Long enough that patches should have been out for months now.
On a Fedora Core 3 machine I manage,/etc/localtime has a November, 2005, timestamp. It displays the correct DST settings for 2007 when parsed with zdump. So I guess somebody was paying attention when the bill was passed. That's sixteen months if I counted correctly.
I have some servers running RH 7.3 for which there are no rpm-based updates that I could find (fedoralegacy having closed down). I followed the instructions in the article to update/usr/share/zoneinfo, but that alone doesn't do the trick. The file/etc/localtime on these systems is a static binary, not a link to/usr/share/zoneinfo/America/New_York or whatever's appropriate for your timezone. The fix is simply to delete/etc/localtime and create a symlink with the same name to the correct zone data in/usr/share/zoneinfo.
You seem pretty happy having made the switch. Let me toss out some advice about some of the issues you raised.
Configuring a dual-boot system took me 4-6 hours to figure out, setting up the right partitions (making sure nothing on my windows partitions got erased) took me wayy too long (screwed it up twice).
Configuring dual-boot on a single-drive system is hard. Installing a second drive makes the task much easier. Move the Windows drive to the secondary, install the new drive as the primary, install Linux. When I did this with Fedora it detected the Windows partition on the secondary drive and set up dual-boot for me without a hitch. I didn't have to play with the partitioning of the Linux drive either; I could just accept the defaults.
Figuring out how to move from firefox 1.5 to firefox 2.0 was surprisingly difficult. I don't really understand why that particular thing isn't part of the yum update process but that's just an outsider's perspective.
Most distros take a snapshot at release time and update that software throughout the life cycle of the distro. Firefox 2.0 will be in Fedora 7, but Fedora 6 will continue to stick with 1.5.x until FC6 reaches end-of-life. That's a good decision from a support perspective, but problems can arise as you discovered when you try to update something directly from the developers rather than through the distro's package-management mechanisms. Hell, the Linux kernel in things like RedHat Enterprise 4.4 is still around 2.6.9 while the current kernel release version is 2.6.19. The RHEL kernel is heavily patched to keep up with security fixes, etc., but it's still fundamentally 2.6.9 which shipped quite a few years back now.
I haven't figured out Samba yet--this seems like it should be easy but so far it's not.
I often install webmin, a web-based administrative tool. You can install this with "yum install webmin" if you have the "extras" repository enabled. Afterwards, use a browser to open https://localhost:10000/, and log in as root, and you'll have access to a very nice and ever-growing collection of graphical Linux management tools. (You might want to change the default webmin admin account or the webmin port if you're concerned about local security.)
This is much simpler way to configure Samba if you're not used to editing config files. You do have to set up separate user accounts in Samba, but after doing that, all you need do is issue two commands at the prompt (as root):
#chkconfig smb on #service smb start
The first tells the machine to start Samba upon reboot; the second starts the server right now. Once nice feature of having separate accounts for Samba is you can use a Windows login that doesn't match your Unix login. For instance, I have the same username in both cases, but different passwords.
There are GUI tools that you can use for service management, but I prefer to use the command-line for simple tasks like this.
I'm probably going to roll out a test of the Linux Terminal Server Project for one of my healthcare clients in the next few months. We think thin desktops with no local storage and a single shared OS image is the way to go for healthcare providers that need to meet HIPAA privacy and security regulations. Having a single OS image has a lot of appeal to their support staff as well.
I'm out of date it appears. I know for certain that early 802.11b Linksys routers limited you to 192.168.1.0/24, and I thought that was true for competing products at the time as well. This was an obviously arbitrary limitation that I believe was intended to maintain a distinction between "consumer" and "professional" product lines. The latter allowed you to entire address space on the inside. That's probably half-a-dozen years ago now. I'm glad to see this isn't true any more.
In the legal case at issue the events took place on August 7, 2004. I obviously don't know the state of routers now, so I'm certainly not competent to comment on what was available then. Perhaps you know?
My experience with consumer routers like the ones by Linksys and Netgear is that they always assign internal addresses in the 192.168 space. Some Linksys routers only made the 192.168.1.0/24 space available; my Netgear permits 192.168.0.0/16. Perhaps there are some other consumer routers out there that let me assign addresses outside of the RFC1918 specs, but I haven't seen any.
All these things may be true, but it's unlikely that a jury would believe that this particular defendant went to all this trouble. After all, one line of argument in this case is that the defendant doesn't even use computers, much less play tricks with her IP address.
I read most, but not all, of the testimony. The most compelling argument he makes is the match between the IP address reported by KaZaa and the IP address assigned by Verizon. If the computer were behind a router, KaZaa would see that computer's router-assigned address in a space like 192.168.0.0/16. As an informed juror, I would tend to disregard the various "you can't really know it was this computer, can you?" lines of argument as a result.
