While I praise this article for pointing out the fact that AOL's filtering software is not at the level it should be, I think any accusations of outright liberal bias are premature and ignore the fundamental differences between whitelist and blacklist filtering.
Whitelist filtering requires that the filter provider (e.g. AOL) add the site to a list of approved sites. i.e. rather than specifically 'banning' sites that kids can visit, the whitelist method approves sites. Obviously, in order for a site to make it onto this list, a site has to be reviewed. The fact that there aren't as many liberal sites on the whitelist may only go to show a much larger weakness with whitelists, namely, that you can't give the user every acceptable site on the Internet. I could probably test the filtering software and turn up some assinine conclusion like 'there is a clear bias towards Coke over Pepse b/c my favorite Pepsi sites weren't on the whitelist.' There is simply too many sites for any 'human' means of review. If there are complaints about not enough liberal sites being on the whitelist, try submitting them to AOL and see if they get reviewed and added (and maybe this article will be the impetus for that). Until AOL blantantly rejects liberal sites from being added, it's hard to accuse them of any conspiracy to 'republicanize' our youth.
All arguments about the appropriateness of filters, whitelist filtering is the only filtering option that really has any promise for establishing the goal of a safe sandbox for kids to surf the Internet in. Blacklist filtering techniques will often either (a) miss inappropriate sites or (b) ban appropriate sites because their spider turned up the word 'sex'. Whitelist filtering does not have either of these weaknesses. But if Yahoo's indexers can't keep up with the growth of the Internet, I doubt any whitelist filtering company could either.
If Microsoft actually makes the source available to more than a few select people it will escape.
I don't necessarily agree with this point. Microsoft has _thousands_ of developers already working on the Windows code base (much more than a few select people), yet the code has not "escaped." Yes, this is a possibility when another company is allowed to develop it's own Windows product, but that company also has the same interest as Microsoft in keeping that code base secret (probably b/c they'll have to pay a large amount of money to get the license). First off, you don't allow any developer full access to the code, and second, you tightly monitor your office environment. For example, at the Microsoft office my friend worked at, an alarm would go off if you connected the Microsoft internal LAN to an outside Internet connection.
From my interpretation, it's not "open source" per se. What would happen is that Microsoft would sell of licenses to use their source code for Windows. Corporations (not individuals) could purchase these licenses and use them to market their own branded version of windows.
This wouldn't be much more of a security risk because you would actually have to purchase a very expensive license to look at the code. Your average hacker/cracker still would not be able to get their hands on the code. Keep in mind that Microsoft has licensed it's code out before to other companies, though not to sell their own version of Windows, without the rest of the world being able to view it.
Thus, to summarize, I think this could only improve the security of Windows, as many more developers would be poring over the code. Also, the security risks would be minimized because you would actually have to be a developer at one of the licensees in order to view the source code. Then, eventually, the Windows code base could reach a point where it could be completely open sourced without the entire Windows population suffering a host of security exploits.
Since it appears that Tech Review didn't post the article online, here's the full text (it's just a little text box at the bottom of a page). Please excuse any transcription errors.
Geek Show
Listen closely and you can hear it: the gentle sound of millions of folks nervously scratching their heads wondering just what the heck is going on with the open-source, or "free" sofware that has suddenly gotten so much attention (see "Programs to the People," TR January/February 1999). There's no better source of information, rumor and opinion on this phenomenon than Slashdot, which titles itself: "New for Nerds. Stuff that Matters."
The creation of 22-year-old Rob Malda -- whose nom de Web is "Commander Taco" -- Slashdot feels like a geek clubhouse. Malda and others post articles on topics ranging from "the personalities behind Linux" to an essay predicting the arrival of "sexbots" that never have headaches.
Anti-Microsoft sentiment permeates the site; these people know their code and they're offended by what many perceive to be bloated, inefficient software. The "Ask Slashdot" part of the site ranges beyond programming, as knowledgeable participants weigh in on quirky questions like, "What is the bandwidth of a nerve?" (Answer: Well, it depends on what you mean by "bandwidth" and "nerve").
