In the US, most people with PhDs in Astrophysics are teaching math/physics to high schoolers for about $40,000-50,000 a year, or working in textbook publishing for about $45,000/year. And they're there because they're getting paid more than they would working at a college/university ($25,000 unless you're lucky enough to find a tenure-track job, then about $35,000) or for government labs.
Ironically, I know quite a few people with PhDs in Astrophysics. Four of my past co-workers (textbook industry)+ one of their husbands (gov't lab), my high school math teacher, and a couple of people that went back to school to learn CS and finally have decent-paying jobs.
But did you have utilities or an apartment lease prior to purchasing your house? Those can count too--although they're a bit more "generous" in their terms. Often, they have an entry in your credit report, but they don't report unless you do something pretty awful (refuse to pay hundreds of dollars to the phone company over many months; default on a lease).
Student loans also count, and can be a major positive if you've paid on time. I was told that one of the reasons my credit score was very high was that although I was carrying a heavy student loan debt, it had a positive history. The first thing that many twenty-somethings do is go into some type of default on their student loans, so the banks view not going into default as a sign you understand the concept of debt and understand the concept of repayment.
Ad. to the above: The RDSLAM I referred in the previous post is in the subdivision 1/2 mile to the west. The problems with the cable company is why I suspect the RDSLAM was probably pretty quickly overloaded with customers--1,500 homes, no cable internet. It went live before the other subdivison finally got cable access.
The road they need to cross was completely reconstructed about 24 months ago (as in, being closed for seven months, taken all the way down to the road bed and reconstructed; most of the utilities in the area were relaid/upgraded at that point). The town claims that Comcast was offered the opportunity to trench/cable during the reconstruction, but they refused to extend their network. That's the reason the crappy little cable company had its monopoly renewed a couple of months ago: the town actively went after Comcast service but were rebuffed.
(The town actively went after Comcast service because the crappy little cable company refused to cable up the new subdivisions, thus violating their contract; after being rebuffed by Comcast, the town had to take the crappy little cable company to court for everyone to get cable. I guess we should be glad we had cable when we moved in--the subdivision about 1/2 mile to the west of us, across the little creek, finally got cable service about five months ago. The oldest homes in that subdivision were about six years old at that point.)
That's bizarre: although Comcast has almost a virtual monopoly in our area, they refused to extend out to our town even when requested by town gov't (I believe their lines end just across the road from us ~500 feet away; the older homes in the unincorporated area of the next town all have Comcast service). Instead we get to pay for the hell which is a small cable company: $110/month for basic cable and internet.
And SBC/at&t refuses to allow us to connect to the RDSLAM they put in about a cable mile away--we didn't have phone service when the RDSLAM was put in, and for some reason (I'm guessing it got overloaded pretty fast; we called about a month after it was installed and they laughed at us) they're restricting it to homes that had phone service at the time when the DSLAM was installed. So, the neighbor who is 15 feet to the north of us can get DSL, but we can't.
What you really need to worry about is when companies Google-bomb the index in order to push "bad" info about their products 20-30+ pages down in the listings.
For example, a couple of months ago, my husband's sister called us to complain that her newish Dell has taken to running slowly, with close to 100% CPU utilization at all times. Our immediate response: malware. Our question: what did you install prior to this problem? Her answer: Party Poker.
I pop over to Google and search for "party poker" malware; "party poker" trojan; "party poker" adware; and various other combinations.
FOR TWENTY+ pages of results, every link returned was to a site praising or advertising Party Poker. The shit-page bombing of the Google index was complete. I think that I got to about the 50th page of results before I started to find any information on the malware that party poker installs and it still wasn't that great of information. GoogleGroups finally got me the info that pointed me in the right directions. I'd expect pages like this if I'd just searched on Party Poker. But for them to poison the index in such a way that searches on "Party Poker" in combination with words that people might be using once their system had been infested by that program--malware, trojan, adware--was very disturbing. Given that the GoogleGroups postings I found about it were both old (1 year+) and newish (a couple of months), this program has been a long-term issue. And now, someone who doesn't understand how the Google index can be poisoned isn't going to find information about what installing the program might do/have done to their system.
I don't know if the index is still poisoned in this way--as I said, it was a couple of months ago. But it has really made me look at Google differently, recently.
