Hey, I have only ever taken one course in statistics, but the simplest thing I know is that outliers can ruin data.
Take away the single largest source of "bandwidth", one internet site, one mail order company.
Then compare. There is virtually no snail mail based "bandwidth" besides NetFlix. This is just fodder for the people that like pushing numbers around and saying, "Wow, look!"
It's very simple! The more rare (or in demand perhaps) your skill, the more you are paid for performing it.
A very very skilled artist is a rare thing, and they make good money. A skilled programmer is an important asset, and they get paid a lot of money. A skilled garbageman is anyone who can lift 25 pounds. They don't get paid a lot.
When reading this I am strongly reminded of a lesson I learned back in 8th grade history. See, in early America, there weren't all the laws that protect companies now. Monopolies and trusts abounded. Big companies trounced little ones.
A specific example (well not specific enough, it was way back in 8th grade ya know) was the steel industry. The big monopoly in the steel industry stayed that way. Why? When someone else came along they would undersell the startup, who couldn't afford to keep up. Then big monopoly raised it's prices again and continued making money without competition.
This situation has many analogies, but also differences. The situation was just a few years ago when the internet was taking off with a bang. Netscape was (nearly) the only player in town. This was back when Netscape was version 2 mind you. Then Microsoft sees the internet is taking off and decides to take over. They toss some resources at throwing together an internet browser. They (of course) don't take as much effort as Netscape did, they just copy Netscape. Now there's a competiton. Over the course of few months, both Netscape and IE jumped a few major versions. To any lightly computer-savvy individiual, it's clear that the rate these improvements were coming out, the code behind them was sloppy. The products just needed something else to have a chance at catching the consumer.
So, bloat, bloat, bloat, feature here feature there. Then MS turns IE free. BAM end of development. NS can't afford to continue. MS slowly recoups IE as a decent product since they are of course still around and NS has turned around like a hurt puppy licking it's wounds. NS is crap just like IE is crap because they were very rapidly thrown together, then the potential profit from increasing the quality fell to near zero.
No, not any good can be considered a service that way.
When I go to McDonalds and buy a hamburger, I own that hamburger. It is mine and I eat it. If I pay someone to mow my lawn, I have a mown lawn, but I have no lawn mower, and nothing besides what I paid for, a mown lawn.
When I buy a car I do exactly that. I own the car, and I can do whatever (within legal bounds of course) I want with it. I can drive to work, I can put it in my garage and drool on it, whatever. It is mine. But you cannot own a service. I don't own the cleaning my dentist did on my teeth. He did it for me, and I paid him for it.
And the grease that slips the gears of this whole situation is that computer software is very different from a product or service. Although it is somewhat of a product, and somewhat of a service (I don't think so but some companies would like it to be).
Now of course the manual, the box, and whatever media be it floppy or CD or whatever is a product. But it is very different from every other product that exists. The reason is that it can, without cost or significant effort, be duplicated beyond legal bounds. When I buy a book, I can loan it to my friend. And he can read it, and life is good. But when I want to read the book again, he *HAS* to give me back my book. With software, I can 'give' my program to my friend, and we both have it.
I've thought about this before and never came to a solution. Software can be copied, and with little argument we can say that will never stop, but that is not the issue here. In it's basest form, software is a product, and I can (should? be able to) sell that product, after which point the buyer can use it. But the little legal point is once I've sold it I can't use it anymore, while that is often not the case.
I couldn't agree more. The ISP is the only body that can do anything to remedy the problem from the infected server perspective.
@home already runs scans of computers, for NNTP servers AFAIK, maybe more. They should scan for infected computers (that get blabla/cmd.exewhatever) and disconnect them, and contact the owners. They're the only ones that have firm connectionowner records through billing.
Unfortunately there's nothing that legally forces ISPs to do this and they won't. Bah.
I think that's the very basic reason. It solves problems very very well. I know I've wanted to partake in OS projects just to 'give something back' as I use almost ONLY free as in beer software for personal uses.
But my main point is that OS software really works. It's as if every expert in the world that wanted to had a say in the design of [insert object here]. I myself have on multiple occasions tweaked OS programs that I found and liked, to be programs I liked even more. Just the other day, I modified smb2www to work by passing username/password pairs to my Win2K hosts. If it wasn't open source I would have just said "Oh crap, that would be a really useful program but it doesn't work. Poo." But, I could fix it. I had an OS instant message client I liked a while back, except it displayed idle times as raw minutes. I made it do hour:minute.
