And stop bitching about XBOX just because it comes from Microsoft. Look past the freakin' name for once and see that MS might just have something good on their hands.
While I won't disagree that there're some people bitching over the MS issue, game console preference seem to be as equally religiously charged as text editor preference. You'll find numerous people all claiming how great the suckitude is of any current (or even unreleased) game system from the PS2 to the N64 to the Dreamcast to the GameCube to the XBOX. It's just a given.
However, it's worth pointing out that if I ignore the anti-Microsoft bias, all I'm left with is the fact that it's a game console from a company that hasn't produced a console before. That's not to say they can't pull it off -- just that there's nothing to get me downright excited until I reach out and touch one.
But do you really need Microsoft Word? Or do you need a functional word processor? There is a world of difference between the two.
It's somewhere in-between -- I need a functional word processor that's capable of properly reading and writing Microsoft Word documents. I am required to deal with them as part of my job. Ideally, everyone would use PDF or, even better, plain text. But that isn't the world I live in.
Others have already pointed out the game issue. I think what Loki is doing is great, but for the most part, there is a significant lack of games for Linux.
Finally, despite all the effort being put into Gnome and KDE, it just doesn't have a nice, unified GUI "feel" to the desktop. Most of my Linux use is done on the text console -- X is used only for websites that require Netscape, and for the occasions when I need to fight with StarOffice.
Just out of curiosity, what's missing that you need?
Quality. When using StarOffice 5.2, I was running into significant bugs every 5 minutes or so. Microsoft may have their own share of bugs, but StarOffice makes 'em look downright perfect in comparison.
Just don't install Windows XP. I don't use Windows anything. So I'm not succeptible.
Well good for you. Unfortunately, those of us not living in never-never land have discovered that, contrary to popular belief, there's a lot of software out there that hasn't been ported to Linux.
Don't get me wrong -- I love Linux. I use it both at home and at work. It's been my general operating system of choice for about 6 years now. However, that doesn't change the fact that there're some computer-related activites for which Windows is the better answer, either because more commercial developers supports it or the software requires less effort to install-and-use.
So until Linux fully addresses these issues (something which, as much as I'm a fan of Linux, I doubt will every fully happen), there will be a need for some people to install and use Windows. Claiming otherwise is at best misguided advocacy and at worst trolling.
If so, will that even hold water? Barney is a name. It's also commonly given to animals.
Some random (presumably trademarked) common names/terms:
Apple Computers
Apple Records
Intel (think spy movies)
Denny's
Wendy's
McDonald's
Chevrolet Corvettes
Think of trademarks kind of like a (more properly run) set of domain names. You have to be actively using it to claim it. You only get an entry in the TLDs that directly relate to your product. So "wendys.food" is exclusive to the Wendy's corporation, but wouldn't have much of a claim if Wendy Johnson wanted to open "Wendy's Car Repair".
In general, trademarks seem to be (at least in my understanding) one of the better-run intellectual property things. I think they just get a bad rap due to the unfortunate interaction between trademarks and domain registries (where all commercial entities have been more-or-less lumped together in.com).
But you'd have been right if you'd said something like "AmigaOS and Windows NT are not competitors either, they run on different hardware."
But that assumes that someone picks the hardware first and then picks the operating system. Often times, this isn't the case. Most consumers, for example, wouldn't buy a computer that's three times as fast as an Intel-based machine of the same price, but that doesn't run Windows.
you don't want to be bothered with such things when trying to write a quick one-off script.
However, there are more uses for a scripting language than just quick one-off scripts. As mentioned in the article blurb, this language can be embedded within your own applications to provide scripting support.
"Password" does not necessarily need to be a "stupid" password.
If you need significant security, it's a stupid password because it's guessable. If you don't need significant security, it's a stupid password because it's (relatively) long. You might as well go with "1234". It's equally guessable, but more than twice as easy to type (if you factor in both length and the ease of typing something sequential).
I suspect the relevant scenario might be more like:
Buy tokens with credit card. Play with tokens. Lose. Dispute charges on credit card bill.
