If I weren't so lazy, I'd setup mozilla's e-mail proggie with a bayesian filter or something else. There ARE ways to conquer advertisers, and the people already have weapons like the ones I mentioned to combat it that are far more powerful than the advertisers can think up.
Sure, on the individual level, I can set up bayesian filtering to throw away junk that I don't want to read, or opt to just deal with it and honestly only waste 5 or 10 minutes a day. But that's not where the true "waste" exists. The real waste is when the email server I admin jumps from an average of 30K emails a day to 130K over a two month period. The liscense we have with the vendor who provides us our MTA software charges us not per user, but per email delivered. You can see my problem here. We're literally paying for crap. This is the part of the equation where a good Linux user should shout "why not use open source MTA's?" (which is certainly not a new idea around here). We face other problems with that like: support contracts, admin re-training, hardware shifts, etc., all of which involve $$$ which makes open source not quite as free as the penny-pinchers would like. Moreover, I'm having to spend a huge amount of time and money installing and maintaining bayesian filtering on the email server itself in a futile attempt to battle the spam.
It may only cost an end user a few minutes, but it can easily cost an organization tens of thousands of dollars. That's my beef with the "is it really that big of a problem?" argument.
Your point is well taken, but I'd offer a bit of a follow up. Here we are, a relatively young civilization when it comes to broadcasting signals out into the universe. However, we're already looking (via Seti and other programs) for someone who's trying to communicate with us. I'd argue that as technology advances and costs associated with signal monitoring generation drop, we'll hit a point where it will be so cost effective to look for neighbors, that we'll go ahead and do it, regardless of any questions about the very existence of said neighbors. Right now, it's a tough sell because it takes a substantial amount of funding to keep programs like Seti running. That money could arguably be spent on "better" things. But at some point, as costs fall, money will be taken out of the equation. Then we'll not only look for someone who is broadcasting, but I'd imagine we'll actually attempt to broadcast back at candidates, and randomly "out there." In such a case, I'd propose that we'd generate a carrier signal with basic variation (primes or some equally fundamental principle). I think that's what Seti is banking on: that someone out there is trying to say something, and knows that they need to use techniques which catch the eye of any observers. RF seems to make sense, in this case.
"I'm not sure that a device that would allow streaming but wouldn't allow you to copy would be very popular," he said.
I think you're right. He missed the point entirely. The problem is that he can't even think about the possibillity of radio being improved, enhanced, or otherwise changed. The RIAA and all of it's members are stuck with the mindset that the world will continue to accept their outdated model for music distribution. Their playing catchup (to be honest their not even playing catchup... they're just not playing) to new music distribution methods like epitonic.com, napster.com, and others, and their doing everything in their power to stop or otherwise hinder more revolutionary distribution methods. It all comes down to the simple fact that the RIAA was created as a method to protect the artist's rights and distribute music. Right now, it's not doing either particularly well. Out with the old, in with the new. And I'm not just speaking of an easily breakable DRM scheme... a totally new concept of how to profit from music is needed. Free mp3's and expensive concerts? Free mp3's and bonus material for cd buyers? Whatever method the future holds, everyone already knows that the RIAA won't be in the picture.
I believe the accusation that "Apple Lied" stems from the statements in their TV ads that the new G5's are "the world's fastest personal computer" and the statements on their website claiming "the world's first 64 bit personal computer." We all know that we can debate the speed issue to no end, but it's aparent that the G5, sexy as it may be, isn't the first 64-bit desktop.
I'd be interested in seeing what the thoughts are of the IT staffers. Where I work, keeping legacy protocols like AppleTalk up (letting people easily abuse printing rights) and dealing with the fact that Rendezvos sets up a Class B on our Class C partitioned network (and then decides to throw multicasts and broadcasts out at strange times) is a pain in my ass. As handy as I find my PowerBook, I'd be a happier person if I didn't have to deal with the "ease of use" that Apple puts into it's products.