Unless they expected -- indeed counted upon -- the post invasion Iraq situation to be so rapturous that it would immediately erase any question of going in.
I believe it was this.
I think there were many people in the administration who saw Iraq as an "easy" war to win points at home and win ground in the Middle East.
But I also believe this because--at the time of 9/11, a coworker I was with on that day--was someone had connections to the Bush administration. As the towers went down, she stated: now we go to war with Iraq. We all looked at her in puzzlement and asked how she knew that Iraq was at fault. She said that it didn't matter, but that "everyone knew" it would be an easy war and good win for the administration.
She was a former energy lobbyist from D.C., BTW. So, add in the energy/oil aspect to the delusion: I think there was a belief that the war in Iraq would be "easy" and that the result would be a country with vast oil reserves under US control.
That, in itself, is mystifying to me, unless the administration believed it's own WMD propaganda.
No, I think the adminstration believed its own "dancing and singing" propaganda. The propaganda that stated that once the US eliminated Saddam, Iraq would fall in line as a good little US ally in the Middle East. It's been stated and proven that there were no post-invasion, worse-case plans, because the administration wouldn't believe there was that possibility.
And many MLS programs (Masters of Library Systems) are becoming increasing technical. Most MLS grads now graduate with basic computer science, basic relational database, and basic XML knowledge under their belt. There are also LIS (Library Information Systems) programs that cross over with computer science programs, specifically database systems, to the point of sharing classes with the CS people.
When it comes to the article in the encyclopedia or the Tom Clancy novel, COPYRIGHT and FAIR USE don't distinguish. They're both written sources of information. One is entertainment information and one is education information--but they're both written.
And one day, both the Tom Clancy novel and the article in the encyclopedia will enter into the public domain, should the public domain still exist at that point. And then, information DOES want to be free--that's the PURPOSE of the public domain.
There are pages that use the Window-Target header, and even some that use the Link header (though we haven't yet checked what for!). There are even some pages that include the Content-Style-Type header.
Wasn't creating a Window-Target HTTP header a trick for always breaking out of other people's frames (if someone links to your site and framed your site content within their own). I thought it was more reliable (back in 1999/2000) than the various JS tricks for breaking out of frames.
No, the problem is that it proves nothing except a possible reporting bias.
Someone with a brain tumor knows it's on the left side of his head. When asked which side he holds his cell phone on, he reports that he holds it on the left side of his head.
For those that report that they hold the cell phone on the left side of their head--if cell phones caused an increased risk of brain tumors--you would see an increased risk of left-side tumors and a STANDARD risk of right-side tumors. But what the study actually found was that there was an "increased" risk of left-side tumors and a "decreased" risk of right-side tumors with left-side cell phone holders.
What that indicates is that the reporting of which side of their head they have historically held the cell phone on may have become biased due to them knowing that there's a tumor in the left side of their head. They might have been right-side holders, but "recall" that they were left-side holders because, of course, everybody knows that holding the cell phone on one side of the head causes brain tumors. It's an indication of a possible self-reporting bias, rather than an actual connection.
So, basically, what the study said was that people with left-side tumors reported that they held the cell phone on the left side, while people with right-side tumors reported that they held the cell phone on the right side, WHETHER OR NOT it was reality. The decreased risk of other-side tumors indicates that it may not be reality--that the public assocation between brain tumors and cell phones causes a person to report that they held the cell phone on the side of their head with the tumor even if they did not.
I think things have worsened, but mainly because most low-level techs--male or female--are hired for their script reading skills and not their technical skills. So, the majority of the low-level technical support is useless . . . *sigh*
But yea, back when I did some technical support (on-campus, unpaid, at university) those who were most vocal about questioning my abilities for simply being a female were female. But that may have partly been because I was kicking most of the males' asses in our shared CS courses.
Actually, I could see a usage for this "ping" function that sort of makes me wish that it was implemented for all browsers . . .
I help run a distributed web site--our content is mirrored across multiple servers in multiple countries. To do statistics for that "site as a whole" requires statistics to be run on each server and then combined. Since the multiple servers are "owned/operated" by different people and there's nobody that has admin access on all, this doesn't happen. The last time we got accurate statistics for all our sites was back in 1999 or so.
