There is no problem that one can solve with object oriented design that cannot also be solved with structs and function pointers.
There are some problems that can be solved with structs and function pointers which cannot be solved with object oriented design (without duplicate code).
I'm going to lose the mod points I gave, but whatever...
RAII is only useful for pointers. Dtors cannot throw any exceptions, that means that RAII is useless for anything else. See my post above.
Dafuq? Are you serious?
A GC is a more elegant solution for memory management.
There are entire classes of problems for which a GC is not acceptable, not just in the embedded realm, but in the application/system/large memory realms. There are reasons why the Java VM provides for direct byte buffers or why.NET allows you to jump from managed code to un-managed code.
I'm a recurring canceler on Netflix... as a Canadian with half of the content, I run out of content after a while and simply cancel my subscription to renew it a few months later and have new stuff pup up that interests me.
That's the cool thing with Netflix (and Hulu). We can cancel and re-join at any time without paying penalties. Try that with cable:/
While Netflix left me wanting when they originally split streaming off as a separate service, I canceled the DVD portion in favor of keeping the on-demand streaming. They don't have EVERYTHING in their catalog, but compared to Hulu and Amazon Prime I am more likely to find what I want to watch in Netflix. That is true both for original content and regular content. Hulu I got solely to get 11.22.63 since I liked the book. After that I didn't have a reason to keep it based on the few shows I had a vague interest in. Amazon Prime I kept primarily because of its connection to Prime Shipping, but I have watched Prime few enough times that I'm ready to cancel it next month. Maybe it's just because I'm so accustomed to Netflix, but I feel I can find something to watch on Netflix that I will enjoy nearly 100% of the time. And with Prime and others I just don't have that hit rate. Could even be that Netflix UI is just better for me that the others.
Interesting. I'm about to cancel Amazon Prime because 1) I don't use shipping that much, and 2) most of the shows I watch are already on Netflix and Hulu. As for these two, sometimes I want to cancel one or the other, depending what's on the show.
I'm keeping Hulu because it carries the latest Naruto Shippuden episodes (yeah, I'm a fan) and The Daily Show, my kids watch Sailor Moon and Dragon-Ball and my wife watches Dance Moms and Modern Family. Plus Hulu has a ton of really good old movies from the Criterion Edition (we are old-movie buffs.)
Netflix on the other hand, I've been close to cancelling it because it doesn't really update their movie catalogs or TV shows that often. What is keeping me with Netflix are its originals (Peaky Blinders, Marco Polo, Narcos) plus "Jane the Virgin".
I think for binge-TV-watchers like us, two streaming services are always needed. For us, Hulu and Netflix fit the bill.
And if I see any ads, I'm canceling. Why would I pay for service with ads? Makes no damn sense.
And if you cancel, where do you go? Other than Hulu and Amazon Video, there is not much else that can compete in terms of content for a dirt-cheap price.
Even with adds, it would still be cheaper than cable. A combination of netflix with some other streaming services and an internet connection is the closest thing we have to a-la-carte cable.
So until cable companies decide to provide a-la-carte plans that people can cancel at any times without penalties, I'd be willing to see adds on Netflix. I hate adds, but I cannot be so dogmatic when it comes to what I'm getting for every buck I pay.
Hey google - why don't you train 2 million Americans?
Google is a multinational company, and most of their profits come from outside the USA. To add, India and China are the two major areas of growth into the future. The USA is that faded boxer who doesn't have it any more but still hasn't come to grips that he is washed up.
The answer to your question is one you will not like but is no less true even if you don't: They are training Indians instead of Americans because India is the future, and America is the past.
India is the future? Tell me something, has the "future" managed to put a fucking toilet above a cell phone on the priority list yet?
Based on this latest job push, I'm guessing not. Good to know our "future" has its priorities in order.
Unfortunately India is plagued by the same problems that plague Brazil. Wealth accumulated on top, an educated middle class, and then swats and swats and more swats of people living in filth and misery. Both countries excel at not making their poor's well being a priority. **
** With that said, I think India will fix that before Brazil. India is improving little by little. Brazil keeps sliding in the opposite direction.
Hey google - why don't you train 2 million Americans?
Google is a multinational company, and most of their profits come from outside the USA. To add, India and China are the two major areas of growth into the future. The USA is that faded boxer who doesn't have it any more but still hasn't come to grips that he is washed up.
The answer to your question is one you will not like but is no less true even if you don't: They are training Indians instead of Americans because India is the future, and America is the past.