Here at Denison University, we were lucky enough to catch wind of this perl script, written by Josh Richard of the University of Minnesota-Duluth and enhanced by Mike Lang of the University of Connecticut enhanced it. We modified our standard registration web page (unknown mac-addresses are handed a dummy ip and all traffic redirects to a registration page. Once they register, DHCP hands them a "real" ip) to scan for the DCOM vulnerability using the UCONN script. Users that fail the test are redirected to a page offering links to the patches. Users that pass are directed to the standard registration page, including virus scanning downloads. UConn also includes handy suggestions for using TCP dump to listen on port 135 and for ICMP, note it in a log, giving you a great list of IPs that need to be cleaned. Read UConn's entire summary page here. It saved us.
Check out LaunchCast They've been doing this for several years now (create a personal radio station). They get by the rules that which "explicitly prevent radio stations from doing things like allowing listeners to democratically select which songs to play" by letting listeners rate music, which performs two tasks: 1) a rather TiVo like function, using your ratings to find new music you might also like and 2) to help decide what songs you get to listen to. Note that listeners aren't saying "I'd like to hear song X next." Instead, listeners are simply showing preference for a song, artist, album, genre, or other member's preferences. The best feature is the "Red X" option, to ban a song, artist, or album from your station. It's quite swank. Best of all, it's free for basic service, and an actually reasonable subscription for enhanced features.
The fact that cisco has a webpage dedicated to their aquisitions really demonstrates their corporate mentality, for better or worse. It's a slippery slope for me. On the one hand, I am a huge fan of the products that Cisco has aquired and (in my opinion) improved on... take the VPN 3000 Series Concentrator line, or the AP 1200 series wireless access point. Both technologies built from aquired technology, and both have been improved in ways that only Cisco could improve them (using both corporate clout and Cisco owned technologies). I find it really comforting to know that when I have a new project at work, I only have to call my Cisco VAR and she and my regional Cisco rep show up with great ideas and help. But I find it unsettling somehow. I'm torn between wanting them to innovate, but at the same time, who really competes with Cisco. Juniper? Foundary? Certanly not 3Com anymore... The only way you can get your foot in the door against Cisco is to "pull a Juniper" and get your routers used instead of theirs on research networks like I2. Oh! The moral dilema...
A friend of mine wrote pop2blog, which is essentially like the Hiplog service for hiptop owners, only it posts to your MovableType blog via the Blogger API. The script is out under the GPL at http://bixworld.com/pop2blog/
I work for a small higher ed institution, facing this exact problem. Being higher ed, the term "ban" doesn't go over well with anyone (particularly myself), but usabillity dictates you do something. My solution was to install a PacketShaper 4500 behind our gateway router, which in the first fifteen minuites, recovered it's value twice over in terms of recovered bandwidth. Users performing "academic" related persuits get priority, as dictated by university policy, but the students still get unlimited access to do whatever they want when the "academic" applications aren't hogging the line. This box is disturbingly smart, paying attention to the application type, not the port number, meaning napster on port 80 is still treated like napster, not http. Very slick.
I guess this is why I don't use outlook to check my email: I just got an email from "beta@winzip.com" with the nimda readme.exe file attached. Joy. And I was just starting to forget why I should have a personal email account and a spam account.
That's just crummy admin on your Admin's part. Segment off your users, rather than having every user under the same branch of the tree. This will both boost performance (by cutting the brodcast traffic on each subnet) and allow for easier admin. Just my thougths.
Sure, LDAP and Active directory are great things. But let's take a quick look at what they are. Directory based management of all nodes/users/devices/whatever jacked into a network. They meet the same need that NDS met when it came out around 10 years ago. Now, think back to the days when NDS first came out on the scene. Netware 3. Did anyone really enjoy that product? No. Novell themselves openly admit that it was the first attempt at a rather complicated project. It wasn't until Netware 4 that NDS really became a powerful and useful tool. So, NDS has had 10 years lead (give or take) over AD and LDAP. Am I certain that these will suffer similar problems? Of course now, but NDS is tried and true. Netware has a userbase tens of millions (this isn't exactly a Banyan here) And NDS can manage both LDAP, thereby making it a more useful tool than either of the others. Expect to see Novell begin to sidestage their other products, like Netware and Groupwise, and focus on the one key product which will allow them to move forward. Expect to see an even greater cross platform push for NDS, allowing them to be the central authentication and management point for all nodes.
