better and better at measuring, we can increase the accuracy of of our measurements with a fixed goal in mind. Even if that goal is arbitrary, it doesn't matter, everything is relative. The notion of concrete concepts are really just abstract concepts of emergent behaviors of an underlying relational system.
Money is a funny thing. Sometimes is closely represents value, other other it's way off. An example is in Minnesota, the state subsidized farmers getting fiber internet. Looking only at the farmer's profit, it's a loss in the number of $$$ at the micro-level. At the macro level, $$$$$ in extra money is now coming into the state as a whole. There are situations where spending X money in one area will gain you Y money in another area, where Y is greater than X.
UW-Madison has a strong computer engineering course. Freshmen that made it past the crazy hard 101 classes were getting called from IBM, Intel, and AMD asking what their plans were post-graduation for designing CPUs.
Wisconsin companies get to use the patent for free and profits are dumped back into the University state system, not just UW-Madison. I was paying $1.6k/sem back when they still help patents on certain ways to work with stem cells. Once the patent ended, prices jumped up. Of course if there is any federal money, the research needs to be open, but UW-Madison has a lot of private funding from alumni and keeps out federal money and uses only state tax money when it looks lucrative.
According to another site, Playboy's website had 4mil visitors per month, but after they switched to no longer have nudes on their website, their visitors went up to 16m per month.
Few sites that give you knowledge compare and contrast the pros and cons among the many different algorithms and their potential uses in hypothetical situations. Assuming you have a decent teacher, you'll get that.
Code that technically works is the worst kind of code. Don't breath, lest you topple over the jinga tower. Many people's code looks like a random walk through a maze. They eventually make it out, but there was no planning involved.
Insult to injury is many times the code is also technically follow best practices. It's so hard to explain to these people that even if the code technically works and follows best practices, it doesn't mean it's good code. It just means it's better code. Following the letter of the law is only so good, you also need to follow the spirit of the law, and that's much harder to convey in an objective way.
I am a developer. My manager is awesome, my job is awesome, the people I work with are awesome. I love going to work. I work with Directors, VPs, and have gotten recognition from a board member. I've never gotten a formal promotion, but my pay keeps skyrocketing. Our number one issue is we can't hire new programmers fast enough.
The other big issue is that programmers are not "cogs". There's a 6 month-1 year learning curve just to break even on a new hire. There's a lot to learn, and very few people have all of the knowledge. We're used to training people. We actually get better results with fresh college students because they listen better. We're stuck with someone for 6 months to a year before we even know if they're any good, at least for the average. Of course I was told 3 months into starting that I was well past expectations and that I was going to get a "correctional" raise. Ever since then, I've been getting a steady set of strong raises.
We don't want more developers, we want more good developers. Increasing the pay just exacerbates the already horrible signal to noise ratio, plus put you on the hook for more payout when you find out how horrible your new hire is. Paying more money wouldn't be an issue if we were guaranteed to get a good hire.
If you're going to college to learn how to code, you're doing it wrong. You learn a lot more much more important and fundamental skills that are nearly impossible to practice in the real world. From day one, they said we're not going to learn how to program, but how to think and solve problems and not make mistakes that other have before us. You cannot program if you don't understand the problem, but you can throw crap at a wall and some times something sticks just enough to be kind of useful.
Even if you pay enough to retaining talent, you still have the issue that finding talent is a mostly fruitless effort. For some companies, they already have enough talent to attract more talent, but most companies don't have any pull, all they have is money. Throwing money at the issue of the lack of programming talent is likely to give you similar results of throwing money at the USA's education system.
The problem with any tech knowledge based education is the knowledge is outdated in 3-5 years. Code monkeys need to be constantly reeducated every few years, like upgrading a computer. If A then B. If C then D. Yep, great education. Now everything falls into the buckets of A, B, C, or D, when it may be an E.
Cheaper in the short run, more expensive in the long run.
I'd be more interested to see a comparison of a 4 years CS grad with another 4 years of work experience up against someone with 8 years of work experience.
