This happens every day when I drop mail into the postbox.
I think your analogy actually cuts the other way. When you drop your mail into the mailbox, it enters a highly regulated, automated, centralized system that collects fees (i.e., stamps) of which the government gets a cut. Yes, it's true that you do not know the people, but you sure know who they work for.
By contrast, Negroponte seems to be suggesting that you would (in effect) hand your letter to a stranger on the street, who would hand it off to another, who hands it off to another, etc., until it gets to where it's going, with no intervention by a centralized agency.
It's an interesting theory, but we'll never see it happen, for one obvious reason: it does not lend itself well to being taxed.
Also, a cost-benefit ratio is not a static thing that applies the same to everyone. Rather, it is directly affected by where you are in your life -- your resources and your interests. If you've just graduated from school and have a mound of debt and an entry-level job, this product might indeed seem too extravagant. But if you're an established professional with disposable income, and this tool would make your life easier or more enjoyable, the ratio for you is different. That's why there is a market for the Kia as well as a market for the BMW. The people who say "it's too expensive" are really only saying, "it's too expensive for me." Calculate your own cost-benefit ratio, because it's the one that matters.
This would also be a good time to remember Edmund C. Berkeley. He was an insurance executive (an actuary, I believe) who saw the commercial possibilities of the digital computer at a time when it was generally regarded as only an expensive military tool. He was instrumental in convincing Prudential to buy the Univac I. He then left the insurance industry and became the first advocate of computer education, developing some great logic toys (e.g., the Brainiac, the Geniac) and writing some great books for students (e.g., Giant Brains, Symbolic Logic and Intelligent Machines). He was one of the founding editors of Computers and Animation. Berkeley rocked.
Here is the issue in a nutshell. The world is full of kids who love to draw anime and comics and such, and some are quite good. The world will always be full of such people, millions and millions of them. But the industry does not need ten million anime artists, or even one million anime artists, or even twenty thousand. So there will always be someone good who is willing to work for peanuts. It's just like music: ninety-nine percent of musicians never rise above the bar band level, even though many are highly talented, but that doesn't stop them all from trying. In the popular arts, the supply of willing candidates will always be orders of magnitude beyond what the industry can absorb.
No, not with line numbers, and not with GOTOs. QBasic doesn't need them. If you teach it with some structure, and make sure that she declares her variables, she can have a total blast and get a feel for what programming is about. Fast and fun results will prevent loss of interest, which is probably the biggest threat to your project. Further, the knowledge that she gains will not be obsolete because the procedural statements are almost identical to VBScript. And the built-in help file is actually useful. Face it, you have to start with something fast and easy if you want to hold her interest.
Please try to take your own advice. . . . Sometimes I wish that PC monitors came with a big stick attached to them so that their operators could be thwacked about the head when they mail or post something completely inane.
Colin, pick up your telephone directory. Now turn to the business section. Now look up all of the businesses that start with a generic word like "Quality" or "National" or even a common last name like "McDonalds." If you live in a medium to large city, you will find many businesses that have similar names using generic words or common names. Multiply by all of the cities in your country. How do you suppose all of those businesses co-existed before the age of the Internet? How do you suppose McDonalds fast food and McDonalds car rental co-exist in your city today? What does any of this have to do with TLDs? That's why I'm asking you to stop and think. It's not a TLD issue. It's a trademark law issue.
Watch out for that stick attached to your monitor!
Here is my point: Take it out of the realm of domain names, and apply the same analysis that you would apply in the physical world. I open a bar called "Johnny's Bar." You open a bar down the street called "Johnny's Pub." Why does the anlysis change if we compare johnnys.bar to johnnys.pub? Or acme.com to acme.net? If the name is actually trademarked (like Kodak) then you can't do it anyway, if it isn't, well, there are a thousand bars out there than include the name "Johnny." And finally, if someone is actually committing fraud or an IP violation, that has nothing to do with the TLD process -- it's just fraud or theft. Those things are already unlawful.
And to carry the thought one step farther: if the domain name does NOT incorporate a trademarked name, such as when the word is generic, then that is exactly the kind of situation in which a lot of TLDs would actually be useful (bank.com, bank.net, bank.blah). So it works either way. If the name is trademarked, it doesn't matter if there are a lot of TLDs because you can shut down infringers, and if the name is not trademarked, it doesn't matter if there are a lot of TLDs because they don't infringe in each other. Where is the problem?
