Yes. If anything, Comcast should be paying Netflix's ISP because it's their customers requesting the content.
In truth though, the end customer and Netflix are both using "the internet" so their respective ISPs should be billing their customers appropriately and providing the service paid for. I pay Comcast to provide me with the service as advertised. In fact, I changed carriers to them to *improve* my Netflix experience so they should damn well be providing.
Yes. Java was way too heavyweight. Which is not really a problem with Java but more with the way people were using it. Which is not really their fault either since there was a gap. Which Flash filled.
I guess it should also be remembered that alongside flash was Shockwave which was also fairly heavyweight and bloomed for a while and then fell from grace. I worked with that with some non-web stuff and it was fairly nice and easy to extend with the xtra API. I never really worked with Flash but my understanding of it is that early versions were not particularly powerful but it ultimately ended up adding more features as time went on until it edged Shockwave out.
A stack trace *can* be much more helpful. But when there is a simple error, it is usually simply unnecessary. It's true that mostly it's likely an implementation issue than a fault of Java though.
Do you think statistical physics, astronomy or quantum thermodynamics would most help?
When the one line error message tells you that your SQL connection has failed because the login or password was incorrect (typically nicely also telling you the username used and the host it attempted to connect to and usually the driver used also) and the stack trace simply tells you the there was an "SQL Exception" followed by irrelevancy after irrelevancy, I'll take the message and check the password used.
But perhaps I'm being unfair. It may just be JBoss or the developer who coded it but many other languages give you useful info by default.
Java: Why explain everything you need to know about what went wrong in a one line error message when a three page stack trace can leave you totally confused instead?
But it's also possible that the other path might have resulted in much faster development - we'll never know.
True, it's impossible to be sure but one might attempt to measure it up to the growth of the internet which has, in a decade or two since it became consumer ready, brought us vast information resources at a cheap cost and has, in the process, totally buried some technologies that telcos were attempting to bring us in their half-hearted locked-in manner (Video calls, information services etc).
I don't see the issue with duplication of infrastructure anyway. It's like when people complain about multiple brands of conflakes on the supermarket shelf and the associated advertising costs. It totally fails to account for the value proposition that competition brings.
Sure, no one wants a dozen fiber optic cables strung down the street (though I wish the US would bury the cables but that's another discussion) but that's not likely to happen anyway. The issue is that if you allow competition, the maximum that can be charged is that at which point it becomes financially worthwhile for another company to come along and start stringing cable. If the government is providing a monopoly, it becomes a "my way or the highway" proposition and those companies that have the monopolies can get away with a whole bunch of crap. Just look at what is happening where Google has said "damn the expense" and started putting fiber in anyway. Prices from the incumbent ISPs have plummeted.
The ISPs would be bribing the state to raise barriers to new ISPs having access to the fiber, of course.
The more you put under the power of the state, the more opportunity for corruption.
To be sure there is a role for state regulation but when you're introducing regulations to fix problems cause by regulation, it's time to take a step back and reconsider what the fuck you're doing.
The metric system rationalizes and reforms a whole bunch of physical phenomena.
You cant do that with dates & times. They're intrinsically arbitrary and irrational. The current system is set up to account for human needs so it likely the best compromise that's possible. Even the metric system had to bend a little and have centimeters for common use.
m/d/y is by no means internationally accepted. It seems to be mostly an american thing. The UK (and I think most of Europe) uses d/m/y and I'm sure it varies around the world. I much prefer YYYY-MM-DD anyway. It squares with how we do time and is an ISO standard. m/d/y is just nonsense whichever way you slice it.
I question your assumption. Third party payment processors for $ have costs not associated with Bitcoin and something of a lock-in due to network effects (Don't take Visa? You just cut out 80% or more of your potential customers. Take Bitcoin? Use any one of an increasing number of processors).
Of course, hopefully the processors will fade with time anyway as suggested below by the AC.
Mining fees will be kept low by intense competition amongst miners. This is also an incentive for merchants to accept and use Bitcoin rather than directly converting to $.
Slashdot Mirror
Usually but not always.
As a Comcast customer, I agree.
Yes. If anything, Comcast should be paying Netflix's ISP because it's their customers requesting the content.
In truth though, the end customer and Netflix are both using "the internet" so their respective ISPs should be billing their customers appropriately and providing the service paid for. I pay Comcast to provide me with the service as advertised. In fact, I changed carriers to them to *improve* my Netflix experience so they should damn well be providing.