Most people don't use VMs for stuff like HA, they use it for environmental isolation. Need to spin up a new Apache instance? Ohh, put Apache on a VM. WTF?!
Your example is a simple case. My employer is a multi-national head-quartered in a small city. They pay decent compared to the city, but poorly compared to the nation. Where they do well is their benefits are top notch. Most people here get more in benefits than in salary. We get something like $30k+ in benefits, because of great health insurance, life insurance, accident insurance, longer term disability insurance any many other things that all employees get.
VMs are rarely the correct answer for environment isolation anyway. VMs are when you absolutely need to run a different OS than the host. For everything else, use jails.
In this hypothetical world of non-scarcity, people don't work. You seem to have missed that part. No one fixes your car, no one makes you a cake, everything does itself because everything is automated.
Some of the higher end DDOS services involve not announcing your blocks on the Internet, but making sure the routes between your and your DDOS service has your blocks announced. This way only your DDOS service can talk directly to you. They act as your gateway to the Internet for your hosted services.
P.S. I love the point you made. I did not think of that hole in the armor.
Slashdot Mirror
Ignoring the quality difference and only focusing on the tech difference for the purpose of argument. Thanks for the reference though!
Not MU-MIMO, only MIMO. Big difference. Sub 900Mb is pretty slow.
I'm willing to pay upwards of $400 for an AP, and I really want one because I want to do VLANs, but I have yet to see a 6 antenna MU-MIMO 1700+ AP.
Another issue that I realized is that I have a better chance of a home wireless router letting me install OpenWRT than an AP. One of the eventual changes to OpenWRT is per device wifi queues. Wifi is only fast because it supports batching, but the network stacks are so dumb that they only batch when there are 2 or more packets in a row destined for the same device. If you get two devices using a lot of bandwidth, instead of batching, you get the wifi sending out one packet at a time since it send out packets FIFO and only batches sequential packets. the proposed changes for OpenWRT have been tested to no do strict FIFO, but instead FIFO per device queue, but round robin the queues instead of the packets. This dramatically increased effective bandwidth while reducing latency for gaming and VoIP.
One of the biggest benefits of MU-MIMO is that they have N number of queues because they need to support transmitting to multiple devices concurrently. This effectively gives your per device queues so long as your number of devices do not exceed the number of concurrent transmissions and the beams do not overlap.
We've increase the precision of what defines a kilogram, but made sure it stayed within the accuracy of modern day measurements. As we get better and better at measuring, we can increase the accuracy of of our measurements with a fixed goal in mind. Even if that goal is arbitrary, it doesn't matter, everything is relative. The notion of concrete concepts are really just abstract concepts of emergent behaviors of an underlying relational system.
Money is a funny thing. Sometimes is closely represents value, other other it's way off. An example is in Minnesota, the state subsidized farmers getting fiber internet. Looking only at the farmer's profit, it's a loss in the number of $$$ at the micro-level. At the macro level, $$$$$ in extra money is now coming into the state as a whole. There are situations where spending X money in one area will gain you Y money in another area, where Y is greater than X.
UW-Madison has a strong computer engineering course. Freshmen that made it past the crazy hard 101 classes were getting called from IBM, Intel, and AMD asking what their plans were post-graduation for designing CPUs.
Wisconsin companies get to use the patent for free and profits are dumped back into the University state system, not just UW-Madison. I was paying $1.6k/sem back when they still help patents on certain ways to work with stem cells. Once the patent ended, prices jumped up. Of course if there is any federal money, the research needs to be open, but UW-Madison has a lot of private funding from alumni and keeps out federal money and uses only state tax money when it looks lucrative.
So no, you're wrong.
According to another site, Playboy's website had 4mil visitors per month, but after they switched to no longer have nudes on their website, their visitors went up to 16m per month.
Few sites that give you knowledge compare and contrast the pros and cons among the many different algorithms and their potential uses in hypothetical situations. Assuming you have a decent teacher, you'll get that.
Why would you get a private loan when the government ones are so easy to get? They hand those out like candy after Halloween.
