QA takes money, and Android devices are sold with basically no margin. How are you going to pay for that team?
Answer: you aren't.
Apple's QA team has enough problems with its limited set of updates and devices, and Apple has a huge pile of money and presumably decent processes. Any android manufacturer would find it impossible to handle multiple update streams on multiple hardware platforms.
Node.js is neat. We have a production service running on it that handles a couple of hundred thousand messages a day sent by tens of thousands of devices. It's not much, but it's only about 700 lines of javascript. That's including comments and whitespace.
The callback mechanism takes some getting used to. If you haven't programmed interrupt handlers or async I/O it'll be confusing, because the flow-of-control isn't really that obvious.
Why did we choose node? Our understanding was that it's great at concurrency, and this was a small and fast project to test it out. We're using it as a stepping stone to Apache Lambda, with the goal of throwing our message infrastructure into Apache at some point.
We didn't try to replace a LAMP stack with it. I suspect that would be a bad use for node...although given a low workload it could probably be done.
I brought my Echo into the office, and it's been able to recognize at least 5 separate people with no issues from across the room. At home, it recognized everyone's voice, kids and adults. Three people in the office have bought one after using mine.
So your anecdote, like mine, means nothing overall.
I had to bring my Echo in because my kids were continuously asking Alexa for jokes, which gets unbelievably annoying.
This "thing" shows the interesting interaction between the engineering community and the scientific community. This is why you should take scientist statements with a grain of salt.
Engineers: look, this works! Scientists: that violates the laws of science and is impossible. Engineers: who gives a sh*t what you think? Here's the data Scientists: the data must be wrong Engineers: you try it Scientists: we have no f*cking idea what's happening, but it's happening Engineers: f*cking pinheads Scientists: oh, maybe this is what's happening
If you take scientists too seriously, you never get past step #1.
The food scientist consensus was that saturated fats etc made people fat, so they switched America to a low-fat high-carb diet. Well evidence shows that's total BS, yet they discount it.
They don't teach error handling either. How many handouts in CS have said "error handing as an exercise left for the reader?" if it's mentioned at all.
However, it's arguably one of the most difficult designs you can make when you write software.
Android would have died without the write-once run-badly-everywhere aspects of Java. Google could have written its own, but it chose to copy Java's APIs.
It's not like there were a lot of multi-platform options out there. Tk/TCL?
Time to pay up, google. Your technical debt is calling.
Slashdot Mirror
QA takes money, and Android devices are sold with basically no margin. How are you going to pay for that team?
Answer: you aren't.
Apple's QA team has enough problems with its limited set of updates and devices, and Apple has a huge pile of money and presumably decent processes. Any android manufacturer would find it impossible to handle multiple update streams on multiple hardware platforms.
Node.js is neat. We have a production service running on it that handles a couple of hundred thousand messages a day sent by tens of thousands of devices. It's not much, but it's only about 700 lines of javascript. That's including comments and whitespace.
The callback mechanism takes some getting used to. If you haven't programmed interrupt handlers or async I/O it'll be confusing, because the flow-of-control isn't really that obvious.
Why did we choose node? Our understanding was that it's great at concurrency, and this was a small and fast project to test it out. We're using it as a stepping stone to Apache Lambda, with the goal of throwing our message infrastructure into Apache at some point.
We didn't try to replace a LAMP stack with it. I suspect that would be a bad use for node...although given a low workload it could probably be done.
If headers aren't copyrightable, why do headers have copyright statements in them?
Yeah, how many white people have actually seen White Chicks? Anyone? Anyone?
It's funny, pharmas were just busted for promoting off-label use. Why isn't this considered just another off-label use?
I asked my dog and he didn't say anything so we're good, right?
I brought my Echo into the office, and it's been able to recognize at least 5 separate people with no issues from across the room. At home, it recognized everyone's voice, kids and adults. Three people in the office have bought one after using mine.
So your anecdote, like mine, means nothing overall.
I had to bring my Echo in because my kids were continuously asking Alexa for jokes, which gets unbelievably annoying.
Does the FTC have the authority to compel the production of this information?
We want all your money, and we want it now!
- your government
This book already documented the effect.
If the vendors are illegal, they should just call the police/HOA to get rid of them.
You know, there's like no chance of winning the lottery, but someone usually wins the lottery every week.
It's hilarious that someone is complaining that an Appl product has too many buttons
If you think it sucks donkey balls, write a competitor. It's only been the dominant music app for more than a decade.
Yet engineers do not deny what's in front of them because there's no theory explaining it.
This "thing" shows the interesting interaction between the engineering community and the scientific community. This is why you should take scientist statements with a grain of salt.
Engineers: look, this works!
Scientists: that violates the laws of science and is impossible.
Engineers: who gives a sh*t what you think? Here's the data
Scientists: the data must be wrong
Engineers: you try it
Scientists: we have no f*cking idea what's happening, but it's happening
Engineers: f*cking pinheads
Scientists: oh, maybe this is what's happening
If you take scientists too seriously, you never get past step #1.
My cablecard is free. Isn't everyone's? ISP: comcast.
The food scientist consensus was that saturated fats etc made people fat, so they switched America to a low-fat high-carb diet. Well evidence shows that's total BS, yet they discount it.
So what does a scientific consensus really mean?
It's weird - why didn't they take all 10?
You don't have to insult them. Just do their job for them and they'll get the point.
They don't teach error handling either. How many handouts in CS have said "error handing as an exercise left for the reader?" if it's mentioned at all.
However, it's arguably one of the most difficult designs you can make when you write software.
5 hit product lines
Three (or four, if x86 -> x86_64 counts) architecture transitions inside one product line (68k -> PPC -> x86 -> x86_64)
Two industries destroyed (music, mobile phones).
One industry revolutionized (computing)
That's pretty good for a company that's been going out of business for 40 years
What did the FBI say when he asked them for help?
So did the Apple ][. Plus it had slots.
Android would have died without the write-once run-badly-everywhere aspects of Java. Google could have written its own, but it chose to copy Java's APIs.
It's not like there were a lot of multi-platform options out there. Tk/TCL?
Time to pay up, google. Your technical debt is calling.