Slashdot Mirror
Sure, on the individual level, I can set up bayesian filtering to throw away junk that I don't want to read, or opt to just deal with it and honestly only waste 5 or 10 minutes a day. But that's not where the true "waste" exists. The real waste is when the email server I admin jumps from an average of 30K emails a day to 130K over a two month period. The liscense we have with the vendor who provides us our MTA software charges us not per user, but per email delivered. You can see my problem here. We're literally paying for crap. This is the part of the equation where a good Linux user should shout "why not use open source MTA's?" (which is certainly not a new idea around here). We face other problems with that like: support contracts, admin re-training, hardware shifts, etc., all of which involve $$$ which makes open source not quite as free as the penny-pinchers would like. Moreover, I'm having to spend a huge amount of time and money installing and maintaining bayesian filtering on the email server itself in a futile attempt to battle the spam.
It may only cost an end user a few minutes, but it can easily cost an organization tens of thousands of dollars. That's my beef with the "is it really that big of a problem?" argument.
Your point is well taken, but I'd offer a bit of a follow up. Here we are, a relatively young civilization when it comes to broadcasting signals out into the universe. However, we're already looking (via Seti and other programs) for someone who's trying to communicate with us. I'd argue that as technology advances and costs associated with signal monitoring generation drop, we'll hit a point where it will be so cost effective to look for neighbors, that we'll go ahead and do it, regardless of any questions about the very existence of said neighbors. Right now, it's a tough sell because it takes a substantial amount of funding to keep programs like Seti running. That money could arguably be spent on "better" things. But at some point, as costs fall, money will be taken out of the equation. Then we'll not only look for someone who is broadcasting, but I'd imagine we'll actually attempt to broadcast back at candidates, and randomly "out there." In such a case, I'd propose that we'd generate a carrier signal with basic variation (primes or some equally fundamental principle). I think that's what Seti is banking on: that someone out there is trying to say something, and knows that they need to use techniques which catch the eye of any observers. RF seems to make sense, in this case.
"I'm not sure that a device that would allow streaming but wouldn't allow you to copy would be very popular," he said.
I think you're right. He missed the point entirely. The problem is that he can't even think about the possibillity of radio being improved, enhanced, or otherwise changed. The RIAA and all of it's members are stuck with the mindset that the world will continue to accept their outdated model for music distribution. Their playing catchup (to be honest their not even playing catchup... they're just not playing) to new music distribution methods like epitonic.com, napster.com, and others, and their doing everything in their power to stop or otherwise hinder more revolutionary distribution methods. It all comes down to the simple fact that the RIAA was created as a method to protect the artist's rights and distribute music. Right now, it's not doing either particularly well. Out with the old, in with the new. And I'm not just speaking of an easily breakable DRM scheme... a totally new concept of how to profit from music is needed. Free mp3's and expensive concerts? Free mp3's and bonus material for cd buyers? Whatever method the future holds, everyone already knows that the RIAA won't be in the picture.
I believe the accusation that "Apple Lied" stems from the statements in their TV ads that the new G5's are "the world's fastest personal computer" and the statements on their website claiming "the world's first 64 bit personal computer." We all know that we can debate the speed issue to no end, but it's aparent that the G5, sexy as it may be, isn't the first 64-bit desktop.
my 2 cents...
-p
I'd be interested in seeing what the thoughts are of the IT staffers. Where I work, keeping legacy protocols like AppleTalk up (letting people easily abuse printing rights) and dealing with the fact that Rendezvos sets up a Class B on our Class C partitioned network (and then decides to throw multicasts and broadcasts out at strange times) is a pain in my ass. As handy as I find my PowerBook, I'd be a happier person if I didn't have to deal with the "ease of use" that Apple puts into it's products.
Here at Denison University, we were lucky enough to catch wind of this perl script, written by Josh Richard of the University of Minnesota-Duluth and enhanced by Mike Lang of the University of Connecticut enhanced it. We modified our standard registration web page (unknown mac-addresses are handed a dummy ip and all traffic redirects to a registration page. Once they register, DHCP hands them a "real" ip) to scan for the DCOM vulnerability using the UCONN script. Users that fail the test are redirected to a page offering links to the patches. Users that pass are directed to the standard registration page, including virus scanning downloads. UConn also includes handy suggestions for using TCP dump to listen on port 135 and for ICMP, note it in a log, giving you a great list of IPs that need to be cleaned. Read UConn's entire summary page here. It saved us.