Forget domain names for a moment. Think generally. What stops anyone from choosing a business name that unlawfully incorporates another company's name? What stops anyone from creating the "Kodak Cafe" or the "Microsoft Bar and Grill"? The answer is: trademark law. Why isn't this enough? Why make such a big deal about trying to solve a problem that's already solved? Create all the TLDs that you want. I guarantee that if someone other than Kodak tries to register Kodak.blah, the registrant of Kodak.blah will be shut down. It's a non-issue.
And pardon me if you think that comments about submissions are off-topic, but once again, there are way too many hyperlinks in the submission. I do not need to know the web address of the supermarket chain's corporate headquarters, or the charity's corporate headquarters, or the event campaign's home page, or the sponsoring gallery's home page, or even the artist's home page. I just want to see the damned shopping cart helix. Pardon me for sounding like a curmudgeon, but nine times out of ten, I am only interested in one link: the link to the subject of the submission, not every related entity (which I can ferret out from the aricle if I really want to). Am I the only one who thinks so?
Out of curiousity, were you here then? I mean, your user ID is 600,000+. I was just wondering if it's a new account, and you used to post as someone else.
There is no reliable correlation between a poster's UID and whether a person was "here then." On one hand, some people (such as me) were reading Slashdot for years without having a UID, and on the other hand, many accounts with low UIDs seem to have been dormant for years. No reliable correlation.
Claude Shannon's theories underlie almost every aspect of the digital computer and digital communications. His master's thesis (1938) established that logic circuits can be simplified mathematically rather than by trial and error. His mathematical theory of communication (1948) established the entire field of information theory, making possible digital communications (modems, networks). In terms of his importance to the field, he is miles beyond most of the people on the list and most of the people already inducted.
Yes, I believe Moglen is wrong on this point. As a preliminary matter, any license requires a return promise from the licensee: not to violate the terms of the license. This is so basic it is almost circular. If you accept the benefits of the license, you must comply with the terms of the license. If Moglen is correct that the licensee has made no such promise, then how could any licensee ever be found to have committed a violation? Moglen is failing to distinguish express promises from implied promises. Sure, the licensee might not say "I promise," but the law will imply that the licensee said "I promise" when he accepts the benefits of the license.
The law.com definition of "license" also provides definitions for the parties to a license, i.e., the licensor and licensee. A "licensee" is defined as "a person given a license by the government or under private agreement." Further, according to law.com, the term "agreement", when used legally, is exactly synonymous with "contract": "in law, another name for a contract including all the elements of a legal contract: offer, acceptance, and consideration (payment or performance), based on specific terms." So the law.com definitions do support a connection between licensees and traditional contract law.
But I also see something in the law.com definition that may explain some of the controversy. There seem to be two sources of license, those granted by the government (e.g., driver's licenses, building permits) and those that arise by agreement between parties. Perhaps once of the sources of confusion is that some people regard the grant of rights under the GPL to be of the "governmental" type of license, because the underlying rights are based on copyright. My view, on the other hand, is that the grant is of the "agreement" type of license.
I hate to say it, but you're completely wrong. . . . The GPL is not a contract.
Pardon me for converting this to a meta-discussion, but your post provides an example of why the controversy will continue to rage until a court rules authoritatively on the question of whether or not the GPL is enforceable. There is no consensus even among knowledgeable people (indeed, even among knowledgeable lawyers) on the legal status of the GPL, which is why a test case is significant. But as a matter of basic contract law, a license is a form of contract and will be subject to a determination of its enforceability if one of these cases should proceed to a final resolution. Even when the terms of a contract are unilateral (i.e., not separately negotiated between the parties), the issues of contract formation and enforceability are still present (e.g., did the purported licensee accept the terms of the license, either expressly or by conduct?). These questions will not be sidestepped if this case proceeds to trial.
The GPL doesn't need tested, it's simple just copyright law . ..