Now not everyone is a developer, or even able to code at all. But surely you've seen the OS spinoffs! Linux itself is an amazing example of this. Linux was one thing once. But, it is OS, and someone saw it and said "I wish it..." and a bit later, it DID! Then it's made available for everyone else. The original person gets a solution that fits perfectly for them, everyone else with the same problem gets the same great fitting solution.
OS software really works. But by it's nature, it is not as "beginner friendly". I have in all honesty see people struggle with a web browser interface. Compiling a program from downloaded source is FAR beyond these people, even when the instructions are laid out in plain simple language. Even when the instructions are shouted to them, I bet. These people often can't even find the installation program they downloaded when running Windows.
We have, on the other hand, made quite a bit of progress. My first unix/linux experience was with FreeBSD. Once I installed RedHat a few months later I was amazed with how easy it was. It still brought some fdisking into the picture, which is too much for some people though.
I SERIOUSLY disagree with your opinion that BattleBots is so much better than Robot Wars. It is a bit exciting to watch, but it's really already gotten blah. There's basically the flippers, and the spinners and that's it. There is no point but destroy.
In the Robot Wars of yore though, there were GOALS. Your robot had to be agile, and quick, and able to fight it out a bit. It took a lot of thinking to design a robot that could get the most points across a wide range of competitons, instead of just designing a few spinning hammers on wheels.
No my friend, yours is the flawed analogy
on
Software Aesthetics
·
· Score: 1
Code does in fact need to be "pretty". But what means "pretty" for code is different than for something else. Good code needs to be maintainable, because it is almost inevitably: an integral part of a larger whole, and going to be changed in the future.
If a car manufacturer upgraded your car's engine every year, they would probably go to greater lengths to make sure that it fit in such a way that they could easily remove the existing engine, then replace it with the new one.
The way code is *VERY* similar to the auto industry as you mention, is that extreme care is required to make each part not "pretty" but "properly". Every part has to meet strict requirements and tolerances. This is where the analogy begins to show differences.
For the most part, pieces of cars are very carefully designed very smooth operating pieces of machinery. The bearings that let the wheels turn were designed with great care, and the people designing the bearings were probably very careful to talk to the people designing the wheels to make sure that either part would do exactly what they needed, and work together well.
Two interfacing routines in code often do not have the same characteristics. A large number of programmers in today's field either: don't care or don't know how to make good code. I've been pained watching my coworkers code in the past, replaced existing routines with amazingly smaller and more efficient pieces, cringed as I watch "computer experts" just mash the arrow key and wait to be at the beginning of the line instead of just pressing "home".
I think pretty is a very subjective word. Things will be pretty only from the right perspective. I believe the right eye can look at a house's wiring in the same way another would look at fine art in a museum. I've seen a lot of machines that nearly took my breath away, just because they accomplished their task so elegantly and cleanly. I've rarely seen code that did the same (in the workplace).
I'm afraid you've gotten it a bit wrong. The point is that AMD is simply changing the naming policy.
For example. As it exists now, an Athlon processor running at 1.2GHz is called "Athlon 1.2GHz". NOW they will call it "Athlon 1400" without saying Hz. Right next to a "Pentium IV 1400MHz" for example, it will appear the same to the kind of consumer that will be basing their purchase decision on the CPU speed.
Doesn't this make sense:
In addition to the requirement that nobody else be using the trademark you want to use (or something similar), your trademark must also be distinctive. That means it cannot be so generic that it just describes the product instead of the source of it. For example, if you sell timber, you cannot trademark the word "wood" or "timber," but you could trademark "wood-o-rama" or "timbermania."
From http://www.soyouwanna.com/site/syws/trademark/trad emarkFULL.html
*pout*
...the 'managers' in this case cannot be ignored, they are there. BUT they are not the least able to contribute, but often the most able.
I see a lot of OS projects start out small, and grow and grow. The original, and often very very talented, individuals usually wield the most control. From what I've seen of the business world (a few internships), the managers are usually the least competent. So these projects are "self-managed" if you look at it the right way, it's managed by specific people, it's just not managed by some seperate entity.
I'm currently a college student, in computer engineering. I definetly have seen these problems first-hand. Here in the US computer schooling is defintely horrible, demand far outweighs supply. But it's even worse abroad. Over this past summer I had an internship where I worked with an indian person who was schooled there. He was not exactly the brightest crayon in the box. His computer skills are atrocious. It's a tough problem, but the solution is simply to get the teachers so our OWN people learn what we need to know!
Slashdot Mirror
Hey, I have only ever taken one course in statistics, but the simplest thing I know is that outliers can ruin data.
Take away the single largest source of "bandwidth", one internet site, one mail order company.
