The scary thing is: this could happen with every brand/vendor/type of plane.
There was an Airbus accident where the pilot got killed because the plane did not brake good enough and crashed with the nose into a building at the airport, reason: because not all wheels had ground contact, the computer did not allow to use reverse engine power to brake the plane. A stupid engineering error...
but often can't be used when later changing things because customizations have been added that would be overwritten Then your generator sucks.
When generating OO code you use a sandwich architecture. Assume you want to have a Customer class, it would look like this: BaseCustomer <- GenCustomer < Customer.
Initially all 3 are generated, because they don't exist. The "generated" code goes into GenCustomer. Now you have two hooks where you can do your manual customization, BaseCustomer (you most likely never touch it, unless you have a special need, e.g. a computed ID or something) and Customer. All your code only uses classes on the level of "Customer"... the rest is managed behind the scene by the generator framework.
Regenerating anything only touches and rewrites the "Gen*" classes and never touches your code in the Base* and "*" classes.
And then again the old war starts: do you put generated code under source control? Yes, you do. And you put it into a proper branch. Because if something goes wrong you simply can merge the code to your liking. Fiddling with two versions of the generator is just more work and more errornous.
I forgot the english name of the pattern, something like "extended inheritance".
I was once in a software project with a team of about 6 or 8, where a friend of mine and I were joking: we don't measure efficiency by the amount of code we write but by the amount of code we delete during refactorings.
activeCustomerAccount = null; Please please please, don't do this. After the CPU returns from the current subroutine, that stack frame is gone... why the funk are there still people who set local variables to null?
So you have a variable, v, and you write the dot behind it, v., the methods applicable pop up? If I see you naming methods in a Customer class containing the word "Customer" again, you will be chained in the basement with water and bread until you repent. And yes, the methods are verbs and start with the most significant verb.
There is nothing wrong with: Customer::create, File::create, Window::create... and so on.
while coders solve the last generation of problems on the next generation of computers (and thus make the code more elegant but less efficient). That is nonsense. A for loop is a for loop is a for loop. It does not lose efficiency just because a coder gets older and a computer gets faster.
We lose efficiency because problems and solutions become more complex and coders are forced to deliver in relatively short time. So they rely on frameworks that focus on simplifying work and ensuring "safety" from mistakes (e.g. in multi threading). Most perceived performance loss is simple a reason of the increased complexity. E.g. facebook serves a billion concurrent users. A systems architecture that scales that way is not the same as a small embedded C program with half a dozen variables running single threaded on a 8-bit micro controller with 16 glorious input/output pins.
I use GPs navigation extremely rarely. Interestingly most map apps, unless they are commercial ones like tom tom, are so often wrong, that GPS basically is only good enough to get a rough clue about your location and direction... is it already the next crossing I have to turn? Or the one after?
I can not understand people who use GPS all the time for simple drives. Looking at the news how often people get lost in simple circumstances, like ending up on a runway, it is just ridiculous.
Close to the equator it is even more irritating. During day time the sun goes that way (e.g. slightly south of you, in case yo are e.g. at 10degrees north), and at night the moon goes another way, because of its inclination versus the earth orbit the moon is slightly north of you. So standing at the same spot at noon your shadow shows north and at midnight (full moon) your shadow points south.
The people did not even know what the experiment was about, moron.
The experimenters measured the brain waves of the test subjects. And they clearly saw that the test subjects noticed the changes in the magnetic fields.
They did not ask them; uh, do you feel anything?
No idea why you want to argue bullshit about stuff you have no clue about and are on top of that to stupid to either read the summary, the article or to grasp it.
Most people don't know that they are sensitive to EM or magnetic fields.
I have a strong sense for orientation, perhaps I can feel magnetic fields... never made such an experiment, so I don't know. I always guessed I use the sun and the stars to orientate... or simply know from looking at a map days ago where north is... no idea.
Basically every migrating bird can sense the earth magnetic field, and that to a degree that they can follow the twists and whirls like the flow of a river to pinpoint their location and not only follow direction.
but I'm betting you will fairly quickly find out that they really don't "feel" it. I know enough people who actually do feel it. I guess if you pay the flight and some compensation for your time, they happily come to you and you can do your "double blind study" (Hint: it is not double blind when the objects know what is tested).
The magnetic fields from wires in the environment are far too weak to have any effect. No they are not. They are ten to hundred times stronger than the earth magnetic field.
(just like cellphone radiation could warm up your brain if it was in the kilowatt range, but it isn't...) Define warm up. 1/100 degree celsius is a warm up.
No, but claiming that it is better than the insurance-based healthcare the US has, tend to make you appear dumb. I think the one who looks dumb is you, after all in Germany we also use an insurance based system... obviously.
But seeing the world today....in western countries, where they are wanting to pick socialism or communism over democracy?!?! But you do know that in the real world you can be communist and democratic same time, or socialist and democratic?
Slashdot Mirror
The scary thing is: this could happen with every brand/vendor/type of plane.
There was an Airbus accident where the pilot got killed because the plane did not brake good enough and crashed with the nose into a building at the airport, reason: because not all wheels had ground contact, the computer did not allow to use reverse engine power to brake the plane. A stupid engineering error ...
Both sensors are for stall warnings. ... which sounds not plausible/logical.
But it seems only one is used for the MCAS system
Yes, it could.
In real world such cases don't exist, just read the rest of the SO article.
Setting a local reference to null is in 99.9999% of all cases WRONG, especially in a finally block as the SO question was asking.
The mystery is: why does it have two sensors, when only one is used for the MCAS system? https://qz.com/1575509/what-we...