It would be nice to have this "ping" functionality to ping a single log server each time a link is clicked, resulting in a single source to do statistics from. (The HTML pages are the same cross-site, so it could be easily done.) But some have mentioned this can be done via Javascript . . . per link? I probably should investigate that.
It depends on a lot of factors, many of which aren't addressed in this article.
If you want to provide full-text access/searching to the content on your site, but don't want to pay for the processing power required to do that on the back end, allowing Google to fully index your site may be valuable to you. And the trade-off--that the indexed content will also be avaliable via Google's primary search engine, not just the search box on your site--may not be a concern.
If you have reasons why you don't want the full indexing/searching of your site avaliable via a service that is not under your control, then that might be a trade off you don't want to make. I work on a web site were we DON'T want that, for a whole variety of reasons. But we've had the meta headers and robots.txt tuned and in place since 1997, because we knew we needed them.
Usually, the problem with plagiarism is that copying the material was illegal.
No. Plagiarism is a concept concerning moral rights, while copyright is a legal term concerned with assigned rights. Something may both be copyright infringement and plagiarism, or may be just copyright infringement, or may just be plagiarism.
If I republish a newspaper article in my blog and claim that I wrote it, it is both copyright infringement and plagiarism. I have violated US copyright law, and I have violated the original author's moral rights (to receive credit for what he/she created).
If I copy that article 50,000 times and distribute it as part of a sales package without the permission of the original publishing source/creator, it is copyright infringement but not plagiarism (as long as the author's byline remains on the article and I'm not trying to claim that I wrote it myself).
If I attempt to republish a novel by Mark Twain and claim that I wrote it, it's plagiarism BUT NOT copyright infringement. Mark Twain's work is in public domain, so there are NO legal rights assigned to that work any longer. But there are moral rights--the ongoing right of Mark Twain/Sam Clemens to have his name attached to the words he wrote/novels he created. Moral rights-only infringement can only be punished socially, not legally. (Although, "social" punishment can have lasting effects--explusion from school, firing from a job, are all "social" punishments under this model.)
When copyleft is brought in, the whole issue becomes a little more complicated. Copyleft is not public domain (case 3 above). The copier's rights under copyleft are defined by which copyleft scheme is being licensed by the originator/creator (there are several, and at least one creative commons license even allows the release from moral rights).
As a followup to my other post: what I'm trying to say is that I find the Britannica versus Wikipedia arguments both useless and amusing. Not because Wikipedia is (or should be acknowledged to be) that good a source of information, but because--within K-12 education and academia--Britannica is widely acknowledged to be just about as academically useless a resource as Wikipedia.
the Wikipedia articles that aren't as popular and aren't as well-researched
A lot of those aren't even going to be in Britannica. No, Wikipedia isn't authoritative from an academic standpoint, but neither is Britannica once you get past the 4th grade. In reality, they're just about equal in the academic viewpoint: they're both pretty much a pile of crap.
(Or, the 4th grade was the last time I was allowed to use any of the Britannica-like sources for much of anything in school.)
Secondly, Britannica has much better fact checking than Wikipedia.
Actually, having known people who have worked for Britannica, that's not quite true. None of them were surprised by the recent research that showed that Wikipedia's accuracy and Britannica's accuracy were pretty similar on well-researched and popular topics.
(In fact, at least one of the said something along the lines that *only* finding three major errors per article in Britannica was pretty damn good.;)
And as I said below, I think it's nothing more than a pretty "dumb" computer algorithm using assigned keywords to link together an unfortunate collection of DVDs.
If you've seen Planet of the Apes, and were a data-entry person putting that DVD into the Wal-mart database, and were presented with an internal list of keywords to assign to that movie which included a keyword along the lines of "race relations" or "black/white relations," would you chose that keyword to assign to Planet of the Apes?
Since, to me, Planet of the Apes is nothing more than a pretty poorly done allegory for "race relations in the United States," I probably would. And I wouldn't expect a $10/hour data entry person to realize the further effects that choice might have on the online recommendations system. Hell, both helped design keyword-based systems like this, and having done data entry for those type of systems, I'm not sure that *I* would recognize the wierdnesses that *could* arise from that assignment.
From a data control prespective, that's when you need further constraints built into the system (for example, if a DVD contains both "race relations" and "nonfiction" as keywords, maybe you only want to return keyword matches/recommendations that contain BOTH "race relations" and "nonfiction"; that would have avoided this particular mess without removing what's probably an appropriate keyword from either Planet or the MLK documentaries).