With the new law, the company can't turn around and fight the fact that you're disputing the charges. They can just cancel your account and blacklist your credit card. Given that some players might win, it's going to be a huge headache for the casinos.
So the casinos can either take the risk of credit cards (depending on how often people pull that stunt) or they can require tokens to be purchased with a check-equivilant (which lowers the convenience and impulse factor). Overall, the law certainly doesn't seem to prevent Australians from online, off-shore gambling (although the article implied it did), but (unless I've misinterpreted the credit card/legal interation) it does put a damper on the casinos' profitability.
They are not routable on the net, but they _are_ routable on gnutella via push messages.
Unless both machines have unroutable IP addresses. My work machine uses 10.0.0.3, but gets masqueraded through a Linux box. As such, it generally can only initiate connections, but it can't receive them. (Presumably, the push message tells such a machine to initiate a connection to the client, to get around this limitation.) Now someone else who's in the same situation wouldn't be able to connect with me, since neither of our machines could be the one to receive a new, incoming connection. As such, it'd be nice if the client (I use limewire) were smart enough to let me filter out all the non-routable IP addresses. I know it can recognize them, as it highlights them in red. But I found nothing in the options that would let me auto-ignore what're more-or-less worthless search results.
No. This has nothing to do with musicians. Record companies already distribute musical recordings in a digital form. Furthermore, I can't imagine them not including "online distribution rights" within the contract that the musicians must sign when signing with a record label. Finally, no one has questioned whether or not the record companies are in control of the electronic distribution rights of songs. Instead, the two main issues have been whether private users have a valid fair use exception to allow P2P file sharing and whether the RIAA has the right to directly sue Napster for the actions of people using their system (especially since the file gets sent directly from one client to another).
In this case, it's talking about organizations that have specifically paid a freelancer in the past for the right to use that person's work. At the time, electronic distribution was not considered in the contract. It's not that the news agencies were attempting to distribute content that they weren't allowed to distribute. Rather, they were trying to distribute it via a new medium. (As opposed to the musician issue, where even before Napster, I couldn't sit on a street corner and give out illegal copies of a commercially-produced tape to every car that drove by.)
I can see toasters and coffepot weeb cams with IP numbers. but light bulbs?
They didn't say IP numbers. They said URLs. If every image, page, frameset, and CGI script on the web can each have their own URL, providing one to each lightbulb isn't going to be that difficult:
So they aren't necessarily doing things the smart way. But my guess is that they use something simpler than TCP/IP for communicating to the actual bulb sockets (unless this is one of those projects that is just mindlessly burning money). In reality, it wouldn't surprise me if the URLs turned out to actually be something closer to:
I have no doubt that one day AI will come to pass. [...] I *DO* have doubts that Cyc will be at all related to this outcome.
To me, it seems that Cyc is working on the second step of creating a "true" AI -- first you build it, then you teach it. While the scientists may be teaching something that's more just a smoke-and-mirrors database, that doesn't mean that Cyc wouldn't be an effective training tool to use to teach a true AI. The knowledge being placed within Cyc is a useful part of thought, even if it isn't inherently self-aware.
In the same way you know the content of a book by reading the cover?
...which doesn't work too well for newsgroups. Sure it looks like there are nearly 300 posts in alt.binaries.pictures.erotica.unix, but when I open the newsgroup happily expecting "NAKED LINUX BOX (1/1)" and "watchmesutorootandgetnasty.mpg", instead I'm greeted with typical porn spam. Life is cruel.
Anyone who thinks this action is somehow illegal or unwarranted should read our trademark laws. The names of the groups alone violate those laws.
A quick count shows that only 4 of the 10 newsgroups listed would've had a potential trademark conflict. The other 6 were either completely generic (alt.binaries.movies), named after a standard multimedia fileformat (alt.binaries.movies.divx; I'm sure the DivX people don't object to people using their name to describe files that are, in fact, encoded in DivX), or named after a pirate group that releases movies (all the ones toward the bottom of the list; I'm basing this last one on another Slashdot post and not first-hand knowledge). Besides, I don't see anyone going after, say, the comp.os.ms-windows.* hierarchy or the comp.sys.ibm.* hierarchy. Overall, I suspect that group content played a much, much larger role than any trademarks issues with the name.
well, I will agree. That is a useful idea. But I still have trouble getting past the fact that with this plan, you wouldn't be altering your content, but the content of other sites. As a content creator, this makes me really uneasy.