Actually, the GPL does need to be tested if you want legal certainty that it is enforceable. The GPL is a license, not a copyright (that's what the "L" in GPL stands for). Licenses are contractual, and contracts are not merely enforceable on their own self-expressed authority. Contracts can be held unenforceable on a number of grounds -- lack of consent (i.e., you can't contractually bind a party that didn't actually enter the contract), lack of consideration (i.e., a contract requires that both sides give something up), violation of public policy (i.e., you can't enforce a contract that would require a party to commit an unlawful act), etc. Further, the GPL is an unusual license (in fact a unique license), so existing legal precedents do not provide a lot of predictive value. The fact that the GPL has not yet been fully adjudicated and found to be enforceable has been the source of considerable controversy.
Yes, you are correct. A preliminary injunction is not a final adjudication on the merits of the dispute. It is an interim remedy to stop the disputed activity only during the period while the case in pending. But because a preliminary injunction is a drastic remedy (it basically shuts down the defendant's activities until the case is resolved) a court will not grant one unless the plaintiff has enough evidence to demonstrate a strong likelihood of success at trial. So this means that the judge presumably reviewed the terms of the GPL, reviewed the defendant's activities, allowed the defendant to raise objections, weighed the evidence, and still granted the motion.
Stephenson makes great use of speculative history. He postulates some great "what if" scenarios arising from past events and uses them to weave an alternative present. He always succeeds in grabbing my attention. And then -- and then his male protagonist tries to talk to a woman. And that is where his novels fall apart. His dialog does not ring true. Every conversation sounds contrived. I think it's supposed to be banter, but it's just stilted. Is it any wonder he chose the name "Eliza" for the female protagonist in Quicksilver?
So the veneer glue can get some tooth into the substrate. With a wooden substrate, the glue penetrates the pores of the wood for a better bond. Otherwise, it tends to peel off when the piece is stressed.
I've worked with hardwood veneers on musical instruments. Two points: (1) you want a porous substrate so the veneers adhere properly; (2) you want a reasonably rigid (non-flexible) substrate so the veneers adhere properly. I can't image a worse substrate for fine veneer work than sheet metal. How intact will the veneers be after a year of normal use?
Someone with points to spare should mod up the parent post, since it is an important point an no one seems to have noticed it. Yes, this is the anniversary of the System/360 family of mainframes, which is a very significant anniversary for sure. But no, this is certainly not the 40th anniversary of the mainframe. The Univac I, II, III. The early Ferranti models. The IBM 700/7000 series. Lots more. Mainframes left and right, throughout the 1950s.
Last week there was a long Slashdot thread on the so-called "iPod Killer" proposed by Microsoft. That product was soundly critisized for missing the point -- the consensus among posters appeared to be that the product was too big and combined too many capabilities into one device, and thereby had lost the single-purpose elegance of the iPod. Now comes another product that appears to be conceptually similar to that Microsoft multi-purpose device. But this time it's being offered by a smaller company that a number of people like. I think it will be very interesting to compare the two threads. Was it just the words "iPod Killer" that prompted the strong opinions, or was it the underlying concept of a multi-purpose device?
For what I've seen, popup sales help is not particularly impressive. The "first level" of help appears to be essentially automated (basically a chatbot) that burns up several minutes establishing contact with an "Eliza-like" series of questions that you have to answer before you can get any further. Once you get to a real human, the level of support is not much different than what you would find if you searched the FAQ for the site (for all I could tell, it might still have been a bot except that I specifically asked if it was a person and then got a non-automated response). And I found it creepy to be accosted virtually, sort of like being stalked by a popup ad. Until they improve the interface and the substance, I'll pass.
Indeed, once a brand name enters the common parlance, it has a life independent of the company that it stands for, even if that company loses its leadership position.
Boss (handing you a stack of paper as he points to the Canon copier next to your desk): "Please xerox these documents."
Boss (handing you a stack of reference citations as he points to the Microsoft search engine on your desktop): "Please google these terms."
You might think it can't happen, but it can. The fact that Google is so dominant today is no guarantee of anything except that its name will probably remain recognizable as a verb for awhile. Google will have to continue to compete, and compete well, if it wants to stay on top. It was not very long ago that AltaVista ruled the search engine world, and it did not take very long for its user base to erode when Digital/Compaq failed to give it the priority it deserved.
They are also wrong when they say that there are companies that truly want users to hack their mass-market electronic products. Here is the killer counter-argument:
Name one manufacturer of a mass-market electronic product (including the Roomba described in the article) that will not instantly void your product warranty for doing the things described in this article.
Slashdot Mirror
This happens every day when I drop mail into the postbox.