Based on what's been said in a couple of other comments, it sounds like the Wal-mart recommendations system operates via a set of assigned keywords or metadata. Having worked on a similar type of system in the past--they probably have a defined set of keywords that can be assigned for each DVD. When the DVD is entered into the database, it is the data entry person's responsibility to chose a proper set of keywords for the DVD.
This type of system develops strange biases in several ways, most noteably through human interpretation. Say you have a keyword "black/white relations." One data entry person might only assign that keyword to nonfictional documentaries, while another might assign that keyword to based-on-real-life movies as well. And another person who's particularly sensitive to the underlying messages of movies might assign that keyword to Planet of the Apes (as well as possibly to box collections of ST:TOS).
Somebody selects one of those movies, and gets a bizarre selection of "related" movies which simply reflects the fact that three different people viewed the use of a defined keyword and thus assigned it in three different ways. It's hard to even design business rules to prevent this from happening because it overly limits what the system was designed to do. If a business rule says that only nonfiction documentaries and based-on-real-life movies can receive the "black/white relations" tag, you might end up missing a movie like Crash. If the business rule says instead that you can't assign a tag based on the "underlying" message of a movie, how do you define underlying message? Racism or "black/white relations" (my bet is that the Wal-mart keyword was closer to "black/white relations" rather than "racism" because all the movies that apparently popped up as suggestions were about that particular subset of racism) is the in-your-face message of Planet of the Apes. It's so thinly masked by the story that I'm not sure I'd define it was the "underlying" message. I'm the type of person that probably *would* assign Planet to the "black/white relations" tag, because its consideration of that theme is about the only redeeming factor of the movie.
Of course, I grew in an area where--due to integration--racism was a pretty major issue and I thought I'd learned most of the various "bad" terms that members of one race (hell, one European background) called members of another race (or other European backgrounds) when I was young. "Monkey" had definitely fallen out of use in my area by the 1980s; first time I was ever introduced to it as a racist term was online about four years ago.
Slashdot Mirror
Actually, women who are pregnant during the prime flu months need to hide from their doctors to avoid being stuck with the flu shot.
I lied, and told her that I'd had it already. I don't do flu shots--despite repeated exposures, I haven't had the flu in 10+ years.
In the US, most people with PhDs in Astrophysics are teaching math/physics to high schoolers for about $40,000-50,000 a year, or working in textbook publishing for about $45,000/year. And they're there because they're getting paid more than they would working at a college/university ($25,000 unless you're lucky enough to find a tenure-track job, then about $35,000) or for government labs.
Ironically, I know quite a few people with PhDs in Astrophysics. Four of my past co-workers (textbook industry)+ one of their husbands (gov't lab), my high school math teacher, and a couple of people that went back to school to learn CS and finally have decent-paying jobs.
But did you have utilities or an apartment lease prior to purchasing your house? Those can count too--although they're a bit more "generous" in their terms. Often, they have an entry in your credit report, but they don't report unless you do something pretty awful (refuse to pay hundreds of dollars to the phone company over many months; default on a lease).
Student loans also count, and can be a major positive if you've paid on time. I was told that one of the reasons my credit score was very high was that although I was carrying a heavy student loan debt, it had a positive history. The first thing that many twenty-somethings do is go into some type of default on their student loans, so the banks view not going into default as a sign you understand the concept of debt and understand the concept of repayment.
Ad. to the above: The RDSLAM I referred in the previous post is in the subdivision 1/2 mile to the west. The problems with the cable company is why I suspect the RDSLAM was probably pretty quickly overloaded with customers--1,500 homes, no cable internet. It went live before the other subdivison finally got cable access.
The road they need to cross was completely reconstructed about 24 months ago (as in, being closed for seven months, taken all the way down to the road bed and reconstructed; most of the utilities in the area were relaid/upgraded at that point). The town claims that Comcast was offered the opportunity to trench/cable during the reconstruction, but they refused to extend their network. That's the reason the crappy little cable company had its monopoly renewed a couple of months ago: the town actively went after Comcast service but were rebuffed.
