This was my first question too. Real-world testing is all well and good, but how controlled was it? What assurances do we have that his results really paint the picture he claims and not something else?
Note: The site is down so I haven't read the article yet. I'm guessing it fails to address this concern?
MBAs are good in cutting corners in traditional businesses, but generally have no understanding of technology risks.... This sort of arrogance is so common it's not even funny. I once presented a GIS plot to such a person. You know, the kind of thing that crunches so much data it takes a cluster of machines upwards of several minutes just to produce a single frame? Well this guy argued if I needed so much computer power to make a simple picture I must be doing something wrong.
... legislation requiring mommy to wipe your ass until age 18, at which time it becomes the responsibility of your employer (or the EDD if you are jobless).
No, the law does not require material to be taken down just because somebody asks. In fact, the law quite explicitly states that a takedown notice is only valid if it comes from a duly authorized party. If you read the synopsis carefully, you'll see that it does not say anything about what YouTube is required to do, rather it says what they are not required to do. They do not have to verify whether a takedown notice is legit. They could choose to do so, but it would seem they don't.
This approach is probably a safe and prudent practice. IANAL, but by my cursory examination of the law's text, I see nothing granting a service provider the time to validate a claim. Hence, if YouTube left the content online while they investigated a takedown notice, and that notice turned out to be legitimate, they might be held liable for everything that transpired during that interval. Thus, to be safe, they take the material down straight away and only put it back once they've refuted the complaint.
Perhaps the point is to create something that oneself or another finds beneficial. Perhaps the point is to do something one enjoys. There are many reasons why somebody might volunteer their time for an OSS project. The prospect of earning money is not the only motivator in this world. Not everybody feels a need for material reward in exchange for their effort. Are you suggesting they should?
As for OSS developers not getting a good deal, that's your opinion. What constitutes a good or bad deal is entirely subjective. Who are you to decide for everyone what is and is not satisfactory?
The problem with your argument here is that, like many, you do not understand fairness. The word fair refers to equal opportunity, not equal occurrence.
Give two people each a coin with a perfect 50/50 probability of landing on either side and have each flip it 1,000 times. It is entirely possible that one may get all heads while the other gets a mostly even split. This is not unfair. They both had the same opportunity of getting the same result.
The same applies to the discussion here. If we assume that both yourself and an OSS developer have the same general breakdown of hours (40 working for money, 78 free, the rest sleeping), it is in no way unfair that you spend your 78 free hours watching TV while the OSS dev spends it writing software. Both have the same opportunity to code during those hours. There is nothing to stop you from doing the same. That's not unfair. That's your choice.
I'm a programmer that worked in the software industry professionally for a good 13-years or so before switching to science. I personally find that I'm always writing little scripts and other programs to carry out all manner of tasks. Everything from gathering data to organizing it, analyzing it, charting it, reporting it, sharing it... you name it I do it. Now I cannot say whether this is a truly necessary function or whether it's due to my background. I am, if you will, a biased sample. It is a valuable skill, nonetheless. It saves a tremendous amount of time.
Ah, but this is not what Apple is doing. They do not require the purchasers of Leopard to also buy a Mac. Anybody is free to just buy the OS. It may not be usable on anything other than Apple machines, but this fact is not exactly hidden from consumers. It is plainly stated on Apple's website as well as in the license agreement, and even on the software box itself. All of these are freely available to anyone prior to purchase. There is no justification, other than sheer ignorance, for buying Leopard and being surprised that it cannot be used on a Dell.
Don't even think about trying to put OS X on your PC without first purchasing a legitimate copy of Mac OS Leopard.
This is good advice. However, I would also recommend reading the Leopard SLA too, particularly section 2:
This License allows you to install, use and run one (1) copy of the Apple Software on a single Apple-labeled computer at a time. You agree not to install, use or run the Apple Software on any non-Apple-labeled computer, or to enable others to do so.
There is nothing in the distributed systems that precludes one from using them in a centralized fashion if so desired. The term decentralized, as used in this context, simply means there is no inherent central point. It does not mean one cannot be established.
I said Internet servers, not web servers. Thus, I was including all of the various things one might wish to do with a remote shared-machine. Many of these things do require one to have a system account.
Additionally, I said modern system models were reminiscent of the older mainframe approach, not the same as. The point I was making is that multi-user, remote-computing is experiencing a resurgence of popularity; and that UNIX has its roots in that way of doing things, whereas Windows does not. I never said Windows lacked the capability.