You make some good points that I'm not sure how to fully address. Part of the way I see SmartTags is not as adding content, per se, but rather providing a means of accessing an external reference on data you're studying (akin to going to look something up in the encyclopedia, for example). The fundamental problems, however, is that both the data you're studying and the external reference happen to be "the web" and the access is being facilitated in such a way that makes it appear that the data being read is different from what originally written.
One partial solution might be some user interface changes to make it more obvious to the reader what's part of the real website and what's being added client-side. Also, I think education will help to a degree. In the real world, for example, a reader has no problem understanding that in a Norton annotated edition of Hamlet, the stuff on the left is what Shakespeare wrote while the stuff on the right was written by a bunch of Shakespearean scholars. A more web-oriented comparison might be how most users (I hope) have come to understand that when you click on an external link, you're actually going to another site, maintained by a different person. While some sites have external link disclaimers, most don't bother.
On the PHB issue, I think you're fairly safe in that either they'll never figure out how to turn SmartTags on in the first place or they'll see SmartTags on everything. Even if they don't put two and two together and realize that the SmartTags are being generated by their web browser, they'll think that the website you created is just conforming to standard Internet conventions or something.
In the case of commercial software, the decisions are based on committee and entirely profit driven. In the end, the objective is to ship more units.
On the other hand, the goal of Linux is "World domination. Fast.", so it's not just the profit-driven entities who want to see their product everywhere. The more people who use Linux, the more the platform benefits as a whole. Every new Linux user may turn out to be a Linux developer. Every new Linux developer may go on to write a really great application for Linux (or port an existing one over). Every great application for Linux makes it that much more appealing. That appeal is what turns around and sucks in new Linux users, creating a bit of a cycle.
After all, critics who don't contribute are so highly respected in every field.
One could argue that, since software's dynamic and more functional in nature, most criticism is more likely to be constructive. If I say "I like the painting, but the rays from the sun make it hard to see the people.", there's not much the painter can do (unless he feels like creating a companion painting showing the same scene with different lighting). If I say "Great web browser, but the navigation buttons are a little large.", there's a decent chance that some developer will agree and produce one of the following (in order of increasing developer effort): a patch to use smaller navigation images, a configuration option to let the user select smaller navigation images, or a fundamental redesign of the browser to make it skinnable.
When I've developed software personally, I've added features that I had zero personal interest in, simply because I knew they would be extremely useful to other users. Sure, it's not enough to necessarily motivate me to do an entire from-scratch project that I don't care about, but I still write some code for people other than myself.
The difference is that the end user must actively translate the document, where SmartTags will add them automatically. The real problem with SmartTags though is that MS controls where these new links on your page go.
Babelfish changes the links (both internal and external) in each translated page to point back through Babelfish. So while you must actively invoke it the first time, it continues to operate on its own. Similarly, while SmartTags default off, they stay on once you turn them on. The only major difference between Babelfish and SmartTags in this respect is whether or not the new mode is kept if you quit the web browser.
As for the whole MS/control issue, third-parties can provide SmartTags. However, the decision of which SmartTags the user does or doesn't use is ultimately up to the user. Besides, what's to keep Babelfish from translating "Google" to "Altavista" in an effort to warp reality? And, unlike with SmartTags, a user can't actively decide which word translations to accept or reject.
Mozilla isn't written for you. It's written by the authors, for the authors. After digging through the Mozilla site, I won't disagree that Mozilla's focus is very developer driven, with less focus on a final consumer browser. However, they do specificially acknowledge the encapsulation of Mozilla within a consumer browser, and they seem to be open to input by everyone (as under the FAQ entries was a link to their public wishlist newsgroup). And while there's certainly a "personal itch" factor, I'd be willing to bet some of the Mozilla people wouldn't be doing what they do if they didn't think other people would be using it. A painter may paint a painting simply because he wants to paint, but it's still nice to know that other people enjoyed it, too.