PNGBlocker 2.0 can invalidate DNS records for known ad servers. I don't block ads, I just make them unresolvable via DNS on my home network.
Code that technically works is the worst kind of code. Don't breath, lest you topple over the jinga tower. Many people's code looks like a random walk through a maze. They eventually make it out, but there was no planning involved.
Insult to injury is many times the code is also technically follow best practices. It's so hard to explain to these people that even if the code technically works and follows best practices, it doesn't mean it's good code. It just means it's better code. Following the letter of the law is only so good, you also need to follow the spirit of the law, and that's much harder to convey in an objective way.
I am a developer. My manager is awesome, my job is awesome, the people I work with are awesome. I love going to work. I work with Directors, VPs, and have gotten recognition from a board member. I've never gotten a formal promotion, but my pay keeps skyrocketing. Our number one issue is we can't hire new programmers fast enough.
The other big issue is that programmers are not "cogs". There's a 6 month-1 year learning curve just to break even on a new hire. There's a lot to learn, and very few people have all of the knowledge. We're used to training people. We actually get better results with fresh college students because they listen better. We're stuck with someone for 6 months to a year before we even know if they're any good, at least for the average. Of course I was told 3 months into starting that I was well past expectations and that I was going to get a "correctional" raise. Ever since then, I've been getting a steady set of strong raises.
banks don't even offer low rates
My FAFSA loans have been 2%-3% interest for the past 8+ years. I have no idea how they even make money on student loans if they match inflation.
We don't want more developers, we want more good developers. Increasing the pay just exacerbates the already horrible signal to noise ratio, plus put you on the hook for more payout when you find out how horrible your new hire is. Paying more money wouldn't be an issue if we were guaranteed to get a good hire.
If you're going to college to learn how to code, you're doing it wrong. You learn a lot more much more important and fundamental skills that are nearly impossible to practice in the real world. From day one, they said we're not going to learn how to program, but how to think and solve problems and not make mistakes that other have before us. You cannot program if you don't understand the problem, but you can throw crap at a wall and some times something sticks just enough to be kind of useful.
Even if you pay enough to retaining talent, you still have the issue that finding talent is a mostly fruitless effort. For some companies, they already have enough talent to attract more talent, but most companies don't have any pull, all they have is money. Throwing money at the issue of the lack of programming talent is likely to give you similar results of throwing money at the USA's education system.
The problem with any tech knowledge based education is the knowledge is outdated in 3-5 years. Code monkeys need to be constantly reeducated every few years, like upgrading a computer. If A then B. If C then D. Yep, great education. Now everything falls into the buckets of A, B, C, or D, when it may be an E.
Cheaper in the short run, more expensive in the long run.
I'd be more interested to see a comparison of a 4 years CS grad with another 4 years of work experience up against someone with 8 years of work experience.
What happens when gold is worth less than the value it represents? Go back a century and gold was everything, but that is no longer true.
Most people don't use VMs for stuff like HA, they use it for environmental isolation. Need to spin up a new Apache instance? Ohh, put Apache on a VM. WTF?!
Your example is a simple case. My employer is a multi-national head-quartered in a small city. They pay decent compared to the city, but poorly compared to the nation. Where they do well is their benefits are top notch. Most people here get more in benefits than in salary. We get something like $30k+ in benefits, because of great health insurance, life insurance, accident insurance, longer term disability insurance any many other things that all employees get.
VMs are rarely the correct answer for environment isolation anyway. VMs are when you absolutely need to run a different OS than the host. For everything else, use jails.
The only discoveries that are centuries away are ones we cannot even fathom today
Fusion
In this hypothetical world of non-scarcity, people don't work. You seem to have missed that part. No one fixes your car, no one makes you a cake, everything does itself because everything is automated.
Some of the higher end DDOS services involve not announcing your blocks on the Internet, but making sure the routes between your and your DDOS service has your blocks announced. This way only your DDOS service can talk directly to you. They act as your gateway to the Internet for your hosted services.
P.S. I love the point you made. I did not think of that hole in the armor.