To extend your point, the reason UNIX systems don't have UAC-style privilege elevation is due to its history. UNIX came into being, and was largely developed, during an era in which virtually all computers were large, multi-user systems that sat in a back room. An administrator would have to be sitting at a terminal 24/7 just in case somebody came knocking -- quite an unreasonable expectation. As a result, programmers had to get used to the idea of restricted abilities.
With the desktop computer model, the situation is quite different. Classically-speaking, the user is sitting right at the machine and is the only one using it. They are the administrator as well as the user. There is no expectation of security since nobody else is involved. Windows derives much of its architecture and style from this method of computing.
Modern-day computing is rapidly moving back toward the shared-computer model. This is occurring somewhat on the front-end (e.g. individual user accounts on a desktop machine for different users), but mostly it's happening on the back-end. Internet servers are very reminiscent of the mainframe-era multi-user model. This is why UNIX is such a good fit for such tasks -- it was designed specifically for it, whereas Windows has had to play catch-up. UAC is a good example of single-user thinking applied to a multi-user problem.
If I may, I believe the reason your attempts are being perceived as minimal is because that which you're doing is considered by many to be a normal, everyday thing. You're not being seen as doing anything special or unique. The only thing newsworthy is who's doing it, not what's being done.
"For example, US schools continue to lag behind internationally in science and math education. On the other hand, the US is the largest, single, R&D-performing nation in the world pumping some $340 billion into future-related technologies. The US also leads the world in patent development."
This is surprising? I suppose one could think so if they are expecting said patents and R&D activity to have scientific intent. In my experience, most patents and R&D activity are for commercial purposes. They exist to make money. The science is just a means to that end, rather than the end itself.
I routinely run a comparison between Apple products and other vendors in an attempt to maintain an objective measure of this particular point. In my observation, Apple products are generally cheaper than the competition..... for the same product.
Case in point, my last comparison: Baseline Apple Xserve $3,000. Same configuration from Dell (as best as I could get it): $7,500. The key difference: OS licenses.
So what people really need to consider is not just the price but what one is actually getting for that price. Otherwise, apples and oranges, fruit salad, blah blah blah....
So far the only uses for Cell chips have been research stuff and the PS3.
May one suppose by "research stuff" they are referring to the end-uses of the chip? One could interpret the statement as meaning the chip itself has only (outside of the PS3) been used as the subject of research. That would seem to contradict the server offerings from Mercury Systems.
I've always found it funny how creationists and their supporters argue against the teaching of evolution on the basis that, "It's only a theory." What they seem to forget (or were never taught) is that everything in science is, at most, a theory. Nothing is certain, nothing is absolute. Nothing can be deemed "irrefutable" as such a statement implies one can tell the future. It is always possible that new data will someday come to light and contradict that which we hold to be a "universal truth."
Of course, I suppose if the creationists did understand this concept they would attempt to use it against the scientific community. If nothing is absolutely certain, how then do you know anything, right?
*looks around at all of the gadgets, cars, buildings, medical technology, etc.*
Well, we seem to be applying science quite well despite the fact.:)
... relational databases are the ONLY known method to CORRECTLY reflect the realities of modeling the real world as data...
The only known method or the only method you know of? Not only that, but what exactly is the definition of correctly reflecting the realities of modeling the real world as data? Relational technology may be based upon sound mathematics, but the application thereof is not. Any application of technology to a particular task is subjective. Every situation is different, and everybody involved has different ideas of what is important.
Don't get me wrong. As my other posts in this discussion illustrate, I don't for a minute agree with the notion that relational technology will be completely replaced by this. That's just good old-fashioned FUD, designed to stir up demand for something different. All I'm saying is that there is no one technology that is the best or the only for all situations.
This method, while more modern, does not mean better.
You're right, it doesn't necessarily mean better. Once again, there's a right tool for the right job. As the article describes, a column-oriented database is optimized for a much different purpose than a row-oriented one. The following example may help to clarify this point.
Consider a database such as the one employed by Google Images. Each record represents a single image, and it contains, among other things, the format (e.g. JPEG, GIF, PNG), the dimensions, the file size, the date/time spidered, and of course the pixel data in both original and thumbnail sizes. Now suppose you wanted to see the dimensions of all JPEG images. In a row-oriented system, each JPEG record would have to be loaded from disk in its entirety (including the possibly multi-megabyte image data) just to extract a few bytes of dimension information. Using Google's column-oriented Bigtable system, however, only the file containing the dimension data is accessed. The image data isn't even thought of.