Besides, while a developer-centric focus could be used to somewhat negate complaints that Mozilla has a confusing install process, it doesn't work so well against complaints that Mozilla's buggy. Just because a painter doesn't try and sell his paintings doesn't mean I can't comment on his artistic skill (or lack thereof). Especially when he's placing his photographs up on a street corner where any people strolling along can photograph them (with permission). The only real difference is that you can no longer complain about the cost.
Slashdot Mirror
While I won't disagree that there're some people bitching over the MS issue, game console preference seem to be as equally religiously charged as text editor preference. You'll find numerous people all claiming how great the suckitude is of any current (or even unreleased) game system from the PS2 to the N64 to the Dreamcast to the GameCube to the XBOX. It's just a given.
However, it's worth pointing out that if I ignore the anti-Microsoft bias, all I'm left with is the fact that it's a game console from a company that hasn't produced a console before. That's not to say they can't pull it off -- just that there's nothing to get me downright excited until I reach out and touch one.
Damn. One of my pet peeves has always been the misspelling of Wookiee (14.6k hits) as Wookie (49k hits). I guess I'm officially in the wrong, now.
On the other hand, "Erasmus is right" has 4 hits, while "Erasmus is wrong" has only 1 hit. So I'm still right 80% of the time.
It's somewhere in-between -- I need a functional word processor that's capable of properly reading and writing Microsoft Word documents. I am required to deal with them as part of my job. Ideally, everyone would use PDF or, even better, plain text. But that isn't the world I live in.
Others have already pointed out the game issue. I think what Loki is doing is great, but for the most part, there is a significant lack of games for Linux.
Finally, despite all the effort being put into Gnome and KDE, it just doesn't have a nice, unified GUI "feel" to the desktop. Most of my Linux use is done on the text console -- X is used only for websites that require Netscape, and for the occasions when I need to fight with StarOffice.
Quality. When using StarOffice 5.2, I was running into significant bugs every 5 minutes or so. Microsoft may have their own share of bugs, but StarOffice makes 'em look downright perfect in comparison.
Well good for you. Unfortunately, those of us not living in never-never land have discovered that, contrary to popular belief, there's a lot of software out there that hasn't been ported to Linux.
Don't get me wrong -- I love Linux. I use it both at home and at work. It's been my general operating system of choice for about 6 years now. However, that doesn't change the fact that there're some computer-related activites for which Windows is the better answer, either because more commercial developers supports it or the software requires less effort to install-and-use.
So until Linux fully addresses these issues (something which, as much as I'm a fan of Linux, I doubt will every fully happen), there will be a need for some people to install and use Windows. Claiming otherwise is at best misguided advocacy and at worst trolling.
Some random (presumably trademarked) common names/terms:
Think of trademarks kind of like a (more properly run) set of domain names. You have to be actively using it to claim it. You only get an entry in the TLDs that directly relate to your product. So "wendys.food" is exclusive to the Wendy's corporation, but wouldn't have much of a claim if Wendy Johnson wanted to open "Wendy's Car Repair".
In general, trademarks seem to be (at least in my understanding) one of the better-run intellectual property things. I think they just get a bad rap due to the unfortunate interaction between trademarks and domain registries (where all commercial entities have been more-or-less lumped together in .com).
Not even Pac-man. If I'm not mistaken, it had a problem with the level counter rolling over and messing everything up after 255 boards.
This is all RMS's fault. He kicked off the trend when he decided to name his efforts at creating a Unix-like operating system "GNU is Not Unix".
But that assumes that someone picks the hardware first and then picks the operating system. Often times, this isn't the case. Most consumers, for example, wouldn't buy a computer that's three times as fast as an Intel-based machine of the same price, but that doesn't run Windows.
However, there are more uses for a scripting language than just quick one-off scripts. As mentioned in the article blurb, this language can be embedded within your own applications to provide scripting support.
If you need significant security, it's a stupid password because it's guessable. If you don't need significant security, it's a stupid password because it's (relatively) long. You might as well go with "1234". It's equally guessable, but more than twice as easy to type (if you factor in both length and the ease of typing something sequential).