The better question to ask yourselves as Americans is where you went wrong. You used to be on top of the world. Now China makes everything you used to make, and India is eating into the programming. You used to make the best cars, now that's Japan.
Look to yourselves before you blame others. It is harder to do, but only by accepting where you went wrong do you have a hope of recovering from a tail spin.
India, like Brazil, has too many systemic problems right now for the world to consider it "the future". China, yes. And to a lesser extend Eastern Europe and some parts of Latin America (in particular Mexico which has been getting its shit together for the last 2 decades.)
Heck, even Japan with its aging population will be part of the future for a long time before India. Why, robotics, efficiency and personal/business savings up to the wazoo.
For India to overcome China (and other regions of the world) as "the future", it needs to overcome them in terms of manufacturing (both low-value added and high-value added manufacturing.)
I find that extremely unlikely (and no, don't construe my words as a bashing of India, I want India to progress and for her citizens to live better.) But it is simply not set up to be a global manufacturing hub (unlike China or Indonesia or Malaysia or Japan or Mexico.)
In the services area, however, I think India is well posed, and it will do well hitting that.
Seagate Fires 6,500, Or 14% of Workforce, Stock Soars
Of those laid off, how much of that was due to improved productivity?
What were the roles?
Were these engineers?
Or workers whose jobs could be automated?
Or how many of these were "dead weight" or in redundant functions?
Or where people blindly laid off due to getting paid more than a certain salary cap, regardless of function just to make the spreadsheets pretty? *** (I've seen this.)
Some layoffs are totally destructive. Others are completely justified. I've seen both types. The article doesn't tell me shit about it.
. Startup culture isn't for the young either; you really have to have the fraternity/sorority member personality type to work there so as people age they're less likely to trade salary for beer pong or free dinner.
I disagree that it is not for the young. They want people who can work 70-100 hours a week. Those people do not have families. If they do their wife does not work. Or if they both work then they have no kids. Startups should have no problem with a 30% turnover rate or higher.
They want people who appear busy for 70-100 hours a week. 70-100 hours a week implies between 10 to 16 hours a day (depending of whether you work 6 days or work without taking a day off.) Then add 1-2 hours of commute, we are talking 10 to 18 hours a day.
No matter how young you are, it is physically impossible to do *actual* work at that rate for more than a few weeks (or months at most.) From experience (and purely anecdotal), the physical and psychological upper limit seems to be 60 hours a week (less if you count commute times). And that translate to 10 to 12 hours every day (whether you work 5 or 6 days) sustained to no more than 8-12 months.
After that, attrition inevitably follows. I've seen companies having attrition rates of 50-60% (and worse) when they force their workforce through that type of grind for more than a year or two.
Unless you are building the next rocket to Mars or some other incredibly cool shit, you are an idiot if you subject yourself to that. You do not need that to gain valuable experience.
Tech Workers Think Silicon Valley and Startups Are Losing Their Luster
The cost of living in the valley is so ridiculous now that one has to make double (or more) than what you could make in Dallas or Denver just to afford a townhome in a decent zip code. This has always been a serious consideration for couples, and it has started to be for the new wave of college grads who are more financially conscious than their predecessors (yes, there is a measurable change in spending habits among younger people.)
Then there is the snobbery of the hiring process. By God, the snobbery by which some companies use when sending rejection letters and e-mails. I've never seen anything like this anywhere.
And last but not least, the ageism. It exists, it is an open secret. So what kind of outcome can a region expect to get when it methodically culls senior people that have accumulated experience on the trenches? Sooner or later that kind of shit is going to come back.
Put all of these three on top of the definition of innovation as "inventing the next mobile/social shit app". Innovation occurs predominantly where software meets hardware to provide unique solutions and opportunities. That is not the valley's bread and butter anymore. Too many unicorns.
So yeah, loss of luster is not surprising. At. All.
You left out the third option.
1,100 jobs where saved by automation. Hostess went out of business and several other companies bought up the rights to the products that Hostess made.
This is the wrong focus. A more accurate (and interesting question) is how often do you switch stacks or architectures?
Embedded programming using, say, the Wind River toolchain can be quite distinct from doing Linux systems programming using the GNU tool chain. And these two are quite distinct from Windows systems programming using (duh) Windows toolchains. And these are all done (typically) in C. Similarly, C++ programming on Linux will be different from Windows programming, and more so when we start adding stacks and libraries like Qt when the objective is to use Qt's higher level capabilities instead of OS portability (which sometimes is not a bad thing.)
