In a matter of speaking, customer's are morons. They don't know what they need, only what they want, and what they want is many times contradictory to itself. One of two things need to happen. The AI needs to dance that fine line between telling the end user they're retarded or we need to also get rid of the human end user.
MS SQL is pretty much feature parity with Oracle, just with different names. The main difference is Oracle charges more so they must be better. MySQL is a pile of crap that horribly violates the principle of least astonishment when it comes to ACID, and when you want ACID, it's slow as fk. Postgresql supports the most micro-optimizations, is wonderfully awesome, but has virtually no mind-share making it hard to find admins.
Row based locking, very expensive, but fine for a few rows.
Page based locking, best when used for contiguous pages for a small portion of the table
Table lock, fastest lock, but Amdahl's law hates this for many write locks.
Data structures are important for performance tuning and in SQL, relations and tables are the structures. You can't design databases with just set theory. You need to understand how the engines work if you want them to perform well. SQL server does a very good job with great performance while remaining ACID. Most databases that are "Faster" are because they gave up some portion of ACID. May as well just use a NO-SQL database./exagerated
AC is very efficient for transitions without using exotic equipment. Barrier to entry is much much lower. And efficiently transmitting DC requires very dangerous voltages.
Seem much easier to search for "pointer confinement" "pointer locking" to get hits on Wayland. Seems Wayland won't let you move the cuusor like X does, but it will, eventually, allow you to confine or lock it. Seems to be taking fooorrrreeeevvvveeeerrrrr.
No wonder Wayland is taking so long. The core Wayland team shares a lot of the same people as the core Xorg team. You do realize that people who made Xorg are mostly the people making Wayland because Xorg is such crap from the multi-decades of backwards comparability and old architecture that is no longer reflective of modern systems. Bailing wire and duct tape.
Value is distributed in a power curve. The top 20% create about 60% of the value. If all is fair, then the top 20% get 60% of the money. That's not even at master level yet. A master may be in the top 20% of the 20% of the 20%. That means the top 0.8% get 21.6% of the money. Of course if you don't want to use value to drive your decisions, you're perfectly allowed to pay the janitors the same amount as the engineers.
I had a senior "software engineer" ask me for a list of IDs from a database. So I sent him the list, just separated by linebreaks. He replied with "can you resend the data, but comma delimited". So I opened up the email I sent, copied the contents into my favorite text editor and did a mass-replace of "\r\n" to "," for him.
I had another "software engineer" that learned about the singleton pattern. Made me an library to call that implemented it as a system wide mutex. I could not create the same object more than once in the entire system. When I confronted him, because I had some situations where I actually needed to call the library on the side, he said that it's the proper pattern since I should not need to create the object more than once. So I just ran the library call from another server.
Math is always tied to reality, but mathematical models are not. Math is a human created language that allows us to describe the rules of our Universe. I don't mean "rules" like laws of physics, but logical rules that apply in all situations. The real "truths" of the Universe. eg If you can prove the A is greater than B, then you know for a fact that B is less than A. This is a fundamental understanding of how our Universe's logic works.
Virtual worlds are perfectly valid worlds. Math in and of itself has no world. Physics is the application of math to the real world and CS is the applied of math to a virtual world that resides inside the real world.
I don't learn well that way. I learn best from learning the 80/20 of theory, then going on to practice. Then revisit the last 20% of the theory. When I was young I had to learn about 25 new words per week in school and I typically got a C or D. Now that I'm in my 30s, I am self-teaching Japanese with about 50 words per week in a language I had no prior background in.
Theory gives me a framework in which to place what I'm learning from practice. Without theory, I am forced to use my memory for everything. With theory, I can at least create some rules to reduce the amount of rote memorization.
When it comes to programming, the first thing I do is go read up on best practice, then I read up on theory, then I dive into the details.
I use algebra just to manage my debt. Anyway, who plans on doing 4 years of statistics? At least when I went to school, we were done with Algebra 2 at the end of 8th grade. Still got 3 more years of required math classes. Or do what I did and transfer schools and retake the same math classes but under a different name. Took Geometry and Algebra 2 at the same time. Slept through the classes and got over 100% in both. Yay, extra credit test questions. Great way to boost your GPA.