Or even better: "Good morning, ma'am. We're taking a survey on what passwords people use. What's your password?"
I suspect the relevant scenario might be more like:
Buy tokens with credit card. Play with tokens. Lose. Dispute charges on credit card bill.
With the new law, the company can't turn around and fight the fact that you're disputing the charges. They can just cancel your account and blacklist your credit card. Given that some players might win, it's going to be a huge headache for the casinos.
So the casinos can either take the risk of credit cards (depending on how often people pull that stunt) or they can require tokens to be purchased with a check-equivilant (which lowers the convenience and impulse factor). Overall, the law certainly doesn't seem to prevent Australians from online, off-shore gambling (although the article implied it did), but (unless I've misinterpreted the credit card/legal interation) it does put a damper on the casinos' profitability.
Unless both machines have unroutable IP addresses. My work machine uses 10.0.0.3, but gets masqueraded through a Linux box. As such, it generally can only initiate connections, but it can't receive them. (Presumably, the push message tells such a machine to initiate a connection to the client, to get around this limitation.) Now someone else who's in the same situation wouldn't be able to connect with me, since neither of our machines could be the one to receive a new, incoming connection. As such, it'd be nice if the client (I use limewire) were smart enough to let me filter out all the non-routable IP addresses. I know it can recognize them, as it highlights them in red. But I found nothing in the options that would let me auto-ignore what're more-or-less worthless search results.
It may be the submitter's opinion, but it was a Slashdot editor who chose to place it on the front page.
No. This has nothing to do with musicians. Record companies already distribute musical recordings in a digital form. Furthermore, I can't imagine them not including "online distribution rights" within the contract that the musicians must sign when signing with a record label. Finally, no one has questioned whether or not the record companies are in control of the electronic distribution rights of songs. Instead, the two main issues have been whether private users have a valid fair use exception to allow P2P file sharing and whether the RIAA has the right to directly sue Napster for the actions of people using their system (especially since the file gets sent directly from one client to another).
In this case, it's talking about organizations that have specifically paid a freelancer in the past for the right to use that person's work. At the time, electronic distribution was not considered in the contract. It's not that the news agencies were attempting to distribute content that they weren't allowed to distribute. Rather, they were trying to distribute it via a new medium. (As opposed to the musician issue, where even before Napster, I couldn't sit on a street corner and give out illegal copies of a commercially-produced tape to every car that drove by.)
They didn't say IP numbers. They said URLs. If every image, page, frameset, and CGI script on the web can each have their own URL, providing one to each lightbulb isn't going to be that difficult:
http://bulb.erasmusdarwinsapartment.org/bulb1
http://bulb.erasmusdarwinsapartment.org/bulb2
http://bulb.erasmusdarwinsapartment.org/bulb3
Of course, on the other hand, they could also mean URLs in the form of:
http://10.0.5.1/
http://10.0.5.2/
http://10.0.5.3/
So they aren't necessarily doing things the smart way. But my guess is that they use something simpler than TCP/IP for communicating to the actual bulb sockets (unless this is one of those projects that is just mindlessly burning money). In reality, it wouldn't surprise me if the URLs turned out to actually be something closer to:
http://www.erasmusdarwinsapartment.org/cgi-bin/bul b?id=1&set=onl b?id=2&set=offl b?id=3&set=toggle
http://www.erasmusdarwinsapartment.org/cgi-bin/bu
http://www.erasmusdarwinsapartment.org/cgi-bin/bu
Overall, though, attempting to idly speculate based on the vague details from an obvious fluff piece isn't really going to get us much.
To me, it seems that Cyc is working on the second step of creating a "true" AI -- first you build it, then you teach it. While the scientists may be teaching something that's more just a smoke-and-mirrors database, that doesn't mean that Cyc wouldn't be an effective training tool to use to teach a true AI. The knowledge being placed within Cyc is a useful part of thought, even if it isn't inherently self-aware.