Same with Java. Right now I'm doing core Java, dealing directly with NIO, threading, etc, all outside the safety net of a container. For better or worse, our work does not use OSGI (and that would have been a very distinct type of work if such an architectural choice have been made.)
That is quite distinct from EE development. Working specifically to stay within a web container can be quite distinct from working on something *that exploits* every nook and cranny off an EE container. Then you can throw Spring (and/or OSGI) into the mix which can cut through every one of these dimensions in the Java world. And let's not get started when you move into a completely different paradigm (say vert.x)
In Javascript, this is even more radical since modern Javascript (for better or worse) has become some sort of meta-language that sits below a higher-level framework or architecture - be it jquery or dojo or extjs in the recent past, AngularJX, backbone, React and what now in our current times. Each of these poses a learning (and possibly architectural) challenge.
One sees the same with Python and Ruby to a lesser extend.
The.NET world provides more continuity in how things are build and a developer can, with relative ease, jump back and forth from C# to Managed C++.
And in the application space you have an additional dimension - application data storage. SQL or NoSQL. That will change how you design and code.
So the question that is more relevant is, how often do you change platforms and stacks. Because you could very well be using the same language for 6 years and have experienced a significant development and architectural shift.
It's hard to believe that testing 32-bit really doubles the testing effort, but whatever.
One of the biggest parts of testing consists of keeping test systems up and running and ready to run the tests, every retired system is less labor.
Virtual machines need updating and maintenance, too, they are not "free" by any stretch, again labor is saved when they are retired.
Often tests are not "totally bulletproof", with failures from system problems or incorrect setup, much labor is involved here in investigation and/or mitigation, again directly related to the number of systems to test.
This and exactly this. Another thing people forget is that testing not only cover fresh installations, but also upgrades. So the testing permutations begin to fly off the handle with the number of supported platforms combined with the number of upgrade paths an organization needs to test.
could turn a 300-mile trip that would normally take 3.5 hours flying into a breezy 28-minute ride
Japan's SCMaglev clocked a speed record of 375 mph in 2015. So that same 3.5 hour flight can be turned into a 1 hour trip. Yes, 28 minutes is better, but economically speaking, a nation could take the 1 hour trip instead with proven technology that already exists.
Sooner or later, maglev trains will achieve speeds over 500 mph. They'll never break the sound barrier (which is Hyperloop's promise), but, economically, they do not have to. One cannot justify the expense, at this time, to create a hyperloop to make a mere 300-mile trip in 30 minutes.
This is also why I wonder the economic sense of building one between Los Angeles and the Valley instead of building a "bullet" train that could make the trip in 1 hour. A hyperloop makes sense for transcontinental travel, to connect distances of over 1,000 miles, and in some cases, to connect sufficiently distant end-points in a country (say, making a 2k-mile trip connecting Kyushu with Hokkaido.)
Don't get me wrong. I like the idea of a hyperloop. We should push our technological capabilities. But we should also see what things make economic sense. I do not see the Hyperloop making sense for a 300-mile trip.
If you have a cell phone, why would having VOIP connection stop you from calling?
The premise was that calling is not optimal and email or text chat should be preferred. Sure, with a cell phone you could call tech support, but if the problem can be resolved more quickly by email or text chat, you could use the internet connection of the phone for that purpose.
From my experience, calling tech support with a cell phone is even more of a nightmare than calling from a a landline due to the inferior quality.
But that would be a problem with the signal and only if you live in the boondocks.
The crux of the problem is in the process of getting your call received and processed, talking to the rep on the other line, etc. That is an invariant whether you call from a land line or a cell phone (who in metro areas have a land line nowadays?) Signal quality takes the back seat on this.
That's in part because new cars are loaded with helpful but expensive safety features like collision-avoidance systems.
Wrong factors. Median income people haven't been able to afford new cars for a while. And this is more a function of loss in purchasing power than about new (and expensive) safety features.
People in such conditions do what they have done for a long time - buy used cars. #firstworldproblem.
You should never spend more than 20% of you annual income on a car.
Where did that rule come from?
Since forever. It's an old personal finance rule regarding how much of your income should go to credit, or home mortgages or car loans. It has to do with your liquidity in cases of emergencies and such.