Now many people using row-oriented systems for such applications tend to keep their image data outside of the database for this very reason. However, one could reasonably argue that this is nothing more than a pseudo-column-oriented approach.
I don't think relational databases are obsolete, as there are a great many applications which the technology fits nicely. I do, however, think they are greatly overused. From what I can tell, people are being trained to use RDBMS anytime they need to persist data. I can't count the number of times I've asked somebody why they're using one and the response is, "It's what I know."
I read a very good statement about relational technology. It said one should not think of an RDBMS as a persistence tool. Sure, it can be used for that, but it's better to think of it as an integration tool. It should be used as a mechanism to permit the uniform sharing of information between different applications, built using different languages, running on different machines, hosted by different operating systems. That's, the statement argued, is the real power.
I'm inclined to agree. If you've got a simple blogger app that doesn't need distribution capabilities, doesn't need language-independence, doesn't need OS-independence, just needs to persist an ordered list of very basic data, an RDBMS is way overkill.
In my view, the problem is more general than the issues surrounding software-as-a-service. In my experience, there is a distinct lack of concern for quality. Everyone involved, the engineers, management, and even the consumers, seem more concerned with getting something done than getting it done right. I can't tell you how many times I've been ridiculed and accused of "making mountains out of molehills" just because I took a few minutes to consider risks and failure modes. I've been accused of impeding progress simply because I refused to put untested code into a live, 24/7 production system. Many people just don't want to give any more time to something than absolutely necessary, even when they have direct evidence showing them how stupid they're being.
I had this one project where, after years of me begging for approval to fix known and very serious bugs, after countless crashes and losses of data, after huge six-figure billing errors, and after tens of thousands of dollars of fees and other penalties, management still refused to acknowledge anything was broken. I had another project where one of the team-leads ran out of logical arguments to support their case and begin ranting about how they were a millionaire.
This is not to say that nobody cares about quality. I've met quite a few who do. However, they are by far the exception and not the rule. The average person would rather have a box of crap than nothing at all. Until that attitude changes, nothing is going to improve. So long as people are accepting of inferiority, there is little incentive to offer them anything better.
Slashdot Mirror
This was my first question too. Real-world testing is all well and good, but how controlled was it? What assurances do we have that his results really paint the picture he claims and not something else?
Note: The site is down so I haven't read the article yet. I'm guessing it fails to address this concern?
... legislation requiring mommy to wipe your ass until age 18, at which time it becomes the responsibility of your employer (or the EDD if you are jobless).
Because features are what sell in the mass-market. Most consumers don't select products based on their merits.
No, the law does not require material to be taken down just because somebody asks. In fact, the law quite explicitly states that a takedown notice is only valid if it comes from a duly authorized party. If you read the synopsis carefully, you'll see that it does not say anything about what YouTube is required to do, rather it says what they are not required to do. They do not have to verify whether a takedown notice is legit. They could choose to do so, but it would seem they don't.
This approach is probably a safe and prudent practice. IANAL, but by my cursory examination of the law's text, I see nothing granting a service provider the time to validate a claim. Hence, if YouTube left the content online while they investigated a takedown notice, and that notice turned out to be legitimate, they might be held liable for everything that transpired during that interval. Thus, to be safe, they take the material down straight away and only put it back once they've refuted the complaint.
Perhaps the point is to create something that oneself or another finds beneficial. Perhaps the point is to do something one enjoys. There are many reasons why somebody might volunteer their time for an OSS project. The prospect of earning money is not the only motivator in this world. Not everybody feels a need for material reward in exchange for their effort. Are you suggesting they should?
As for OSS developers not getting a good deal, that's your opinion. What constitutes a good or bad deal is entirely subjective. Who are you to decide for everyone what is and is not satisfactory?
The problem with your argument here is that, like many, you do not understand fairness. The word fair refers to equal opportunity, not equal occurrence.
Give two people each a coin with a perfect 50/50 probability of landing on either side and have each flip it 1,000 times. It is entirely possible that one may get all heads while the other gets a mostly even split. This is not unfair. They both had the same opportunity of getting the same result.
The same applies to the discussion here. If we assume that both yourself and an OSS developer have the same general breakdown of hours (40 working for money, 78 free, the rest sleeping), it is in no way unfair that you spend your 78 free hours watching TV while the OSS dev spends it writing software. Both have the same opportunity to code during those hours. There is nothing to stop you from doing the same. That's not unfair. That's your choice.