A quick count shows that only 4 of the 10 newsgroups listed would've had a potential trademark conflict. The other 6 were either completely generic (alt.binaries.movies), named after a standard multimedia fileformat (alt.binaries.movies.divx; I'm sure the DivX people don't object to people using their name to describe files that are, in fact, encoded in DivX), or named after a pirate group that releases movies (all the ones toward the bottom of the list; I'm basing this last one on another Slashdot post and not first-hand knowledge). Besides, I don't see anyone going after, say, the comp.os.ms-windows.* hierarchy or the comp.sys.ibm.* hierarchy. Overall, I suspect that group content played a much, much larger role than any trademarks issues with the name.
In order to make it more analagous to spam, you should call him collect and use a fake name.
You make some good points that I'm not sure how to fully address. Part of the way I see SmartTags is not as adding content, per se, but rather providing a means of accessing an external reference on data you're studying (akin to going to look something up in the encyclopedia, for example). The fundamental problems, however, is that both the data you're studying and the external reference happen to be "the web" and the access is being facilitated in such a way that makes it appear that the data being read is different from what originally written.
One partial solution might be some user interface changes to make it more obvious to the reader what's part of the real website and what's being added client-side. Also, I think education will help to a degree. In the real world, for example, a reader has no problem understanding that in a Norton annotated edition of Hamlet, the stuff on the left is what Shakespeare wrote while the stuff on the right was written by a bunch of Shakespearean scholars. A more web-oriented comparison might be how most users (I hope) have come to understand that when you click on an external link, you're actually going to another site, maintained by a different person. While some sites have external link disclaimers, most don't bother.
On the PHB issue, I think you're fairly safe in that either they'll never figure out how to turn SmartTags on in the first place or they'll see SmartTags on everything. Even if they don't put two and two together and realize that the SmartTags are being generated by their web browser, they'll think that the website you created is just conforming to standard Internet conventions or something.
On the other hand, the goal of Linux is "World domination. Fast.", so it's not just the profit-driven entities who want to see their product everywhere. The more people who use Linux, the more the platform benefits as a whole. Every new Linux user may turn out to be a Linux developer. Every new Linux developer may go on to write a really great application for Linux (or port an existing one over). Every great application for Linux makes it that much more appealing. That appeal is what turns around and sucks in new Linux users, creating a bit of a cycle.
After all, critics who don't contribute are so highly respected in every field.
One could argue that, since software's dynamic and more functional in nature, most criticism is more likely to be constructive. If I say "I like the painting, but the rays from the sun make it hard to see the people.", there's not much the painter can do (unless he feels like creating a companion painting showing the same scene with different lighting). If I say "Great web browser, but the navigation buttons are a little large.", there's a decent chance that some developer will agree and produce one of the following (in order of increasing developer effort): a patch to use smaller navigation images, a configuration option to let the user select smaller navigation images, or a fundamental redesign of the browser to make it skinnable.
When I've developed software personally, I've added features that I had zero personal interest in, simply because I knew they would be extremely useful to other users. Sure, it's not enough to necessarily motivate me to do an entire from-scratch project that I don't care about, but I still write some code for people other than myself.
Babelfish changes the links (both internal and external) in each translated page to point back through Babelfish. So while you must actively invoke it the first time, it continues to operate on its own. Similarly, while SmartTags default off, they stay on once you turn them on. The only major difference between Babelfish and SmartTags in this respect is whether or not the new mode is kept if you quit the web browser.
As for the whole MS/control issue, third-parties can provide SmartTags. However, the decision of which SmartTags the user does or doesn't use is ultimately up to the user. Besides, what's to keep Babelfish from translating "Google" to "Altavista" in an effort to warp reality? And, unlike with SmartTags, a user can't actively decide which word translations to accept or reject.
Besides, while a developer-centric focus could be used to somewhat negate complaints that Mozilla has a confusing install process, it doesn't work so well against complaints that Mozilla's buggy. Just because a painter doesn't try and sell his paintings doesn't mean I can't comment on his artistic skill (or lack thereof). Especially when he's placing his photographs up on a street corner where any people strolling along can photograph them (with permission). The only real difference is that you can no longer complain about the cost.