You should never spend more than 20% of you annual income on a car. Median income in the US is ~$50,000 per year. So half the US population can only afford a are of $10,000. Good luck finding a new car at that price.
Which is why people go for used cars, as they always have done. This are news from yesterday. New cars have never been meant for people at or below the median income. They haven't been for a long time. Hell, I alone make 6 figures, and I've only bought a new car once, and that was for my wife and kids (while I still drive the 1998 Honda Civic my wife bought as new). So in our household, we have only bought a new car twice in more than two decades.
And this is not uncommon.
So what makes people think buying a new car is a "new" concern for median-income households? Yes, buying used cars can suck, but that is an indictment in our public transportation systems and not so much on a median-income household purchasing power.
All of those things can be provided, by proper IT.
Remote Monitoring - Virtual Desktop Infrastructure. Only systems inside the firewall can use the CCTV system, and VDI provides a way into the inside of the firewall. The CCTV system is on a non-routable VLAN that traffic cannot leave the premises. No hacking,no DDOS no nothing.
Centralized Monitoring - VLANs and VPNs. By setting up proper VPNs and VLANs, you can properly isolate systems from the outside, while providing the same level of service (perhaps even better service) for properly maintaining a single central monitoring service. The issue here is that in order to do this, you have to have an IT dept that can articulate why it needs to isolate networks from each other properly.
Ethernet Present - Yup, and probably the swiching/routing needed to properly VLAN and VPN the whole thing so that you can use existing infrastructure to isolate traffic from each other on the same equipment. Cheap ass networking gear excepted.
Good IT is expensive, bad IT is costly.
Well, I'm going to lose the mod points I provided, but what the heck.
The type of customer these products are targeted for - small businesses or homes - they do not have proper IT. Now, it is not a fault of these type of customers (to a degree). It is more the manufacturer's faults for not designing products that are *obviously* aimed that does not have dedicated/proper IT.
It should not be impossible to provide a COTS, drop-in CCTV solution that only connects from the cameras to the DVR and to pair the DVR to whatever device the customer wants to use for monitoring, with all other type of network access (local and public) restricted (expect maybe a way to "dial home" for updates.
Such a thing would never be 100% impervious to attacks, but it would be far safer than the current alternatives which are the evil cousins of open smtp relays.
You hypocrites crack me up; are you telling me you won't reinvest in training one of your own to take the job as opposed to letting in a foreigner?
Oh please, it's not like folks here are lining up demanding training. For every one that bitchs about not finding a job and wanting their low-value-added jobs coming back from China in their little, economically depressed towns, there is one going to North Dakota and Texas and finding a job there, or learning a new trade and shit.
You are demanding that we train our local talent, and that's great. But that requires some time of government intervention to either provide that type of education, or to coerce private companies to do so. And that has always been a bridge too far to the general voter.
I remember a time when companies would train their employees, specially their white collar ones. No more. And that's purely a function of how people vote. You do not have a right to demand companies to train you (or not replace you with a possibly better educated worker across the world) unless you have a government that gives you that right.
This country has been content producing HS students who can't add fractions and voting for a government-hands-off approach to adult/vocational training (as it exists in Germany or Japan). And then you act surprised when you reap the fruits of that labor and Chinese workers take your job making cheap trinkets.
[...] are you telling me you won't reinvest in training one of your own to take the job as opposed to letting in a foreigner?
That's what Germany did. In addition to relaxing the stance on immigration, they did reform the education system, introduced subsidies for reeducation of employed, and free education for unemployed.
Which is something I will always applaud Germany for. And this is something the UK and us on the other side of the Atlantic should do (but won't because stupidity is our forte.)
It would be pretty undemocratic and stupid for the new government NOT to invoke Article 50 after such a turnout, even with buyer's remorse. From where I am (on the other side of the Atlantic), I'd say leaving the EU was pretty stupid, but I also firmly believe that nations must reap what they sow. The EU is already getting ready to deal with a UK exit, whether Article 50 is invoked right now or not. They vote for it, let them eat crow (and I would say the same about us in the land of eagles and freedom fries if we vote for Tangerine man.)
Chrome Bug Makes It Easy To Download Movies From Netflix and Amazon Prime
When it comes to Amazon Prime, I like this bug... err feature. Owning content that can't download?
I was a sucker when I bought a few things that I could have gotten on DVD. Never again.
Slashdot Mirror
There is no problem that one can solve with object oriented design that cannot also be solved with structs and function pointers.