I'm a programmer that worked in the software industry professionally for a good 13-years or so before switching to science. I personally find that I'm always writing little scripts and other programs to carry out all manner of tasks. Everything from gathering data to organizing it, analyzing it, charting it, reporting it, sharing it ... you name it I do it. Now I cannot say whether this is a truly necessary function or whether it's due to my background. I am, if you will, a biased sample. It is a valuable skill, nonetheless. It saves a tremendous amount of time.
Did they also get slapped for sending people junk-mail wth signature-required? Oh please, oh please, oh please! :P
Damn right! :)
Ah, but this is not what Apple is doing. They do not require the purchasers of Leopard to also buy a Mac. Anybody is free to just buy the OS. It may not be usable on anything other than Apple machines, but this fact is not exactly hidden from consumers. It is plainly stated on Apple's website as well as in the license agreement, and even on the software box itself. All of these are freely available to anyone prior to purchase. There is no justification, other than sheer ignorance, for buying Leopard and being surprised that it cannot be used on a Dell.
Don't even think about trying to put OS X on your PC without first purchasing a legitimate copy of Mac OS Leopard.
This is good advice. However, I would also recommend reading the Leopard SLA too, particularly section 2:
This License allows you to install, use and run one (1) copy of the Apple Software on a single Apple-labeled computer at a time. You agree not to install, use or run the Apple Software on any non-Apple-labeled computer, or to enable others to do so.
There is nothing in the distributed systems that precludes one from using them in a centralized fashion if so desired. The term decentralized, as used in this context, simply means there is no inherent central point. It does not mean one cannot be established.
I said Internet servers, not web servers. Thus, I was including all of the various things one might wish to do with a remote shared-machine. Many of these things do require one to have a system account.
Additionally, I said modern system models were reminiscent of the older mainframe approach, not the same as. The point I was making is that multi-user, remote-computing is experiencing a resurgence of popularity; and that UNIX has its roots in that way of doing things, whereas Windows does not. I never said Windows lacked the capability.
To extend your point, the reason UNIX systems don't have UAC-style privilege elevation is due to its history. UNIX came into being, and was largely developed, during an era in which virtually all computers were large, multi-user systems that sat in a back room. An administrator would have to be sitting at a terminal 24/7 just in case somebody came knocking -- quite an unreasonable expectation. As a result, programmers had to get used to the idea of restricted abilities.
With the desktop computer model, the situation is quite different. Classically-speaking, the user is sitting right at the machine and is the only one using it. They are the administrator as well as the user. There is no expectation of security since nobody else is involved. Windows derives much of its architecture and style from this method of computing.
Modern-day computing is rapidly moving back toward the shared-computer model. This is occurring somewhat on the front-end (e.g. individual user accounts on a desktop machine for different users), but mostly it's happening on the back-end. Internet servers are very reminiscent of the mainframe-era multi-user model. This is why UNIX is such a good fit for such tasks -- it was designed specifically for it, whereas Windows has had to play catch-up. UAC is a good example of single-user thinking applied to a multi-user problem.
If I may, I believe the reason your attempts are being perceived as minimal is because that which you're doing is considered by many to be a normal, everyday thing. You're not being seen as doing anything special or unique. The only thing newsworthy is who's doing it, not what's being done.
What's the Super Bowl? Is that anything like Puppy Bowl? :P
"For example, US schools continue to lag behind internationally in science and math education. On the other hand, the US is the largest, single, R&D-performing nation in the world pumping some $340 billion into future-related technologies. The US also leads the world in patent development."
This is surprising? I suppose one could think so if they are expecting said patents and R&D activity to have scientific intent. In my experience, most patents and R&D activity are for commercial purposes. They exist to make money. The science is just a means to that end, rather than the end itself.
I routinely run a comparison between Apple products and other vendors in an attempt to maintain an objective measure of this particular point. In my observation, Apple products are generally cheaper than the competition ..... for the same product.
....
Case in point, my last comparison: Baseline Apple Xserve $3,000. Same configuration from Dell (as best as I could get it): $7,500. The key difference: OS licenses.
So what people really need to consider is not just the price but what one is actually getting for that price. Otherwise, apples and oranges, fruit salad, blah blah blah
I'm wondering about the statement:
So far the only uses for Cell chips have been research stuff and the PS3.