I think your analogy actually cuts the other way. When you drop your mail into the mailbox, it enters a highly regulated, automated, centralized system that collects fees (i.e., stamps) of which the government gets a cut. Yes, it's true that you do not know the people, but you sure know who they work for.
By contrast, Negroponte seems to be suggesting that you would (in effect) hand your letter to a stranger on the street, who would hand it off to another, who hands it off to another, etc., until it gets to where it's going, with no intervention by a centralized agency.
It's an interesting theory, but we'll never see it happen, for one obvious reason: it does not lend itself well to being taxed.
Also, a cost-benefit ratio is not a static thing that applies the same to everyone. Rather, it is directly affected by where you are in your life -- your resources and your interests. If you've just graduated from school and have a mound of debt and an entry-level job, this product might indeed seem too extravagant. But if you're an established professional with disposable income, and this tool would make your life easier or more enjoyable, the ratio for you is different. That's why there is a market for the Kia as well as a market for the BMW. The people who say "it's too expensive" are really only saying, "it's too expensive for me." Calculate your own cost-benefit ratio, because it's the one that matters.
Wait, I meant to type "Computers and Automation," not "Animation." Must have been a sort of geek freudian slip I guess.
This would also be a good time to remember Edmund C. Berkeley. He was an insurance executive (an actuary, I believe) who saw the commercial possibilities of the digital computer at a time when it was generally regarded as only an expensive military tool. He was instrumental in convincing Prudential to buy the Univac I. He then left the insurance industry and became the first advocate of computer education, developing some great logic toys (e.g., the Brainiac, the Geniac) and writing some great books for students (e.g., Giant Brains, Symbolic Logic and Intelligent Machines). He was one of the founding editors of Computers and Animation. Berkeley rocked.
Here is the issue in a nutshell. The world is full of kids who love to draw anime and comics and such, and some are quite good. The world will always be full of such people, millions and millions of them. But the industry does not need ten million anime artists, or even one million anime artists, or even twenty thousand. So there will always be someone good who is willing to work for peanuts. It's just like music: ninety-nine percent of musicians never rise above the bar band level, even though many are highly talented, but that doesn't stop them all from trying. In the popular arts, the supply of willing candidates will always be orders of magnitude beyond what the industry can absorb.
No, not with line numbers, and not with GOTOs. QBasic doesn't need them. If you teach it with some structure, and make sure that she declares her variables, she can have a total blast and get a feel for what programming is about. Fast and fun results will prevent loss of interest, which is probably the biggest threat to your project. Further, the knowledge that she gains will not be obsolete because the procedural statements are almost identical to VBScript. And the built-in help file is actually useful. Face it, you have to start with something fast and easy if you want to hold her interest.
Please try to take your own advice. . . . Sometimes I wish that PC monitors came with a big stick attached to them so that their operators could be thwacked about the head when they mail or post something completely inane.
Colin, pick up your telephone directory. Now turn to the business section. Now look up all of the businesses that start with a generic word like "Quality" or "National" or even a common last name like "McDonalds." If you live in a medium to large city, you will find many businesses that have similar names using generic words or common names. Multiply by all of the cities in your country. How do you suppose all of those businesses co-existed before the age of the Internet? How do you suppose McDonalds fast food and McDonalds car rental co-exist in your city today? What does any of this have to do with TLDs? That's why I'm asking you to stop and think. It's not a TLD issue. It's a trademark law issue.
Watch out for that stick attached to your monitor!
Here is my point: Take it out of the realm of domain names, and apply the same analysis that you would apply in the physical world. I open a bar called "Johnny's Bar." You open a bar down the street called "Johnny's Pub." Why does the anlysis change if we compare johnnys.bar to johnnys.pub? Or acme.com to acme.net? If the name is actually trademarked (like Kodak) then you can't do it anyway, if it isn't, well, there are a thousand bars out there than include the name "Johnny." And finally, if someone is actually committing fraud or an IP violation, that has nothing to do with the TLD process -- it's just fraud or theft. Those things are already unlawful.
And to carry the thought one step farther: if the domain name does NOT incorporate a trademarked name, such as when the word is generic, then that is exactly the kind of situation in which a lot of TLDs would actually be useful (bank.com, bank.net, bank.blah). So it works either way. If the name is trademarked, it doesn't matter if there are a lot of TLDs because you can shut down infringers, and if the name is not trademarked, it doesn't matter if there are a lot of TLDs because they don't infringe in each other. Where is the problem?