There are some problems that can be solved with structs and function pointers which cannot be solved with object oriented design (without duplicate code).
Concrete, non-trivial examples please.
RAII is only useful for pointers. Dtors cannot throw any exceptions, that means that RAII is useless for anything else. See my post above.
Dafuq? Are you serious?
A GC is a more elegant solution for memory management.
There are entire classes of problems for which a GC is not acceptable, not just in the embedded realm, but in the application/system/large memory realms. There are reasons why the Java VM provides for direct byte buffers or why .NET allows you to jump from managed code to un-managed code.
I'm a recurring canceler on Netflix... as a Canadian with half of the content, I run out of content after a while and simply cancel my subscription to renew it a few months later and have new stuff pup up that interests me.
That's the cool thing with Netflix (and Hulu). We can cancel and re-join at any time without paying penalties. Try that with cable :/
While Netflix left me wanting when they originally split streaming off as a separate service, I canceled the DVD portion in favor of keeping the on-demand streaming. They don't have EVERYTHING in their catalog, but compared to Hulu and Amazon Prime I am more likely to find what I want to watch in Netflix. That is true both for original content and regular content. Hulu I got solely to get 11.22.63 since I liked the book. After that I didn't have a reason to keep it based on the few shows I had a vague interest in. Amazon Prime I kept primarily because of its connection to Prime Shipping, but I have watched Prime few enough times that I'm ready to cancel it next month. Maybe it's just because I'm so accustomed to Netflix, but I feel I can find something to watch on Netflix that I will enjoy nearly 100% of the time. And with Prime and others I just don't have that hit rate. Could even be that Netflix UI is just better for me that the others.
Interesting. I'm about to cancel Amazon Prime because 1) I don't use shipping that much, and 2) most of the shows I watch are already on Netflix and Hulu. As for these two, sometimes I want to cancel one or the other, depending what's on the show.
I'm keeping Hulu because it carries the latest Naruto Shippuden episodes (yeah, I'm a fan) and The Daily Show, my kids watch Sailor Moon and Dragon-Ball and my wife watches Dance Moms and Modern Family. Plus Hulu has a ton of really good old movies from the Criterion Edition (we are old-movie buffs.)
Netflix on the other hand, I've been close to cancelling it because it doesn't really update their movie catalogs or TV shows that often. What is keeping me with Netflix are its originals (Peaky Blinders, Marco Polo, Narcos) plus "Jane the Virgin".
I think for binge-TV-watchers like us, two streaming services are always needed. For us, Hulu and Netflix fit the bill.
And if I see any ads, I'm canceling. Why would I pay for service with ads? Makes no damn sense.
And if you cancel, where do you go? Other than Hulu and Amazon Video, there is not much else that can compete in terms of content for a dirt-cheap price.
Even with adds, it would still be cheaper than cable. A combination of netflix with some other streaming services and an internet connection is the closest thing we have to a-la-carte cable.
So until cable companies decide to provide a-la-carte plans that people can cancel at any times without penalties, I'd be willing to see adds on Netflix. I hate adds, but I cannot be so dogmatic when it comes to what I'm getting for every buck I pay.
Hey google - why don't you train 2 million Americans?
Google is a multinational company, and most of their profits come from outside the USA. To add, India and China are the two major areas of growth into the future. The USA is that faded boxer who doesn't have it any more but still hasn't come to grips that he is washed up.
The answer to your question is one you will not like but is no less true even if you don't: They are training Indians instead of Americans because India is the future, and America is the past.
India is the future? Tell me something, has the "future" managed to put a fucking toilet above a cell phone on the priority list yet?
Based on this latest job push, I'm guessing not. Good to know our "future" has its priorities in order.
Unfortunately India is plagued by the same problems that plague Brazil. Wealth accumulated on top, an educated middle class, and then swats and swats and more swats of people living in filth and misery. Both countries excel at not making their poor's well being a priority. **
** With that said, I think India will fix that before Brazil. India is improving little by little. Brazil keeps sliding in the opposite direction.
Hey google - why don't you train 2 million Americans?
Google is a multinational company, and most of their profits come from outside the USA. To add, India and China are the two major areas of growth into the future. The USA is that faded boxer who doesn't have it any more but still hasn't come to grips that he is washed up.
The answer to your question is one you will not like but is no less true even if you don't: They are training Indians instead of Americans because India is the future, and America is the past.