May one suppose by "research stuff" they are referring to the end-uses of the chip? One could interpret the statement as meaning the chip itself has only (outside of the PS3) been used as the subject of research. That would seem to contradict the server offerings from Mercury Systems.
I've always found it funny how creationists and their supporters argue against the teaching of evolution on the basis that, "It's only a theory." What they seem to forget (or were never taught) is that everything in science is, at most, a theory. Nothing is certain, nothing is absolute. Nothing can be deemed "irrefutable" as such a statement implies one can tell the future. It is always possible that new data will someday come to light and contradict that which we hold to be a "universal truth."
:)
Of course, I suppose if the creationists did understand this concept they would attempt to use it against the scientific community. If nothing is absolutely certain, how then do you know anything, right?
*looks around at all of the gadgets, cars, buildings, medical technology, etc.*
Well, we seem to be applying science quite well despite the fact.
... relational databases are the ONLY known method to CORRECTLY reflect the realities of modeling the real world as data ...
The only known method or the only method you know of? Not only that, but what exactly is the definition of correctly reflecting the realities of modeling the real world as data? Relational technology may be based upon sound mathematics, but the application thereof is not. Any application of technology to a particular task is subjective. Every situation is different, and everybody involved has different ideas of what is important.
Don't get me wrong. As my other posts in this discussion illustrate, I don't for a minute agree with the notion that relational technology will be completely replaced by this. That's just good old-fashioned FUD, designed to stir up demand for something different. All I'm saying is that there is no one technology that is the best or the only for all situations.
This method, while more modern, does not mean better.
You're right, it doesn't necessarily mean better. Once again, there's a right tool for the right job. As the article describes, a column-oriented database is optimized for a much different purpose than a row-oriented one. The following example may help to clarify this point.
Consider a database such as the one employed by Google Images. Each record represents a single image, and it contains, among other things, the format (e.g. JPEG, GIF, PNG), the dimensions, the file size, the date/time spidered, and of course the pixel data in both original and thumbnail sizes. Now suppose you wanted to see the dimensions of all JPEG images. In a row-oriented system, each JPEG record would have to be loaded from disk in its entirety (including the possibly multi-megabyte image data) just to extract a few bytes of dimension information. Using Google's column-oriented Bigtable system, however, only the file containing the dimension data is accessed. The image data isn't even thought of.
Now many people using row-oriented systems for such applications tend to keep their image data outside of the database for this very reason. However, one could reasonably argue that this is nothing more than a pseudo-column-oriented approach.
I don't think relational databases are obsolete, as there are a great many applications which the technology fits nicely. I do, however, think they are greatly overused. From what I can tell, people are being trained to use RDBMS anytime they need to persist data. I can't count the number of times I've asked somebody why they're using one and the response is, "It's what I know."
I read a very good statement about relational technology. It said one should not think of an RDBMS as a persistence tool. Sure, it can be used for that, but it's better to think of it as an integration tool. It should be used as a mechanism to permit the uniform sharing of information between different applications, built using different languages, running on different machines, hosted by different operating systems. That's, the statement argued, is the real power.
I'm inclined to agree. If you've got a simple blogger app that doesn't need distribution capabilities, doesn't need language-independence, doesn't need OS-independence, just needs to persist an ordered list of very basic data, an RDBMS is way overkill.
In my view, the problem is more general than the issues surrounding software-as-a-service. In my experience, there is a distinct lack of concern for quality. Everyone involved, the engineers, management, and even the consumers, seem more concerned with getting something done than getting it done right. I can't tell you how many times I've been ridiculed and accused of "making mountains out of molehills" just because I took a few minutes to consider risks and failure modes. I've been accused of impeding progress simply because I refused to put untested code into a live, 24/7 production system. Many people just don't want to give any more time to something than absolutely necessary, even when they have direct evidence showing them how stupid they're being.
I had this one project where, after years of me begging for approval to fix known and very serious bugs, after countless crashes and losses of data, after huge six-figure billing errors, and after tens of thousands of dollars of fees and other penalties, management still refused to acknowledge anything was broken. I had another project where one of the team-leads ran out of logical arguments to support their case and begin ranting about how they were a millionaire.
This is not to say that nobody cares about quality. I've met quite a few who do. However, they are by far the exception and not the rule. The average person would rather have a box of crap than nothing at all. Until that attitude changes, nothing is going to improve. So long as people are accepting of inferiority, there is little incentive to offer them anything better.