It's the same network. My point is that, when people text, they are concise, so the amount of network used per communication attempt is lower. More people can tell their loved ones that they are OK per unit of time if they do so with text. If they call, they're going to talk for three/ten/twenty minutes. When they text, they use the network for a faction of a second.
It costs far less to send a text message than to make a 30 second phone call.
Actually the marginal cost of both the call and the text is effectively zero. They both rely on a high fixed cost infrastructure, and have, for all intents and purposes, zero cost for an individual call or text. So I'm curious why we're not talking about the rapacious costs of calling plans as well.
Actually, they wanted you to text during hurricanes and similar emergencies because it uses less bandwidth than a 20 minute phone call, so that more communication could get through. Switch infrastructures have finite capacity, and people tend to talk longer and more frequently in emergency situations, overwhelming the switches. Texting forces brevity. There isn't a huge amount of excess capacity in the networks, and in emergency situations it gets swamped.
These are the kind of assholes who troll around the web looking for any discussion in which to insert their derogatory "I'm smarter than you - it's so obvious!" attitude while ignoring the issue at hand.
Pot, meet kettle.
Those people you are deriding are correct. The companies are charging the price because they can, and people will pay. That is how a market works. Texting is not a necessity, and no-one is holding a gun to anyone's head to buy text service plans. If the price was higher than people's perceived value for the plan, people wouldn't buy it. And yet, they do. They are buying an optional good, and paying willingly.
Whether it is ethically justified is a separate question. The facts seem to be that it's currently legal, so, why wouldn't the telcos charge what the market would bear? Do you work for less than your employer is willing to pay you? Ever asked for a raise, or taken another job because it paid more?
While the marginal cost of sending a text is minimal, the fixed cost of the network is non-trivial. You could look at this as similar to power from a hydroelectric dam. The marginal cost of a kilowatt is nil, but it cost a lot of money to get the capacity into place. So it is with the phone companies.
If you look at the financial statements of the phone companies, or of the oil companies you seem to want to pick on, you'll see that their net profit, as a percent of sales, aren't really that impressive. Verizon makes 6% profit on sales, and a 5% return on their assets. They'd make more money by selling the company and buying a mutual fund. It's not immediately clear from the numbers that they are being 'unfair' in their pricing, as the rate of return isn't that impressive. Google, by comparison, has a 24% profit margin on sales, and a 14.5% return on assets. If profit is evil, then Google is more evil than the telcos.
Probably the most telling picture of the rapacious success can be seen here . If the telcos are such money machines, why is the stock price essentially flat over 5 years?
Now, I'm not going to defend the business practices of the telcos, and I'd like to see them pay rent on the public spectrum. They have done a fine job of manipulating the US congress, and I resent it. But you are overstating/misrepresenting how well they do, and you don't seem to understand the essence of market economics. Wireless service is not forced on anyone, and if you feel that the price is exhorbitant, well, don't pay it. But please refrain from preaching and condescending to the rest of us about what idiots we are for using it.
And yeah, I read your article. I don't think you understand much about economics, finance, or telcos. I wouldn't be bragging about it too much, if I were you.
Yeah, but people buy tons of bottled water in Seattle, too, where the tap water is tasty, and is indeed bottled and resold without additional treatment by such companies as Talking Rain. We have the best tasting water in the country, and my wife still buys cases of bottled water. Go figure.
Technically speaking, there isn't any law that says they have to maximize profits, it's a fiduciary responsibility, for which they could incur civil liability, and/or lose their jobs. They wouldn't be breaking any law to take action in favor of the users/customers/third parties, but the Board of Directors might choose to end their employment for doing so. Or not.
Top level management has a lot of freedom in acting in the interests of the company. The main control is the Board of Directors removing them from management if the board disagrees with the approach.
Data mining is statistically based. The more information that's available to mine, the more accurate the results will be.
A minor quibble. I do data mining for a living. With most data sets, we end up sampling them down, because more data ramps up processing time faster than it improves accuracy. With most problems, more data doesn't improve accuracy measureably, once you've reached a certain critical mass size in the dataset. Simplistically, you don't need to flip the coin a billion times to figure out that it comes up heads 50% of the time.
It's a rare problem that we use more than 100,000 records for. They exist, but they're rare.
Linux will last until UNIX based systems don't work for the needs of the day.
I suspect that Linux can last as long as resources are adequately described by the notion of a file, and work against resources is done by something that can be adequately represented as a process. My simple mind can't conceive of where these two situations cease to be true, so I don't see why you would need to move away from a Linux-ish model.