The better question to ask yourselves as Americans is where you went wrong. You used to be on top of the world. Now China makes everything you used to make, and India is eating into the programming. You used to make the best cars, now that's Japan.
Look to yourselves before you blame others. It is harder to do, but only by accepting where you went wrong do you have a hope of recovering from a tail spin.
India, like Brazil, has too many systemic problems right now for the world to consider it "the future". China, yes. And to a lesser extend Eastern Europe and some parts of Latin America (in particular Mexico which has been getting its shit together for the last 2 decades.)
Heck, even Japan with its aging population will be part of the future for a long time before India. Why, robotics, efficiency and personal/business savings up to the wazoo.
For India to overcome China (and other regions of the world) as "the future", it needs to overcome them in terms of manufacturing (both low-value added and high-value added manufacturing.)
I find that extremely unlikely (and no, don't construe my words as a bashing of India, I want India to progress and for her citizens to live better.) But it is simply not set up to be a global manufacturing hub (unlike China or Indonesia or Malaysia or Japan or Mexico.)
In the services area, however, I think India is well posed, and it will do well hitting that.
Replying to myself, I suggest people follow this: https://www.thelayoff.com/seag...
Seagate Fires 6,500, Or 14% of Workforce, Stock Soars
Of those laid off, how much of that was due to improved productivity?
What were the roles?
Were these engineers?
Or workers whose jobs could be automated?
Or how many of these were "dead weight" or in redundant functions?
Or where people blindly laid off due to getting paid more than a certain salary cap, regardless of function just to make the spreadsheets pretty? *** (I've seen this.)
Some layoffs are totally destructive. Others are completely justified. I've seen both types. The article doesn't tell me shit about it.
. Startup culture isn't for the young either; you really have to have the fraternity/sorority member personality type to work there so as people age they're less likely to trade salary for beer pong or free dinner.
I disagree that it is not for the young. They want people who can work 70-100 hours a week. Those people do not have families. If they do their wife does not work. Or if they both work then they have no kids. Startups should have no problem with a 30% turnover rate or higher.
They want people who appear busy for 70-100 hours a week. 70-100 hours a week implies between 10 to 16 hours a day (depending of whether you work 6 days or work without taking a day off.) Then add 1-2 hours of commute, we are talking 10 to 18 hours a day.
No matter how young you are, it is physically impossible to do *actual* work at that rate for more than a few weeks (or months at most.) From experience (and purely anecdotal), the physical and psychological upper limit seems to be 60 hours a week (less if you count commute times). And that translate to 10 to 12 hours every day (whether you work 5 or 6 days) sustained to no more than 8-12 months.
After that, attrition inevitably follows. I've seen companies having attrition rates of 50-60% (and worse) when they force their workforce through that type of grind for more than a year or two.
Unless you are building the next rocket to Mars or some other incredibly cool shit, you are an idiot if you subject yourself to that. You do not need that to gain valuable experience.
Tech Workers Think Silicon Valley and Startups Are Losing Their Luster
The cost of living in the valley is so ridiculous now that one has to make double (or more) than what you could make in Dallas or Denver just to afford a townhome in a decent zip code. This has always been a serious consideration for couples, and it has started to be for the new wave of college grads who are more financially conscious than their predecessors (yes, there is a measurable change in spending habits among younger people.)
Then there is the snobbery of the hiring process. By God, the snobbery by which some companies use when sending rejection letters and e-mails. I've never seen anything like this anywhere.
And last but not least, the ageism. It exists, it is an open secret. So what kind of outcome can a region expect to get when it methodically culls senior people that have accumulated experience on the trenches? Sooner or later that kind of shit is going to come back.
Put all of these three on top of the definition of innovation as "inventing the next mobile/social shit app". Innovation occurs predominantly where software meets hardware to provide unique solutions and opportunities. That is not the valley's bread and butter anymore. Too many unicorns.
So yeah, loss of luster is not surprising. At. All.
You left out the third option. 1,100 jobs where saved by automation. Hostess went out of business and several other companies bought up the rights to the products that Hostess made.
This is the correct option.
How Often Do You Switch Programming Languages?
This is the wrong focus. A more accurate (and interesting question) is how often do you switch stacks or architectures?
Embedded programming using, say, the Wind River toolchain can be quite distinct from doing Linux systems programming using the GNU tool chain. And these two are quite distinct from Windows systems programming using (duh) Windows toolchains. And these are all done (typically) in C. Similarly, C++ programming on Linux will be different from Windows programming, and more so when we start adding stacks and libraries like Qt when the objective is to use Qt's higher level capabilities instead of OS portability (which sometimes is not a bad thing.)