You underestimate 40Gb/s. That's probably nearly as fast your ISP's local trunk, but instead of sharing that with 100k people, you're only sharing it with 256. Few ISPs use a 256 split because of the signal strength reduction.
WDP-PON is 40Gb split 32 ways with a dedicated 1.25Gb per end point
XG-PON1 is 10Gb down 2.5Gb split 32 ways, non-dedicated but up to 10Gb per device
XG-PON2 is 10Gb down and up split 32 ways, non-dedicated but up to 10Gb per device
TWDM-PON is 80Gb down and up split 32 ways, non-dedicated but up to 10Gb per device
NG-PON2 is similar to TWDM-PON, but caps out at 40Gb for the current implementation. Where it really stands out is the standard should scale up to 25Tb/s with 100Gb per customer. This makes it desirable for a smooth upgrade path into the far future.
I haven't met anyone that dislikes Windows 10. They either like it or have no opinion. My home computer is the last computer that I come in contact with that has Win7. Every work computer in my department and everyone in my family now have Win10 and nearly everyone likes it. There are some Linux users who have to use Windows, but they don't think Win10 is any worse than Win7, but they do abhor Win8.x.
A Mole of Moles: What would happen if you were to gather a mole (unit of measurement) of moles (the small furry critter) in one place? https://what-if.xkcd.com/4/
Slashdot Mirror
In a matter of speaking, customer's are morons. They don't know what they need, only what they want, and what they want is many times contradictory to itself. One of two things need to happen. The AI needs to dance that fine line between telling the end user they're retarded or we need to also get rid of the human end user.
That's just Alaska, not the entire world.
MS SQL is pretty much feature parity with Oracle, just with different names. The main difference is Oracle charges more so they must be better. MySQL is a pile of crap that horribly violates the principle of least astonishment when it comes to ACID, and when you want ACID, it's slow as fk. Postgresql supports the most micro-optimizations, is wonderfully awesome, but has virtually no mind-share making it hard to find admins.
Row based locking, very expensive, but fine for a few rows.
/exagerated
Page based locking, best when used for contiguous pages for a small portion of the table
Table lock, fastest lock, but Amdahl's law hates this for many write locks.
Data structures are important for performance tuning and in SQL, relations and tables are the structures. You can't design databases with just set theory. You need to understand how the engines work if you want them to perform well. SQL server does a very good job with great performance while remaining ACID. Most databases that are "Faster" are because they gave up some portion of ACID. May as well just use a NO-SQL database.
AC is very efficient for transitions without using exotic equipment. Barrier to entry is much much lower. And efficiently transmitting DC requires very dangerous voltages.
That's just asking for trouble. There should be a separate helper interface daemon that is heavily locked down and well tested that can modify files.
Seem much easier to search for "pointer confinement" "pointer locking" to get hits on Wayland. Seems Wayland won't let you move the cuusor like X does, but it will, eventually, allow you to confine or lock it. Seems to be taking fooorrrreeeevvvveeeerrrrr.
... pointer confinement have all landed this cycle"
Good news "MARCH 4, 2016
https://blogs.gnome.org/mclase...
No wonder Wayland is taking so long. The core Wayland team shares a lot of the same people as the core Xorg team. You do realize that people who made Xorg are mostly the people making Wayland because Xorg is such crap from the multi-decades of backwards comparability and old architecture that is no longer reflective of modern systems. Bailing wire and duct tape.
The real question is why the web daemon even had permission to modify files.
Value is distributed in a power curve. The top 20% create about 60% of the value. If all is fair, then the top 20% get 60% of the money. That's not even at master level yet. A master may be in the top 20% of the 20% of the 20%. That means the top 0.8% get 21.6% of the money. Of course if you don't want to use value to drive your decisions, you're perfectly allowed to pay the janitors the same amount as the engineers.