What we may well see is replacement of the desktop/browser metaphor, but any new metaphor is likely to representable with files and processes, just as a command line, then an 80x24 screen, and now our GUI desktops have been. The unix file and process model has been essentially the same at the conceptual level throughout this progression so far, so I don't see why a new innovation would be unimplementable in Linux (or Windows, for that matter).
I'm just measuring by the job postings I see. The majority of the positions I see are for functionality that is delivered via a server through a browser or a device. Those apps are mostly delivered through PHP/Ruby/Python/Java+X/.Net. While I agree that there is still a bunch of core application development out there, my point is just that I think that is a minority of the function points being delivered now.
I agree massively, and would add, C++ is by far the most difficult programming environment I have ever tried to learn. The surface area is large, for many of the reasons you state, and it's brittle as a result. While I understand the reasons that it's adherents appreciate it, there is a reason it occupies an increasing small percentage of the development space. It makes sense for system programming and numerically intensive tasks because of it's undeniable performance advantage, but you have to be a masochist to want to use it for other work.
I'm feeling very old, because I know you didn't make this up.
My dad once almost lost his job because he and his co-worker blew a year's computing budget on an infinite loop one night. It was on a Boeing computer services machine in about 1972, IIRC.
but better suited to tasks where there's a LOT of input/output data,
They're still in heavy use, for just such jobs. Your phone and power bills are probably mainframe generated, but your stock brokerage statement probably isn't. Your credit card bill probably is run on a MF (my company probably does it), but most of the transactions on it probably came from Unix/Windows sources. The MF rules in sequential processing, though mostly from inertia, while Unix/Linux/Windows Server is taking over the DB side.
Mod parent up - he's on the mark. There's a lot of stuff out there for which the source is gone, but it still does the job. Witness the State of California's payroll system we were discussing a bit back.
Aircraft don't fly like bees, don't navigate like bees and don't have the same goals as bees.
Depends on how you look at it. Bees leave a site, navigate through an airspace with other flying critters in it, find a destination, perform a manuever, return to their departure site, land, while avoiding and responding to hostile aggressors. That's a reasonable macro model for a predator drone.
If you're a software manager who doesn't code, then I don't think it's any of your business what an application is coded in.
It's absolutely the manager's business what tools his team uses. The choice of tools affects performance, integration, team composition and productivity, and ability for the application to be supported and extended economically in the future. If the manager isn't thinking about and managing towards these goals, he/she is failing at their primary job.
Anything down by enterprises like banks or fortune 2000 companies, or that is large enough that shared web hosting is irrelevant. The LAMP stack, is popular and useful for smaller endeavors, but most people going large are either on the.Net or Java/JSP/Oracle stacks.
Agreed. I resist using them, and won't let anyone on my team use them for code that gets installed at clients. Our client code looks like a well structured C program, as much as we can make it so.
That said, I like Perl for what we use it for, which is installation glue code, because it's generally easy for client staff to pick up. I like Python myself, but it's harder for marginal IT staff to pick up. Python and Ruby also tend to offend the client IT teams standards folks, who aren't quite sure about this new-fangled stuff. They'd prefer C# or Java, which are completely inappropriate for what we need to do.
I don't don't of Linux distros out of the box that scale to 100+ processors well
Isn't the usual mode of scaling Linux based systems oriented towards bunches of generic boxes with a small number of processors? I don't have broad experience here, but the sites I have worked with don't want to spend the money it costs for large SMP boxes, when, by architecting the application correctly you can scale it across large numbers of commodity boxes more economically, and gain resiliance in the bargain.
I have nothing against Solaris, but neither does it do anything for me. Our company has done well with openSUSE and CentOS/Fedora. As I said above, we haven't had any stability issues, and nothing that would force us to go to Sun. Our load is compute and file system, though, so we don't need a huge Oracle server.
Slashdot Mirror
It's the same network. My point is that, when people text, they are concise, so the amount of network used per communication attempt is lower. More people can tell their loved ones that they are OK per unit of time if they do so with text. If they call, they're going to talk for three/ten/twenty minutes. When they text, they use the network for a faction of a second.
It costs far less to send a text message than to make a 30 second phone call.
Actually the marginal cost of both the call and the text is effectively zero. They both rely on a high fixed cost infrastructure, and have, for all intents and purposes, zero cost for an individual call or text. So I'm curious why we're not talking about the rapacious costs of calling plans as well.