Same with Java. Right now I'm doing core Java, dealing directly with NIO, threading, etc, all outside the safety net of a container. For better or worse, our work does not use OSGI (and that would have been a very distinct type of work if such an architectural choice have been made.)
That is quite distinct from EE development. Working specifically to stay within a web container can be quite distinct from working on something *that exploits* every nook and cranny off an EE container. Then you can throw Spring (and/or OSGI) into the mix which can cut through every one of these dimensions in the Java world. And let's not get started when you move into a completely different paradigm (say vert.x)
In Javascript, this is even more radical since modern Javascript (for better or worse) has become some sort of meta-language that sits below a higher-level framework or architecture - be it jquery or dojo or extjs in the recent past, AngularJX, backbone, React and what now in our current times. Each of these poses a learning (and possibly architectural) challenge.
One sees the same with Python and Ruby to a lesser extend.
The .NET world provides more continuity in how things are build and a developer can, with relative ease, jump back and forth from C# to Managed C++.
And in the application space you have an additional dimension - application data storage. SQL or NoSQL. That will change how you design and code.
So the question that is more relevant is, how often do you change platforms and stacks. Because you could very well be using the same language for 6 years and have experienced a significant development and architectural shift.
It's hard to believe that testing 32-bit really doubles the testing effort, but whatever.
One of the biggest parts of testing consists of keeping test systems up and running and ready to run the tests, every retired system is less labor.
Virtual machines need updating and maintenance, too, they are not "free" by any stretch, again labor is saved when they are retired.
Often tests are not "totally bulletproof", with failures from system problems or incorrect setup, much labor is involved here in investigation and/or mitigation, again directly related to the number of systems to test.
This and exactly this. Another thing people forget is that testing not only cover fresh installations, but also upgrades. So the testing permutations begin to fly off the handle with the number of supported platforms combined with the number of upgrade paths an organization needs to test.
That shit is neither cheap, nor easy.
could turn a 300-mile trip that would normally take 3.5 hours flying into a breezy 28-minute ride
Japan's SCMaglev clocked a speed record of 375 mph in 2015. So that same 3.5 hour flight can be turned into a 1 hour trip. Yes, 28 minutes is better, but economically speaking, a nation could take the 1 hour trip instead with proven technology that already exists.
Sooner or later, maglev trains will achieve speeds over 500 mph. They'll never break the sound barrier (which is Hyperloop's promise), but, economically, they do not have to. One cannot justify the expense, at this time, to create a hyperloop to make a mere 300-mile trip in 30 minutes.
This is also why I wonder the economic sense of building one between Los Angeles and the Valley instead of building a "bullet" train that could make the trip in 1 hour. A hyperloop makes sense for transcontinental travel, to connect distances of over 1,000 miles, and in some cases, to connect sufficiently distant end-points in a country (say, making a 2k-mile trip connecting Kyushu with Hokkaido.)
Don't get me wrong. I like the idea of a hyperloop. We should push our technological capabilities. But we should also see what things make economic sense. I do not see the Hyperloop making sense for a 300-mile trip.
^^^ Winz the internet!
If you have a cell phone, why would having VOIP connection stop you from calling?
The premise was that calling is not optimal and email or text chat should be preferred. Sure, with a cell phone you could call tech support, but if the problem can be resolved more quickly by email or text chat, you could use the internet connection of the phone for that purpose.
From my experience, calling tech support with a cell phone is even more of a nightmare than calling from a a landline due to the inferior quality.
But that would be a problem with the signal and only if you live in the boondocks.
The crux of the problem is in the process of getting your call received and processed, talking to the rep on the other line, etc. That is an invariant whether you call from a land line or a cell phone (who in metro areas have a land line nowadays?) Signal quality takes the back seat on this.
That's in part because new cars are loaded with helpful but expensive safety features like collision-avoidance systems.
Wrong factors. Median income people haven't been able to afford new cars for a while. And this is more a function of loss in purchasing power than about new (and expensive) safety features.
People in such conditions do what they have done for a long time - buy used cars. #firstworldproblem.
You should never spend more than 20% of you annual income on a car.
Where did that rule come from?
Since forever. It's an old personal finance rule regarding how much of your income should go to credit, or home mortgages or car loans. It has to do with your liquidity in cases of emergencies and such.