Check out LaunchCast
They've been doing this for several years now (create a personal radio station). They get by the rules that which "explicitly prevent radio stations from doing things like allowing listeners to democratically select which songs to play" by letting listeners rate music, which performs two tasks: 1) a rather TiVo like function, using your ratings to find new music you might also like and 2) to help decide what songs you get to listen to. Note that listeners aren't saying "I'd like to hear song X next." Instead, listeners are simply showing preference for a song, artist, album, genre, or other member's preferences. The best feature is the "Red X" option, to ban a song, artist, or album from your station. It's quite swank. Best of all, it's free for basic service, and an actually reasonable subscription for enhanced features.
The fact that cisco has a
webpage dedicated to their aquisitions really demonstrates their corporate mentality, for better or worse. It's a slippery slope for me. On the one hand, I am a huge fan of the products that Cisco has aquired and (in my opinion) improved on... take the VPN 3000 Series Concentrator line, or the AP 1200 series wireless access point. Both technologies built from aquired technology, and both have been improved in ways that only Cisco could improve them (using both corporate clout and Cisco owned technologies). I find it really comforting to know that when I have a new project at work, I only have to call my Cisco VAR and she and my regional Cisco rep show up with great ideas and help. But I find it unsettling somehow. I'm torn between wanting them to innovate, but at the same time, who really competes with Cisco. Juniper? Foundary? Certanly not 3Com anymore... The only way you can get your foot in the door against Cisco is to "pull a Juniper" and get your routers used instead of theirs on research networks like I2. Oh! The moral dilema...
A friend of mine wrote pop2blog, which is essentially like the Hiplog service for hiptop owners, only it posts to your MovableType blog via the Blogger API. The script is out under the GPL at http://bixworld.com/pop2blog/
A few hiptop enabled sites for those picking up their Sidekicks today:
m
http://hiptop.bixworld.com
http://hiptop411.co
http://www.dangerinfo.com/sidekick
I work for a small higher ed institution, facing this exact problem. Being higher ed, the term "ban" doesn't go over well with anyone (particularly myself), but usabillity dictates you do something. My solution was to install a PacketShaper 4500 behind our gateway router, which in the first fifteen minuites, recovered it's value twice over in terms of recovered bandwidth. Users performing "academic" related persuits get priority, as dictated by university policy, but the students still get unlimited access to do whatever they want when the "academic" applications aren't hogging the line. This box is disturbingly smart, paying attention to the application type, not the port number, meaning napster on port 80 is still treated like napster, not http. Very slick.
I guess this is why I don't use outlook to check my email: I just got an email from "beta@winzip.com" with the nimda readme.exe file attached. Joy. And I was just starting to forget why I should have a personal email account and a spam account.
That's just crummy admin on your Admin's part. Segment off your users, rather than having every user under the same branch of the tree. This will both boost performance (by cutting the brodcast traffic on each subnet) and allow for easier admin. Just my thougths.
Sure, LDAP and Active directory are great things. But let's take a quick look at what they are. Directory based management of all nodes/users/devices/whatever jacked into a network. They meet the same need that NDS met when it came out around 10 years ago. Now, think back to the days when NDS first came out on the scene. Netware 3. Did anyone really enjoy that product? No. Novell themselves openly admit that it was the first attempt at a rather complicated project. It wasn't until Netware 4 that NDS really became a powerful and useful tool. So, NDS has had 10 years lead (give or take) over AD and LDAP. Am I certain that these will suffer similar problems? Of course now, but NDS is tried and true. Netware has a userbase tens of millions (this isn't exactly a Banyan here) And NDS can manage both LDAP, thereby making it a more useful tool than either of the others. Expect to see Novell begin to sidestage their other products, like Netware and Groupwise, and focus on the one key product which will allow them to move forward. Expect to see an even greater cross platform push for NDS, allowing them to be the central authentication and management point for all nodes.