Actually, they wanted you to text during hurricanes and similar emergencies because it uses less bandwidth than a 20 minute phone call, so that more communication could get through. Switch infrastructures have finite capacity, and people tend to talk longer and more frequently in emergency situations, overwhelming the switches. Texting forces brevity. There isn't a huge amount of excess capacity in the networks, and in emergency situations it gets swamped.
These are the kind of assholes who troll around the web looking for any discussion in which to insert their derogatory "I'm smarter than you - it's so obvious!" attitude while ignoring the issue at hand.
Pot, meet kettle.
Those people you are deriding are correct. The companies are charging the price because they can, and people will pay. That is how a market works. Texting is not a necessity, and no-one is holding a gun to anyone's head to buy text service plans. If the price was higher than people's perceived value for the plan, people wouldn't buy it. And yet, they do. They are buying an optional good, and paying willingly.
Whether it is ethically justified is a separate question. The facts seem to be that it's currently legal, so, why wouldn't the telcos charge what the market would bear? Do you work for less than your employer is willing to pay you? Ever asked for a raise, or taken another job because it paid more?
While the marginal cost of sending a text is minimal, the fixed cost of the network is non-trivial. You could look at this as similar to power from a hydroelectric dam. The marginal cost of a kilowatt is nil, but it cost a lot of money to get the capacity into place. So it is with the phone companies.
If you look at the financial statements of the phone companies, or of the oil companies you seem to want to pick on, you'll see that their net profit, as a percent of sales, aren't really that impressive. Verizon makes 6% profit on sales, and a 5% return on their assets. They'd make more money by selling the company and buying a mutual fund. It's not immediately clear from the numbers that they are being 'unfair' in their pricing, as the rate of return isn't that impressive. Google, by comparison, has a 24% profit margin on sales, and a 14.5% return on assets. If profit is evil, then Google is more evil than the telcos.
Probably the most telling picture of the rapacious success can be seen here . If the telcos are such money machines, why is the stock price essentially flat over 5 years?
Now, I'm not going to defend the business practices of the telcos, and I'd like to see them pay rent on the public spectrum. They have done a fine job of manipulating the US congress, and I resent it. But you are overstating/misrepresenting how well they do, and you don't seem to understand the essence of market economics. Wireless service is not forced on anyone, and if you feel that the price is exhorbitant, well, don't pay it. But please refrain from preaching and condescending to the rest of us about what idiots we are for using it.
And yeah, I read your article. I don't think you understand much about economics, finance, or telcos. I wouldn't be bragging about it too much, if I were you.
Yeah, but people buy tons of bottled water in Seattle, too, where the tap water is tasty, and is indeed bottled and resold without additional treatment by such companies as Talking Rain. We have the best tasting water in the country, and my wife still buys cases of bottled water. Go figure.
At least he used a complete, grammatically correct sentence. ;-)
Oh, my kingdom for a verb...
Technically speaking, there isn't any law that says they have to maximize profits, it's a fiduciary responsibility, for which they could incur civil liability, and/or lose their jobs. They wouldn't be breaking any law to take action in favor of the users/customers/third parties, but the Board of Directors might choose to end their employment for doing so. Or not.
Top level management has a lot of freedom in acting in the interests of the company. The main control is the Board of Directors removing them from management if the board disagrees with the approach.
Data mining is statistically based. The more information that's available to mine, the more accurate the results will be.
A minor quibble. I do data mining for a living. With most data sets, we end up sampling them down, because more data ramps up processing time faster than it improves accuracy. With most problems, more data doesn't improve accuracy measureably, once you've reached a certain critical mass size in the dataset. Simplistically, you don't need to flip the coin a billion times to figure out that it comes up heads 50% of the time.
It's a rare problem that we use more than 100,000 records for. They exist, but they're rare.
Linux will last until UNIX based systems don't work for the needs of the day.
I suspect that Linux can last as long as resources are adequately described by the notion of a file, and work against resources is done by something that can be adequately represented as a process. My simple mind can't conceive of where these two situations cease to be true, so I don't see why you would need to move away from a Linux-ish model.
What we may well see is replacement of the desktop/browser metaphor, but any new metaphor is likely to representable with files and processes, just as a command line, then an 80x24 screen, and now our GUI desktops have been. The unix file and process model has been essentially the same at the conceptual level throughout this progression so far, so I don't see why a new innovation would be unimplementable in Linux (or Windows, for that matter).