Then compare. There is virtually no snail mail based "bandwidth" besides NetFlix. This is just fodder for the people that like pushing numbers around and saying, "Wow, look!"
It's very simple! The more rare (or in demand perhaps) your skill, the more you are paid for performing it.
A very very skilled artist is a rare thing, and they make good money. A skilled programmer is an important asset, and they get paid a lot of money. A skilled garbageman is anyone who can lift 25 pounds. They don't get paid a lot.
Simple supply vs. demand economics rules.
When reading this I am strongly reminded of a lesson I learned back in 8th grade history. See, in early America, there weren't all the laws that protect companies now. Monopolies and trusts abounded. Big companies trounced little ones.
A specific example (well not specific enough, it was way back in 8th grade ya know) was the steel industry. The big monopoly in the steel industry stayed that way. Why? When someone else came along they would undersell the startup, who couldn't afford to keep up. Then big monopoly raised it's prices again and continued making money without competition.
This situation has many analogies, but also differences. The situation was just a few years ago when the internet was taking off with a bang. Netscape was (nearly) the only player in town. This was back when Netscape was version 2 mind you. Then Microsoft sees the internet is taking off and decides to take over. They toss some resources at throwing together an internet browser. They (of course) don't take as much effort as Netscape did, they just copy Netscape. Now there's a competiton. Over the course of few months, both Netscape and IE jumped a few major versions. To any lightly computer-savvy individiual, it's clear that the rate these improvements were coming out, the code behind them was sloppy. The products just needed something else to have a chance at catching the consumer.
So, bloat, bloat, bloat, feature here feature there. Then MS turns IE free. BAM end of development. NS can't afford to continue. MS slowly recoups IE as a decent product since they are of course still around and NS has turned around like a hurt puppy licking it's wounds. NS is crap just like IE is crap because they were very rapidly thrown together, then the potential profit from increasing the quality fell to near zero.
Remember when computer chips were called 486s?
And remember when Intel called their chip the Pentium so they could copyright the name?
Since you can't copyright numbers...
obsequious adj 1: attempting to win favor from influential people by flattery
/. community.
...
Seems like good usage to me. He was trying to seem smart by pointing out someone else's mistake, and win the favor of the influental (?)
No, not any good can be considered a service that way.
When I go to McDonalds and buy a hamburger, I own that hamburger. It is mine and I eat it. If I pay someone to mow my lawn, I have a mown lawn, but I have no lawn mower, and nothing besides what I paid for, a mown lawn.
When I buy a car I do exactly that. I own the car, and I can do whatever (within legal bounds of course) I want with it. I can drive to work, I can put it in my garage and drool on it, whatever. It is mine. But you cannot own a service. I don't own the cleaning my dentist did on my teeth. He did it for me, and I paid him for it.
And the grease that slips the gears of this whole situation is that computer software is very different from a product or service. Although it is somewhat of a product, and somewhat of a service (I don't think so but some companies would like it to be).
Now of course the manual, the box, and whatever media be it floppy or CD or whatever is a product. But it is very different from every other product that exists. The reason is that it can, without cost or significant effort, be duplicated beyond legal bounds. When I buy a book, I can loan it to my friend. And he can read it, and life is good. But when I want to read the book again, he *HAS* to give me back my book. With software, I can 'give' my program to my friend, and we both have it.
I've thought about this before and never came to a solution. Software can be copied, and with little argument we can say that will never stop, but that is not the issue here. In it's basest form, software is a product, and I can (should? be able to) sell that product, after which point the buyer can use it. But the little legal point is once I've sold it I can't use it anymore, while that is often not the case.
Am I the only one that thinks anything but 0 Quake playing at work is too much?
I couldn't agree more. The ISP is the only body that can do anything to remedy the problem from the infected server perspective.
@home already runs scans of computers, for NNTP servers AFAIK, maybe more. They should scan for infected computers (that get blabla/cmd.exewhatever) and disconnect them, and contact the owners. They're the only ones that have firm connectionowner records through billing.
Unfortunately there's nothing that legally forces ISPs to do this and they won't. Bah.
I think that's the very basic reason. It solves problems very very well. I know I've wanted to partake in OS projects just to 'give something back' as I use almost ONLY free as in beer software for personal uses.
But my main point is that OS software really works. It's as if every expert in the world that wanted to had a say in the design of [insert object here]. I myself have on multiple occasions tweaked OS programs that I found and liked, to be programs I liked even more. Just the other day, I modified smb2www to work by passing username/password pairs to my Win2K hosts. If it wasn't open source I would have just said "Oh crap, that would be a really useful program but it doesn't work. Poo." But, I could fix it. I had an OS instant message client I liked a while back, except it displayed idle times as raw minutes. I made it do hour:minute.