You should never spend more than 20% of you annual income on a car. Median income in the US is ~$50,000 per year. So half the US population can only afford a are of $10,000. Good luck finding a new car at that price.
Which is why people go for used cars, as they always have done. This are news from yesterday. New cars have never been meant for people at or below the median income. They haven't been for a long time. Hell, I alone make 6 figures, and I've only bought a new car once, and that was for my wife and kids (while I still drive the 1998 Honda Civic my wife bought as new). So in our household, we have only bought a new car twice in more than two decades.
And this is not uncommon.
So what makes people think buying a new car is a "new" concern for median-income households? Yes, buying used cars can suck, but that is an indictment in our public transportation systems and not so much on a median-income household purchasing power.
This is truly a first world problem.
All of those things can be provided, by proper IT.
Remote Monitoring - Virtual Desktop Infrastructure. Only systems inside the firewall can use the CCTV system, and VDI provides a way into the inside of the firewall. The CCTV system is on a non-routable VLAN that traffic cannot leave the premises. No hacking ,no DDOS no nothing.
Centralized Monitoring - VLANs and VPNs. By setting up proper VPNs and VLANs, you can properly isolate systems from the outside, while providing the same level of service (perhaps even better service) for properly maintaining a single central monitoring service. The issue here is that in order to do this, you have to have an IT dept that can articulate why it needs to isolate networks from each other properly.
Ethernet Present - Yup, and probably the swiching/routing needed to properly VLAN and VPN the whole thing so that you can use existing infrastructure to isolate traffic from each other on the same equipment. Cheap ass networking gear excepted.
Good IT is expensive, bad IT is costly.
Well, I'm going to lose the mod points I provided, but what the heck.
The type of customer these products are targeted for - small businesses or homes - they do not have proper IT. Now, it is not a fault of these type of customers (to a degree). It is more the manufacturer's faults for not designing products that are *obviously* aimed that does not have dedicated/proper IT.
It should not be impossible to provide a COTS, drop-in CCTV solution that only connects from the cameras to the DVR and to pair the DVR to whatever device the customer wants to use for monitoring, with all other type of network access (local and public) restricted (expect maybe a way to "dial home" for updates.
Such a thing would never be 100% impervious to attacks, but it would be far safer than the current alternatives which are the evil cousins of open smtp relays.
You hypocrites crack me up; are you telling me you won't reinvest in training one of your own to take the job as opposed to letting in a foreigner?
Oh please, it's not like folks here are lining up demanding training. For every one that bitchs about not finding a job and wanting their low-value-added jobs coming back from China in their little, economically depressed towns, there is one going to North Dakota and Texas and finding a job there, or learning a new trade and shit.
You are demanding that we train our local talent, and that's great. But that requires some time of government intervention to either provide that type of education, or to coerce private companies to do so. And that has always been a bridge too far to the general voter.
I remember a time when companies would train their employees, specially their white collar ones. No more. And that's purely a function of how people vote. You do not have a right to demand companies to train you (or not replace you with a possibly better educated worker across the world) unless you have a government that gives you that right.
This country has been content producing HS students who can't add fractions and voting for a government-hands-off approach to adult/vocational training (as it exists in Germany or Japan). And then you act surprised when you reap the fruits of that labor and Chinese workers take your job making cheap trinkets.
[...] are you telling me you won't reinvest in training one of your own to take the job as opposed to letting in a foreigner?
That's what Germany did. In addition to relaxing the stance on immigration, they did reform the education system, introduced subsidies for reeducation of employed, and free education for unemployed.
Which is something I will always applaud Germany for. And this is something the UK and us on the other side of the Atlantic should do (but won't because stupidity is our forte.)
It would be pretty undemocratic and stupid for the new government NOT to invoke Article 50 after such a turnout, even with buyer's remorse. From where I am (on the other side of the Atlantic), I'd say leaving the EU was pretty stupid, but I also firmly believe that nations must reap what they sow. The EU is already getting ready to deal with a UK exit, whether Article 50 is invoked right now or not. They vote for it, let them eat crow (and I would say the same about us in the land of eagles and freedom fries if we vote for Tangerine man.)
Chrome Bug Makes It Easy To Download Movies From Netflix and Amazon Prime
When it comes to Amazon Prime, I like this bug... err feature. Owning content that can't download? I was a sucker when I bought a few things that I could have gotten on DVD. Never again.