(The town actively went after Comcast service because the crappy little cable company refused to cable up the new subdivisions, thus violating their contract; after being rebuffed by Comcast, the town had to take the crappy little cable company to court for everyone to get cable. I guess we should be glad we had cable when we moved in--the subdivision about 1/2 mile to the west of us, across the little creek, finally got cable service about five months ago. The oldest homes in that subdivision were about six years old at that point.)
That's bizarre: although Comcast has almost a virtual monopoly in our area, they refused to extend out to our town even when requested by town gov't (I believe their lines end just across the road from us ~500 feet away; the older homes in the unincorporated area of the next town all have Comcast service). Instead we get to pay for the hell which is a small cable company: $110/month for basic cable and internet.
And SBC/at&t refuses to allow us to connect to the RDSLAM they put in about a cable mile away--we didn't have phone service when the RDSLAM was put in, and for some reason (I'm guessing it got overloaded pretty fast; we called about a month after it was installed and they laughed at us) they're restricting it to homes that had phone service at the time when the DSLAM was installed. So, the neighbor who is 15 feet to the north of us can get DSL, but we can't.
What you really need to worry about is when companies Google-bomb the index in order to push "bad" info about their products 20-30+ pages down in the listings.
For example, a couple of months ago, my husband's sister called us to complain that her newish Dell has taken to running slowly, with close to 100% CPU utilization at all times. Our immediate response: malware. Our question: what did you install prior to this problem? Her answer: Party Poker.
I pop over to Google and search for "party poker" malware; "party poker" trojan; "party poker" adware; and various other combinations.
FOR TWENTY+ pages of results, every link returned was to a site praising or advertising Party Poker. The shit-page bombing of the Google index was complete. I think that I got to about the 50th page of results before I started to find any information on the malware that party poker installs and it still wasn't that great of information. GoogleGroups finally got me the info that pointed me in the right directions. I'd expect pages like this if I'd just searched on Party Poker. But for them to poison the index in such a way that searches on "Party Poker" in combination with words that people might be using once their system had been infested by that program--malware, trojan, adware--was very disturbing. Given that the GoogleGroups postings I found about it were both old (1 year+) and newish (a couple of months), this program has been a long-term issue. And now, someone who doesn't understand how the Google index can be poisoned isn't going to find information about what installing the program might do/have done to their system.
I don't know if the index is still poisoned in this way--as I said, it was a couple of months ago. But it has really made me look at Google differently, recently.
Statistically, your daughter has a higher chance of being attacked by a peer at school. Do you homeschool?
I believe it was this.
I think there were many people in the administration who saw Iraq as an "easy" war to win points at home and win ground in the Middle East.
But I also believe this because--at the time of 9/11, a coworker I was with on that day--was someone had connections to the Bush administration. As the towers went down, she stated: now we go to war with Iraq. We all looked at her in puzzlement and asked how she knew that Iraq was at fault. She said that it didn't matter, but that "everyone knew" it would be an easy war and good win for the administration.
She was a former energy lobbyist from D.C., BTW. So, add in the energy/oil aspect to the delusion: I think there was a belief that the war in Iraq would be "easy" and that the result would be a country with vast oil reserves under US control.
That, in itself, is mystifying to me, unless the administration believed it's own WMD propaganda.
No, I think the adminstration believed its own "dancing and singing" propaganda. The propaganda that stated that once the US eliminated Saddam, Iraq would fall in line as a good little US ally in the Middle East. It's been stated and proven that there were no post-invasion, worse-case plans, because the administration wouldn't believe there was that possibility.
My cholestorol is 101, and the hubby's is under 100. (Yes, the COMBINED level). And we're both carnivores.
And many MLS programs (Masters of Library Systems) are becoming increasing technical. Most MLS grads now graduate with basic computer science, basic relational database, and basic XML knowledge under their belt. There are also LIS (Library Information Systems) programs that cross over with computer science programs, specifically database systems, to the point of sharing classes with the CS people.
When it comes to the article in the encyclopedia or the Tom Clancy novel, COPYRIGHT and FAIR USE don't distinguish. They're both written sources of information. One is entertainment information and one is education information--but they're both written.
And one day, both the Tom Clancy novel and the article in the encyclopedia will enter into the public domain, should the public domain still exist at that point. And then, information DOES want to be free--that's the PURPOSE of the public domain.