I had a senior "software engineer" ask me for a list of IDs from a database. So I sent him the list, just separated by linebreaks. He replied with "can you resend the data, but comma delimited". So I opened up the email I sent, copied the contents into my favorite text editor and did a mass-replace of "\r\n" to "," for him.
I had another "software engineer" that learned about the singleton pattern. Made me an library to call that implemented it as a system wide mutex. I could not create the same object more than once in the entire system. When I confronted him, because I had some situations where I actually needed to call the library on the side, he said that it's the proper pattern since I should not need to create the object more than once. So I just ran the library call from another server.
Math is always tied to reality, but mathematical models are not. Math is a human created language that allows us to describe the rules of our Universe. I don't mean "rules" like laws of physics, but logical rules that apply in all situations. The real "truths" of the Universe. eg If you can prove the A is greater than B, then you know for a fact that B is less than A. This is a fundamental understanding of how our Universe's logic works.
Virtual worlds are perfectly valid worlds. Math in and of itself has no world. Physics is the application of math to the real world and CS is the applied of math to a virtual world that resides inside the real world.
I don't learn well that way. I learn best from learning the 80/20 of theory, then going on to practice. Then revisit the last 20% of the theory. When I was young I had to learn about 25 new words per week in school and I typically got a C or D. Now that I'm in my 30s, I am self-teaching Japanese with about 50 words per week in a language I had no prior background in.
Theory gives me a framework in which to place what I'm learning from practice. Without theory, I am forced to use my memory for everything. With theory, I can at least create some rules to reduce the amount of rote memorization.
When it comes to programming, the first thing I do is go read up on best practice, then I read up on theory, then I dive into the details.
Se we need to selectively lengthen telomeres
I use algebra just to manage my debt. Anyway, who plans on doing 4 years of statistics? At least when I went to school, we were done with Algebra 2 at the end of 8th grade. Still got 3 more years of required math classes. Or do what I did and transfer schools and retake the same math classes but under a different name. Took Geometry and Algebra 2 at the same time. Slept through the classes and got over 100% in both. Yay, extra credit test questions. Great way to boost your GPA.
Fixing OpenSSL is a futile endeavor of polishing crap. Some things can only be fixed by being taken out back.
Not the same team, but part of the same community and mindset that enables such poor quality code to exist.
*I use "Linux" to mean the Linux community, and not the kernel, which is of decent standards.
holding down the on/off button to improperly shut the computer off when it freezes
There's an improper way to shut down an unresponsive computer? What about system's without reset buttons?
A 1GW coal power plant produces enough nuclear waste to power a 1.1GW nuclear power plant. It is a horrible way to produce energy.
You underestimate 40Gb/s. That's probably nearly as fast your ISP's local trunk, but instead of sharing that with 100k people, you're only sharing it with 256. Few ISPs use a 256 split because of the signal strength reduction.
WDP-PON is 40Gb split 32 ways with a dedicated 1.25Gb per end point
XG-PON1 is 10Gb down 2.5Gb split 32 ways, non-dedicated but up to 10Gb per device
XG-PON2 is 10Gb down and up split 32 ways, non-dedicated but up to 10Gb per device
TWDM-PON is 80Gb down and up split 32 ways, non-dedicated but up to 10Gb per device
NG-PON2 is similar to TWDM-PON, but caps out at 40Gb for the current implementation. Where it really stands out is the standard should scale up to 25Tb/s with 100Gb per customer. This makes it desirable for a smooth upgrade path into the far future.
I haven't met anyone that dislikes Windows 10. They either like it or have no opinion. My home computer is the last computer that I come in contact with that has Win7. Every work computer in my department and everyone in my family now have Win10 and nearly everyone likes it. There are some Linux users who have to use Windows, but they don't think Win10 is any worse than Win7, but they do abhor Win8.x.
If you assume Earth is the only planet with life, then the Universe is not in a sweet spot, just an unlikely anomaly with a selection bias.
A Mole of Moles: What would happen if you were to gather a mole (unit of measurement) of moles (the small furry critter) in one place?
https://what-if.xkcd.com/4/