Yes. Java was way too heavyweight. Which is not really a problem with Java but more with the way people were using it. Which is not really their fault either since there was a gap. Which Flash filled.
I guess it should also be remembered that alongside flash was Shockwave which was also fairly heavyweight and bloomed for a while and then fell from grace. I worked with that with some non-web stuff and it was fairly nice and easy to extend with the xtra API. I never really worked with Flash but my understanding of it is that early versions were not particularly powerful but it ultimately ended up adding more features as time went on until it edged Shockwave out.
A stack trace *can* be much more helpful. But when there is a simple error, it is usually simply unnecessary. It's true that mostly it's likely an implementation issue than a fault of Java though.
Do you think statistical physics, astronomy or quantum thermodynamics would most help?
When the one line error message tells you that your SQL connection has failed because the login or password was incorrect (typically nicely also telling you the username used and the host it attempted to connect to and usually the driver used also) and the stack trace simply tells you the there was an "SQL Exception" followed by irrelevancy after irrelevancy, I'll take the message and check the password used.
But perhaps I'm being unfair. It may just be JBoss or the developer who coded it but many other languages give you useful info by default.
Actually, I have not. But compilation errors are not my problem anyway. Runtime errors are and I am not a Java developer.
Java: Why explain everything you need to know about what went wrong in a one line error message when a three page stack trace can leave you totally confused instead?
Flash on steroids most likely as it displaced Java in a lot of areas anyway.
That was my thought too. Though 300 might be generous.
Don't Read the Fine Article.
Very pointless and uninteresting.
But it's also possible that the other path might have resulted in much faster development - we'll never know.
True, it's impossible to be sure but one might attempt to measure it up to the growth of the internet which has, in a decade or two since it became consumer ready, brought us vast information resources at a cheap cost and has, in the process, totally buried some technologies that telcos were attempting to bring us in their half-hearted locked-in manner (Video calls, information services etc).
I am altering the deal. Pray I don't alter it any further.
I don't see the issue with duplication of infrastructure anyway. It's like when people complain about multiple brands of conflakes on the supermarket shelf and the associated advertising costs. It totally fails to account for the value proposition that competition brings.
Sure, no one wants a dozen fiber optic cables strung down the street (though I wish the US would bury the cables but that's another discussion) but that's not likely to happen anyway. The issue is that if you allow competition, the maximum that can be charged is that at which point it becomes financially worthwhile for another company to come along and start stringing cable. If the government is providing a monopoly, it becomes a "my way or the highway" proposition and those companies that have the monopolies can get away with a whole bunch of crap. Just look at what is happening where Google has said "damn the expense" and started putting fiber in anyway. Prices from the incumbent ISPs have plummeted.
Competition: Learn it, love it.
Why starve to death when there's free food, housing and money to be had?
If you change "corporations" to "entities", then yes.
Either that or because he's team red and ran in a team red district.
Nor are they Scotsmen.
The ISPs would be bribing the state to raise barriers to new ISPs having access to the fiber, of course.
The more you put under the power of the state, the more opportunity for corruption.
To be sure there is a role for state regulation but when you're introducing regulations to fix problems cause by regulation, it's time to take a step back and reconsider what the fuck you're doing.
Why not either add or remove a dash and be ISO compliant?
The metric system rationalizes and reforms a whole bunch of physical phenomena.
You cant do that with dates & times. They're intrinsically arbitrary and irrational. The current system is set up to account for human needs so it likely the best compromise that's possible. Even the metric system had to bend a little and have centimeters for common use.
Yup, it looks like it is only the USA that uses m/d/y exclusively
http://en.wikipedia.org/wiki/D...
m/d/y is by no means internationally accepted. It seems to be mostly an american thing. The UK (and I think most of Europe) uses d/m/y and I'm sure it varies around the world. I much prefer YYYY-MM-DD anyway. It squares with how we do time and is an ISO standard. m/d/y is just nonsense whichever way you slice it.
Privacy concerns? I can't see that being an issue, somehow.
I question your assumption. Third party payment processors for $ have costs not associated with Bitcoin and something of a lock-in due to network effects (Don't take Visa? You just cut out 80% or more of your potential customers. Take Bitcoin? Use any one of an increasing number of processors).
Of course, hopefully the processors will fade with time anyway as suggested below by the AC.
Mining fees will be kept low by intense competition amongst miners. This is also an incentive for merchants to accept and use Bitcoin rather than directly converting to $.