Forget domain names for a moment. Think generally. What stops anyone from choosing a business name that unlawfully incorporates another company's name? What stops anyone from creating the "Kodak Cafe" or the "Microsoft Bar and Grill"? The answer is: trademark law. Why isn't this enough? Why make such a big deal about trying to solve a problem that's already solved? Create all the TLDs that you want. I guarantee that if someone other than Kodak tries to register Kodak.blah, the registrant of Kodak.blah will be shut down. It's a non-issue.
And pardon me if you think that comments about submissions are off-topic, but once again, there are way too many hyperlinks in the submission. I do not need to know the web address of the supermarket chain's corporate headquarters, or the charity's corporate headquarters, or the event campaign's home page, or the sponsoring gallery's home page, or even the artist's home page. I just want to see the damned shopping cart helix. Pardon me for sounding like a curmudgeon, but nine times out of ten, I am only interested in one link: the link to the subject of the submission, not every related entity (which I can ferret out from the aricle if I really want to). Am I the only one who thinks so?
Out of curiousity, were you here then? I mean, your user ID is 600,000+. I was just wondering if it's a new account, and you used to post as someone else.
There is no reliable correlation between a poster's UID and whether a person was "here then." On one hand, some people (such as me) were reading Slashdot for years without having a UID, and on the other hand, many accounts with low UIDs seem to have been dormant for years. No reliable correlation.
Claude Shannon's theories underlie almost every aspect of the digital computer and digital communications. His master's thesis (1938) established that logic circuits can be simplified mathematically rather than by trial and error. His mathematical theory of communication (1948) established the entire field of information theory, making possible digital communications (modems, networks). In terms of his importance to the field, he is miles beyond most of the people on the list and most of the people already inducted.
Yes, I believe Moglen is wrong on this point. As a preliminary matter, any license requires a return promise from the licensee: not to violate the terms of the license. This is so basic it is almost circular. If you accept the benefits of the license, you must comply with the terms of the license. If Moglen is correct that the licensee has made no such promise, then how could any licensee ever be found to have committed a violation? Moglen is failing to distinguish express promises from implied promises. Sure, the licensee might not say "I promise," but the law will imply that the licensee said "I promise" when he accepts the benefits of the license.
The law.com definition of "license" also provides definitions for the parties to a license, i.e., the licensor and licensee. A "licensee" is defined as "a person given a license by the government or under private agreement." Further, according to law.com, the term "agreement", when used legally, is exactly synonymous with "contract": "in law, another name for a contract including all the elements of a legal contract: offer, acceptance, and consideration (payment or performance), based on specific terms." So the law.com definitions do support a connection between licensees and traditional contract law.
But I also see something in the law.com definition that may explain some of the controversy. There seem to be two sources of license, those granted by the government (e.g., driver's licenses, building permits) and those that arise by agreement between parties. Perhaps once of the sources of confusion is that some people regard the grant of rights under the GPL to be of the "governmental" type of license, because the underlying rights are based on copyright. My view, on the other hand, is that the grant is of the "agreement" type of license.
I hate to say it, but you're completely wrong. . . . The GPL is not a contract.
Pardon me for converting this to a meta-discussion, but your post provides an example of why the controversy will continue to rage until a court rules authoritatively on the question of whether or not the GPL is enforceable. There is no consensus even among knowledgeable people (indeed, even among knowledgeable lawyers) on the legal status of the GPL, which is why a test case is significant. But as a matter of basic contract law, a license is a form of contract and will be subject to a determination of its enforceability if one of these cases should proceed to a final resolution. Even when the terms of a contract are unilateral (i.e., not separately negotiated between the parties), the issues of contract formation and enforceability are still present (e.g., did the purported licensee accept the terms of the license, either expressly or by conduct?). These questions will not be sidestepped if this case proceeds to trial.
The GPL doesn't need tested, it's simple just copyright law . . .