This is one of those virtual communities you hear so much about. Articles generate a torrent of (mostly) well-informed commentary, much of it signed by "Anonymous Coward" -- the epithet Slashdot assigns to contributors unwilling to reveal their names. In one persistent thread, commentators proudly quantify what they call the "Slashdot Effect": the spike in the hit count experienced by Web pages that Slashdot links to. Look to this site as your technical and cultural guide to a powerful movement that is starting to emerge from the computer underground"
"...A built-in digital signal processor compares those images and translates changes into on-screen pointer movements. This technique, called image correlation processing, executes 18 million instructions per second (MIPS) and results in smoother, more precise pointer movement. Current mice execute only about 1.5 MIPS, making a mouse featuring Microsoft IntelliEye about 12 times smarter than ordinary mice."
What's up with this Intelli-blah stuff though? Intellieye, Intellimouse, Intellisense, etc... Is M$ afraid that their technologies would otherwise be perceived as stupid?
Slashdot Mirror
My server can't handle the traffic, but the videos are posted on Google Video:
D-Wave video part 1
D-Wave video part 2
Whitelist filtering requires that the filter provider (e.g. AOL) add the site to a list of approved sites. i.e. rather than specifically 'banning' sites that kids can visit, the whitelist method approves sites. Obviously, in order for a site to make it onto this list, a site has to be reviewed. The fact that there aren't as many liberal sites on the whitelist may only go to show a much larger weakness with whitelists, namely, that you can't give the user every acceptable site on the Internet. I could probably test the filtering software and turn up some assinine conclusion like 'there is a clear bias towards Coke over Pepse b/c my favorite Pepsi sites weren't on the whitelist.' There is simply too many sites for any 'human' means of review. If there are complaints about not enough liberal sites being on the whitelist, try submitting them to AOL and see if they get reviewed and added (and maybe this article will be the impetus for that). Until AOL blantantly rejects liberal sites from being added, it's hard to accuse them of any conspiracy to 'republicanize' our youth.
All arguments about the appropriateness of filters, whitelist filtering is the only filtering option that really has any promise for establishing the goal of a safe sandbox for kids to surf the Internet in. Blacklist filtering techniques will often either (a) miss inappropriate sites or (b) ban appropriate sites because their spider turned up the word 'sex'. Whitelist filtering does not have either of these weaknesses. But if Yahoo's indexers can't keep up with the growth of the Internet, I doubt any whitelist filtering company could either.
I don't necessarily agree with this point. Microsoft has _thousands_ of developers already working on the Windows code base (much more than a few select people), yet the code has not "escaped." Yes, this is a possibility when another company is allowed to develop it's own Windows product, but that company also has the same interest as Microsoft in keeping that code base secret (probably b/c they'll have to pay a large amount of money to get the license). First off, you don't allow any developer full access to the code, and second, you tightly monitor your office environment. For example, at the Microsoft office my friend worked at, an alarm would go off if you connected the Microsoft internal LAN to an outside Internet connection.
From my interpretation, it's not "open source" per se. What would happen is that Microsoft would sell of licenses to use their source code for Windows. Corporations (not individuals) could purchase these licenses and use them to market their own branded version of windows.
This wouldn't be much more of a security risk because you would actually have to purchase a very expensive license to look at the code. Your average hacker/cracker still would not be able to get their hands on the code. Keep in mind that Microsoft has licensed it's code out before to other companies, though not to sell their own version of Windows, without the rest of the world being able to view it.
Thus, to summarize, I think this could only improve the security of Windows, as many more developers would be poring over the code. Also, the security risks would be minimized because you would actually have to be a developer at one of the licensees in order to view the source code. Then, eventually, the Windows code base could reach a point where it could be completely open sourced without the entire Windows population suffering a host of security exploits.