Why did someone mod me/my parent "troll"?
Perhaps you want to read how GC works.
Setting a local variable to null is a waste of time and time of the coder and a reason to fire him.
Which part of: "after the subroutine returns, the stack frame is gone" don't you grasp?
In my experience there is no difference between a C++ or Java programmer.
Can you give an example where one is more aware about costs and writes better code regarding that?
but often can't be used when later changing things because customizations have been added that would be overwritten
Then your generator sucks.
When generating OO code you use a sandwich architecture. Assume you want to have a Customer class, it would look like this: BaseCustomer <- GenCustomer < Customer.
Initially all 3 are generated, because they don't exist. The "generated" code goes into GenCustomer. Now you have two hooks where you can do your manual customization, BaseCustomer (you most likely never touch it, unless you have a special need, e.g. a computed ID or something) and Customer. All your code only uses classes on the level of "Customer" ... the rest is managed behind the scene by the generator framework.
Regenerating anything only touches and rewrites the "Gen*" classes and never touches your code in the Base* and "*" classes.
And then again the old war starts: do you put generated code under source control? Yes, you do. And you put it into a proper branch. Because if something goes wrong you simply can merge the code to your liking. Fiddling with two versions of the generator is just more work and more errornous.
I forgot the english name of the pattern, something like "extended inheritance".
I was once in a software project with a team of about 6 or 8, where a friend of mine and I were joking: we don't measure efficiency by the amount of code we write but by the amount of code we delete during refactorings.
activeCustomerAccount = null; ... why the funk are there still people who set local variables to null?
Please please please, don't do this.
After the CPU returns from the current subroutine, that stack frame is gone
You do know that there are IDEs?
So you have a variable, v, and you write the dot behind it, v., the methods applicable pop up? If I see you naming methods in a Customer class containing the word "Customer" again, you will be chained in the basement with water and bread until you repent. And yes, the methods are verbs and start with the most significant verb.
There is nothing wrong with: Customer::create, File::create, Window::create ... and so on.
Understood?
You reduce that by having a replica for "data mining", you don't do that on the life data base.
while coders solve the last generation of problems on the next generation of computers (and thus make the code more elegant but less efficient).
That is nonsense.
A for loop is a for loop is a for loop. It does not lose efficiency just because a coder gets older and a computer gets faster.
We lose efficiency because problems and solutions become more complex and coders are forced to deliver in relatively short time. So they rely on frameworks that focus on simplifying work and ensuring "safety" from mistakes (e.g. in multi threading). Most perceived performance loss is simple a reason of the increased complexity. E.g. facebook serves a billion concurrent users. A systems architecture that scales that way is not the same as a small embedded C program with half a dozen variables running single threaded on a 8-bit micro controller with 16 glorious input/output pins.
I use GPs navigation extremely rarely. ... is it already the next crossing I have to turn? Or the one after?
Interestingly most map apps, unless they are commercial ones like tom tom, are so often wrong, that GPS basically is only good enough to get a rough clue about your location and direction
I can not understand people who use GPS all the time for simple drives. Looking at the news how often people get lost in simple circumstances, like ending up on a runway, it is just ridiculous.
Close to the equator it is even more irritating.
During day time the sun goes that way (e.g. slightly south of you, in case yo are e.g. at 10degrees north), and at night the moon goes another way, because of its inclination versus the earth orbit the moon is slightly north of you. So standing at the same spot at noon your shadow shows north and at midnight (full moon) your shadow points south.
Did you read the summary?
The people did not even know what the experiment was about, moron.
The experimenters measured the brain waves of the test subjects. And they clearly saw that the test subjects noticed the changes in the magnetic fields.
They did not ask them; uh, do you feel anything?
No idea why you want to argue bullshit about stuff you have no clue about and are on top of that to stupid to either read the summary, the article or to grasp it.
Most people don't know that they are sensitive to EM or magnetic fields.
I have a strong sense for orientation, perhaps I can feel magnetic fields ... never made such an experiment, so I don't know. I always guessed I use the sun and the stars to orientate ... or simply know from looking at a map days ago where north is ... no idea.
Basically every migrating bird can sense the earth magnetic field, and that to a degree that they can follow the twists and whirls like the flow of a river to pinpoint their location and not only follow direction.
but I'm betting you will fairly quickly find out that they really don't "feel" it.
I know enough people who actually do feel it. I guess if you pay the flight and some compensation for your time, they happily come to you and you can do your "double blind study" (Hint: it is not double blind when the objects know what is tested).
The magnetic fields from wires in the environment are far too weak to have any effect.
No they are not. They are ten to hundred times stronger than the earth magnetic field.
(just like cellphone radiation could warm up your brain if it was in the kilowatt range, but it isn't...)
Define warm up. 1/100 degree celsius is a warm up.
No, but claiming that it is better than the insurance-based healthcare the US has, tend to make you appear dumb. ... obviously.
I think the one who looks dumb is you, after all in Germany we also use an insurance based system
Every knight is just a "honorary recognition" ... there is no difference between an UK citizen becoming a knight or a non citizen.
And what has that to do with thermodynamics?
BTW: you are wrong anyway ...
But seeing the world today....in western countries, where they are wanting to pick socialism or communism over democracy?!?!
But you do know that in the real world you can be communist and democratic same time, or socialist and democratic?
Perhaps you want to check who or what the IARC actually is: https://www.iarc.fr/featured-n...
Yes, it is proven. There are dozens of studies that clearly proof it.
Here is one: https://www.iarc.fr/featured-n...