Actually, the GPL does need to be tested if you want legal certainty that it is enforceable. The GPL is a license, not a copyright (that's what the "L" in GPL stands for). Licenses are contractual, and contracts are not merely enforceable on their own self-expressed authority. Contracts can be held unenforceable on a number of grounds -- lack of consent (i.e., you can't contractually bind a party that didn't actually enter the contract), lack of consideration (i.e., a contract requires that both sides give something up), violation of public policy (i.e., you can't enforce a contract that would require a party to commit an unlawful act), etc. Further, the GPL is an unusual license (in fact a unique license), so existing legal precedents do not provide a lot of predictive value. The fact that the GPL has not yet been fully adjudicated and found to be enforceable has been the source of considerable controversy.
Yes, you are correct. A preliminary injunction is not a final adjudication on the merits of the dispute. It is an interim remedy to stop the disputed activity only during the period while the case in pending. But because a preliminary injunction is a drastic remedy (it basically shuts down the defendant's activities until the case is resolved) a court will not grant one unless the plaintiff has enough evidence to demonstrate a strong likelihood of success at trial. So this means that the judge presumably reviewed the terms of the GPL, reviewed the defendant's activities, allowed the defendant to raise objections, weighed the evidence, and still granted the motion.
Stephenson makes great use of speculative history. He postulates some great "what if" scenarios arising from past events and uses them to weave an alternative present. He always succeeds in grabbing my attention. And then -- and then his male protagonist tries to talk to a woman. And that is where his novels fall apart. His dialog does not ring true. Every conversation sounds contrived. I think it's supposed to be banter, but it's just stilted. Is it any wonder he chose the name "Eliza" for the female protagonist in Quicksilver?
So the veneer glue can get some tooth into the substrate. With a wooden substrate, the glue penetrates the pores of the wood for a better bond. Otherwise, it tends to peel off when the piece is stressed.
I've worked with hardwood veneers on musical instruments. Two points: (1) you want a porous substrate so the veneers adhere properly; (2) you want a reasonably rigid (non-flexible) substrate so the veneers adhere properly. I can't image a worse substrate for fine veneer work than sheet metal. How intact will the veneers be after a year of normal use?
Someone with points to spare should mod up the parent post, since it is an important point an no one seems to have noticed it. Yes, this is the anniversary of the System/360 family of mainframes, which is a very significant anniversary for sure. But no, this is certainly not the 40th anniversary of the mainframe. The Univac I, II, III. The early Ferranti models. The IBM 700/7000 series. Lots more. Mainframes left and right, throughout the 1950s.
Last week there was a long Slashdot thread on the so-called "iPod Killer" proposed by Microsoft. That product was soundly critisized for missing the point -- the consensus among posters appeared to be that the product was too big and combined too many capabilities into one device, and thereby had lost the single-purpose elegance of the iPod. Now comes another product that appears to be conceptually similar to that Microsoft multi-purpose device. But this time it's being offered by a smaller company that a number of people like. I think it will be very interesting to compare the two threads. Was it just the words "iPod Killer" that prompted the strong opinions, or was it the underlying concept of a multi-purpose device?
For what I've seen, popup sales help is not particularly impressive. The "first level" of help appears to be essentially automated (basically a chatbot) that burns up several minutes establishing contact with an "Eliza-like" series of questions that you have to answer before you can get any further. Once you get to a real human, the level of support is not much different than what you would find if you searched the FAQ for the site (for all I could tell, it might still have been a bot except that I specifically asked if it was a person and then got a non-automated response). And I found it creepy to be accosted virtually, sort of like being stalked by a popup ad. Until they improve the interface and the substance, I'll pass.
Indeed, once a brand name enters the common parlance, it has a life independent of the company that it stands for, even if that company loses its leadership position.
Boss (handing you a stack of paper as he points to the Canon copier next to your desk): "Please xerox these documents."
Boss (handing you a stack of reference citations as he points to the Microsoft search engine on your desktop): "Please google these terms."
You might think it can't happen, but it can. The fact that Google is so dominant today is no guarantee of anything except that its name will probably remain recognizable as a verb for awhile. Google will have to continue to compete, and compete well, if it wants to stay on top. It was not very long ago that AltaVista ruled the search engine world, and it did not take very long for its user base to erode when Digital/Compaq failed to give it the priority it deserved.
They are also wrong when they say that there are companies that truly want users to hack their mass-market electronic products. Here is the killer counter-argument:
Name one manufacturer of a mass-market electronic product (including the Roomba described in the article) that will not instantly void your product warranty for doing the things described in this article.