I'm just measuring by the job postings I see. The majority of the positions I see are for functionality that is delivered via a server through a browser or a device. Those apps are mostly delivered through PHP/Ruby/Python/Java+X/.Net. While I agree that there is still a bunch of core application development out there, my point is just that I think that is a minority of the function points being delivered now.
and in actual applications programming C++ is still the leader of the pack.
But in terms of what people are doing, how much of that these days is applications programming, in the sense that you are using the term?
I think that the weight of development these days has moved away from things that are delivered in setup.exe's.
I agree massively, and would add, C++ is by far the most difficult programming environment I have ever tried to learn. The surface area is large, for many of the reasons you state, and it's brittle as a result. While I understand the reasons that it's adherents appreciate it, there is a reason it occupies an increasing small percentage of the development space. It makes sense for system programming and numerically intensive tasks because of it's undeniable performance advantage, but you have to be a masochist to want to use it for other work.
I'm feeling very old, because I know you didn't make this up.
My dad once almost lost his job because he and his co-worker blew a year's computing budget on an infinite loop one night. It was on a Boeing computer services machine in about 1972, IIRC.
but better suited to tasks where there's a LOT of input/output data,
They're still in heavy use, for just such jobs. Your phone and power bills are probably mainframe generated, but your stock brokerage statement probably isn't. Your credit card bill probably is run on a MF (my company probably does it), but most of the transactions on it probably came from Unix/Windows sources. The MF rules in sequential processing, though mostly from inertia, while Unix/Linux/Windows Server is taking over the DB side.
Mod parent up - he's on the mark. There's a lot of stuff out there for which the source is gone, but it still does the job. Witness the State of California's payroll system we were discussing a bit back.
The last time I was stung by a wasp, I thought that's what had happened.
I have. They get fooled exactly once. They can also count hunters, and easily recognize guns.
Aircraft don't fly like bees, don't navigate like bees and don't have the same goals as bees.
Depends on how you look at it. Bees leave a site, navigate through an airspace with other flying critters in it, find a destination, perform a manuever, return to their departure site, land, while avoiding and responding to hostile aggressors. That's a reasonable macro model for a predator drone.
Yup. I had a dog once that figured out how to open doors, all on her own. She figured out both door knobs and sliding glass doors.
While not quite the same thing, I watched a crow roll an almost empty latte' cup around on a hill so that it could sip the last few drops out.
Only in Seattle do we have caffiene addicted crows.
If you're a software manager who doesn't code, then I don't think it's any of your business what an application is coded in.
It's absolutely the manager's business what tools his team uses. The choice of tools affects performance, integration, team composition and productivity, and ability for the application to be supported and extended economically in the future. If the manager isn't thinking about and managing towards these goals, he/she is failing at their primary job.
In what context?
Anything down by enterprises like banks or fortune 2000 companies, or that is large enough that shared web hosting is irrelevant. The LAMP stack, is popular and useful for smaller endeavors, but most people going large are either on the .Net or Java/JSP/Oracle stacks.
(and ENCOURAGES it with "implied" scalars)
Agreed. I resist using them, and won't let anyone on my team use them for code that gets installed at clients. Our client code looks like a well structured C program, as much as we can make it so.
That said, I like Perl for what we use it for, which is installation glue code, because it's generally easy for client staff to pick up. I like Python myself, but it's harder for marginal IT staff to pick up. Python and Ruby also tend to offend the client IT teams standards folks, who aren't quite sure about this new-fangled stuff. They'd prefer C# or Java, which are completely inappropriate for what we need to do.
Ask any *real* Unix admin who uses both and more than likely they will say Linux is great for small jobs but Solaris is king for anything else.
Which is why Google runs on Solaris, right?
Oh, wait...
I don't don't of Linux distros out of the box that scale to 100+ processors well
Isn't the usual mode of scaling Linux based systems oriented towards bunches of generic boxes with a small number of processors? I don't have broad experience here, but the sites I have worked with don't want to spend the money it costs for large SMP boxes, when, by architecting the application correctly you can scale it across large numbers of commodity boxes more economically, and gain resiliance in the bargain.
I have nothing against Solaris, but neither does it do anything for me. Our company has done well with openSUSE and CentOS/Fedora. As I said above, we haven't had any stability issues, and nothing that would force us to go to Sun. Our load is compute and file system, though, so we don't need a huge Oracle server.