Wasn't creating a Window-Target HTTP header a trick for always breaking out of other people's frames (if someone links to your site and framed your site content within their own). I thought it was more reliable (back in 1999/2000) than the various JS tricks for breaking out of frames.
No, the problem is that it proves nothing except a possible reporting bias.
Someone with a brain tumor knows it's on the left side of his head. When asked which side he holds his cell phone on, he reports that he holds it on the left side of his head.
For those that report that they hold the cell phone on the left side of their head--if cell phones caused an increased risk of brain tumors--you would see an increased risk of left-side tumors and a STANDARD risk of right-side tumors. But what the study actually found was that there was an "increased" risk of left-side tumors and a "decreased" risk of right-side tumors with left-side cell phone holders.
What that indicates is that the reporting of which side of their head they have historically held the cell phone on may have become biased due to them knowing that there's a tumor in the left side of their head. They might have been right-side holders, but "recall" that they were left-side holders because, of course, everybody knows that holding the cell phone on one side of the head causes brain tumors. It's an indication of a possible self-reporting bias, rather than an actual connection.
So, basically, what the study said was that people with left-side tumors reported that they held the cell phone on the left side, while people with right-side tumors reported that they held the cell phone on the right side, WHETHER OR NOT it was reality. The decreased risk of other-side tumors indicates that it may not be reality--that the public assocation between brain tumors and cell phones causes a person to report that they held the cell phone on the side of their head with the tumor even if they did not.
I think things have worsened, but mainly because most low-level techs--male or female--are hired for their script reading skills and not their technical skills. So, the majority of the low-level technical support is useless . . . *sigh*
But yea, back when I did some technical support (on-campus, unpaid, at university) those who were most vocal about questioning my abilities for simply being a female were female. But that may have partly been because I was kicking most of the males' asses in our shared CS courses.
Actually, I could see a usage for this "ping" function that sort of makes me wish that it was implemented for all browsers . . .
I help run a distributed web site--our content is mirrored across multiple servers in multiple countries. To do statistics for that "site as a whole" requires statistics to be run on each server and then combined. Since the multiple servers are "owned/operated" by different people and there's nobody that has admin access on all, this doesn't happen. The last time we got accurate statistics for all our sites was back in 1999 or so.
It would be nice to have this "ping" functionality to ping a single log server each time a link is clicked, resulting in a single source to do statistics from. (The HTML pages are the same cross-site, so it could be easily done.) But some have mentioned this can be done via Javascript . . . per link? I probably should investigate that.
It depends on a lot of factors, many of which aren't addressed in this article.
If you want to provide full-text access/searching to the content on your site, but don't want to pay for the processing power required to do that on the back end, allowing Google to fully index your site may be valuable to you. And the trade-off--that the indexed content will also be avaliable via Google's primary search engine, not just the search box on your site--may not be a concern.
If you have reasons why you don't want the full indexing/searching of your site avaliable via a service that is not under your control, then that might be a trade off you don't want to make. I work on a web site were we DON'T want that, for a whole variety of reasons. But we've had the meta headers and robots.txt tuned and in place since 1997, because we knew we needed them.
*snicker*
;)
And some of us even can do it without the pretty IDE!
No. Plagiarism is a concept concerning moral rights, while copyright is a legal term concerned with assigned rights. Something may both be copyright infringement and plagiarism, or may be just copyright infringement, or may just be plagiarism.
If I republish a newspaper article in my blog and claim that I wrote it, it is both copyright infringement and plagiarism. I have violated US copyright law, and I have violated the original author's moral rights (to receive credit for what he/she created).
If I copy that article 50,000 times and distribute it as part of a sales package without the permission of the original publishing source/creator, it is copyright infringement but not plagiarism (as long as the author's byline remains on the article and I'm not trying to claim that I wrote it myself).
If I attempt to republish a novel by Mark Twain and claim that I wrote it, it's plagiarism BUT NOT copyright infringement. Mark Twain's work is in public domain, so there are NO legal rights assigned to that work any longer. But there are moral rights--the ongoing right of Mark Twain/Sam Clemens to have his name attached to the words he wrote/novels he created. Moral rights-only infringement can only be punished socially, not legally. (Although, "social" punishment can have lasting effects--explusion from school, firing from a job, are all "social" punishments under this model.)