Now not everyone is a developer, or even able to code at all. But surely you've seen the OS spinoffs! Linux itself is an amazing example of this. Linux was one thing once. But, it is OS, and someone saw it and said "I wish it..." and a bit later, it DID! Then it's made available for everyone else. The original person gets a solution that fits perfectly for them, everyone else with the same problem gets the same great fitting solution.
OS software really works. But by it's nature, it is not as "beginner friendly". I have in all honesty see people struggle with a web browser interface. Compiling a program from downloaded source is FAR beyond these people, even when the instructions are laid out in plain simple language. Even when the instructions are shouted to them, I bet. These people often can't even find the installation program they downloaded when running Windows.
We have, on the other hand, made quite a bit of progress. My first unix/linux experience was with FreeBSD. Once I installed RedHat a few months later I was amazed with how easy it was. It still brought some fdisking into the picture, which is too much for some people though.
I SERIOUSLY disagree with your opinion that BattleBots is so much better than Robot Wars. It is a bit exciting to watch, but it's really already gotten blah. There's basically the flippers, and the spinners and that's it. There is no point but destroy.
In the Robot Wars of yore though, there were GOALS. Your robot had to be agile, and quick, and able to fight it out a bit. It took a lot of thinking to design a robot that could get the most points across a wide range of competitons, instead of just designing a few spinning hammers on wheels.
Code does in fact need to be "pretty". But what means "pretty" for code is different than for something else. Good code needs to be maintainable, because it is almost inevitably: an integral part of a larger whole, and going to be changed in the future.
If a car manufacturer upgraded your car's engine every year, they would probably go to greater lengths to make sure that it fit in such a way that they could easily remove the existing engine, then replace it with the new one.
The way code is *VERY* similar to the auto industry as you mention, is that extreme care is required to make each part not "pretty" but "properly". Every part has to meet strict requirements and tolerances. This is where the analogy begins to show differences.
For the most part, pieces of cars are very carefully designed very smooth operating pieces of machinery. The bearings that let the wheels turn were designed with great care, and the people designing the bearings were probably very careful to talk to the people designing the wheels to make sure that either part would do exactly what they needed, and work together well.
Two interfacing routines in code often do not have the same characteristics. A large number of programmers in today's field either: don't care or don't know how to make good code. I've been pained watching my coworkers code in the past, replaced existing routines with amazingly smaller and more efficient pieces, cringed as I watch "computer experts" just mash the arrow key and wait to be at the beginning of the line instead of just pressing "home".
I think pretty is a very subjective word. Things will be pretty only from the right perspective. I believe the right eye can look at a house's wiring in the same way another would look at fine art in a museum. I've seen a lot of machines that nearly took my breath away, just because they accomplished their task so elegantly and cleanly. I've rarely seen code that did the same (in the workplace).
I'm afraid you've gotten it a bit wrong. The point is that AMD is simply changing the naming policy.
For example. As it exists now, an Athlon processor running at 1.2GHz is called "Athlon 1.2GHz". NOW they will call it "Athlon 1400" without saying Hz. Right next to a "Pentium IV 1400MHz" for example, it will appear the same to the kind of consumer that will be basing their purchase decision on the CPU speed.
Don't underestimate the power of stupidity.
Doesn't this make sense: In addition to the requirement that nobody else be using the trademark you want to use (or something similar), your trademark must also be distinctive. That means it cannot be so generic that it just describes the product instead of the source of it. For example, if you sell timber, you cannot trademark the word "wood" or "timber," but you could trademark "wood-o-rama" or "timbermania." From http://www.soyouwanna.com/site/syws/trademark/trad emarkFULL.html
*pout*
...the 'managers' in this case cannot be ignored, they are there. BUT they are not the least able to contribute, but often the most able. I see a lot of OS projects start out small, and grow and grow. The original, and often very very talented, individuals usually wield the most control. From what I've seen of the business world (a few internships), the managers are usually the least competent. So these projects are "self-managed" if you look at it the right way, it's managed by specific people, it's just not managed by some seperate entity.
I'm currently a college student, in computer engineering. I definetly have seen these problems first-hand. Here in the US computer schooling is defintely horrible, demand far outweighs supply. But it's even worse abroad. Over this past summer I had an internship where I worked with an indian person who was schooled there. He was not exactly the brightest crayon in the box. His computer skills are atrocious. It's a tough problem, but the solution is simply to get the teachers so our OWN people learn what we need to know!