When copyleft is brought in, the whole issue becomes a little more complicated. Copyleft is not public domain (case 3 above). The copier's rights under copyleft are defined by which copyleft scheme is being licensed by the originator/creator (there are several, and at least one creative commons license even allows the release from moral rights).
As a followup to my other post: what I'm trying to say is that I find the Britannica versus Wikipedia arguments both useless and amusing. Not because Wikipedia is (or should be acknowledged to be) that good a source of information, but because--within K-12 education and academia--Britannica is widely acknowledged to be just about as academically useless a resource as Wikipedia.
the Wikipedia articles that aren't as popular and aren't as well-researched
A lot of those aren't even going to be in Britannica. No, Wikipedia isn't authoritative from an academic standpoint, but neither is Britannica once you get past the 4th grade. In reality, they're just about equal in the academic viewpoint: they're both pretty much a pile of crap.
(Or, the 4th grade was the last time I was allowed to use any of the Britannica-like sources for much of anything in school.)
Actually, having known people who have worked for Britannica, that's not quite true. None of them were surprised by the recent research that showed that Wikipedia's accuracy and Britannica's accuracy were pretty similar on well-researched and popular topics.
(In fact, at least one of the said something along the lines that *only* finding three major errors per article in Britannica was pretty damn good. ;)
And as I said below, I think it's nothing more than a pretty "dumb" computer algorithm using assigned keywords to link together an unfortunate collection of DVDs.
If you've seen Planet of the Apes, and were a data-entry person putting that DVD into the Wal-mart database, and were presented with an internal list of keywords to assign to that movie which included a keyword along the lines of "race relations" or "black/white relations," would you chose that keyword to assign to Planet of the Apes?
Since, to me, Planet of the Apes is nothing more than a pretty poorly done allegory for "race relations in the United States," I probably would. And I wouldn't expect a $10/hour data entry person to realize the further effects that choice might have on the online recommendations system. Hell, both helped design keyword-based systems like this, and having done data entry for those type of systems, I'm not sure that *I* would recognize the wierdnesses that *could* arise from that assignment.
From a data control prespective, that's when you need further constraints built into the system (for example, if a DVD contains both "race relations" and "nonfiction" as keywords, maybe you only want to return keyword matches/recommendations that contain BOTH "race relations" and "nonfiction"; that would have avoided this particular mess without removing what's probably an appropriate keyword from either Planet or the MLK documentaries).
Based on what's been said in a couple of other comments, it sounds like the Wal-mart recommendations system operates via a set of assigned keywords or metadata. Having worked on a similar type of system in the past--they probably have a defined set of keywords that can be assigned for each DVD. When the DVD is entered into the database, it is the data entry person's responsibility to chose a proper set of keywords for the DVD.
This type of system develops strange biases in several ways, most noteably through human interpretation. Say you have a keyword "black/white relations." One data entry person might only assign that keyword to nonfictional documentaries, while another might assign that keyword to based-on-real-life movies as well. And another person who's particularly sensitive to the underlying messages of movies might assign that keyword to Planet of the Apes (as well as possibly to box collections of ST:TOS).
Somebody selects one of those movies, and gets a bizarre selection of "related" movies which simply reflects the fact that three different people viewed the use of a defined keyword and thus assigned it in three different ways. It's hard to even design business rules to prevent this from happening because it overly limits what the system was designed to do. If a business rule says that only nonfiction documentaries and based-on-real-life movies can receive the "black/white relations" tag, you might end up missing a movie like Crash. If the business rule says instead that you can't assign a tag based on the "underlying" message of a movie, how do you define underlying message? Racism or "black/white relations" (my bet is that the Wal-mart keyword was closer to "black/white relations" rather than "racism" because all the movies that apparently popped up as suggestions were about that particular subset of racism) is the in-your-face message of Planet of the Apes. It's so thinly masked by the story that I'm not sure I'd define it was the "underlying" message. I'm the type of person that probably *would* assign Planet to the "black/white relations" tag, because its consideration of that theme is about the only redeeming factor of the movie.
Of course, I grew in an area where--due to integration--racism was a pretty major issue and I thought I'd learned most of the various "bad" terms that members of one race (hell, one European background) called members of another race (or other European backgrounds) when I was young. "Monkey" had definitely fallen out of use in my area by the 1980s; first time I was ever introduced to it as a racist term was online about four years ago.