Of course that says nothing about whether the defendant personally was involved.
I also find the line of argument which suggests that someone's DHCP-assigned address can arbitrarily change while connected implausible. It's highly unlikely that the address would change within a single online session, much less at the frequency you suggest through your line of questioning on p. 56. I understand that the fundamental purpose in all of this is raising doubt, but there are many more dubious things in his deposition than this.
I was really puzzled by all the dancing around the question of MAC addresses on page 60 and following. He seemed to be dodging a lot here for no obvious reason. I guess the evidentiary point is that the MAC address is somehow more "innate," though as you point out hardly immutable, while the IP address is more transitory. That's such an obvious technical point, though, that I didn't understand why there was so much bobbing-and-weaving going on.
My favorite was the one guy who actually talked about Linux, then said that Linux wasn't ready for the corporate desktop because he couldn't get OpenSuSE to play a DVD. Funny, I didn't know corporations required that their employees be able to play DVDs on the office workstations. Of course, he attributed the problem to Linux rather than to IP encumbrances, and then said that he didn't want to go the "wink, wink, nudge, nudge" route of obtaining non-free items of questionable legality. Fine, so go buy a commercial Linux with proprietary codecs already.
ISPs (or phone companies) have ligitimate (sic) reasons to be concerned about what you run. A misbehaving application can cause a computer to cause endless crap on a network
Funny how most of the large ISPs don't seem to give a damn about all those customers with compromised computers sending out thousands of spams and trying to break into my firewall every day. The fact is that ISPs don't seem to care what you're doing unless it might affect their ability to oversell capacity. Then they'll suddenly discover the need for usage caps, etc.
For the first half-dozen years that cable companies sold Internet service they never offered their customers even a cheapo firewall router to protect both the network from the customers and vice versa. They'd just happily connect your Windows PC directly onto the Internet without any firewalling and leave it to be hammered day and night. This negligence brought us years of malware and spam that we'll all be coping with for years to come.
You don't say where the Kubuntu box is located in your network. If you have a Linux box running as a firewall, I'd suggest looking at Dan's Guardian, a content filter than runs on top of Squid. You can configure iptables to push all outbound web traffic through Squid/DG so you can't avoid the proxy by changing the browser's settings. For more fine-grained control, you could write rules push traffic from some internal IPs through the proxy but let pass unfiltered traffic from other IPs like you own.
you strongly imply that Linux somehow takes more time to configure and administer than windows and that is simply false.
That may well be true when supporting first-time or novice users, the kinds of people who are going to be calling Dell when they can't figure out how to configure a printer.
Customer: "I can't figure out how to install a printer. Where's the Control Panel?" Support: "Are you using GNOME or KDE?" Customer: "GNOME? There's a statue of one in my mother-in-law's garden."
Before you run out and buy this, I suggest you read the EULA for Windows XP Professional:
You may install, use, access, display and run one copy of the Product on a single computer, such as a workstation, terminal or other device ("Workstation Computer"). The Product may not be used by more than two (2) processors at any one time on any single Workstation Computer. You may permit a maximum of ten (10) computers or other electronic devices (each a "Device") to connect to the Workstation Computer to utilize the services of the Product solely for File and Print services, Internet Information Services, and remote access (including connection sharing and telephony services). The ten connection maximum includes any indirect connections made through "multiplexing" or other software or hardware which pools or aggregates connections. Except as otherwise permitted by the NetMeeting, Remote Assistance, and Remote Desktop features described below, you may not use the Product to permit any Device to use, access, display or run other executable software residing on the Workstation Computer, nor may you permit any Device to use, access, display, or run the Product or Product's user interface, unless the Device has a separate license for the Product.
I'd be curious how xpunlimited.com explains that their software conforms to the Windows EULA.
While it would certainly help if American ISPs did more to control things like outbound port 25 connections, you still have the rest of the world to consider. A large fraction of the traffic that compromised machines send to my SMTP servers now originates outside North America. Bogus SMTP connections from outside the US have grown considerably faster over the past two or three years when compared to the same traffic originating on American hosts.
Though spam now makes up a very high percentage of all SMTP traffic, that traffic in toto is still not very substantial in comparison to torrents and web services that distribute images and video. Plain-text email messages, whether legitimate or spam, just aren't that large in comparison to other types of payloads. It now seems impossible to find good estimates in the public arena of the distribution of Internet traffic by type of service, so I can't point to statistics in support of my argument. (I tried a few Google searches for these figures without much luck. There are some proprietary reports that claim to include such data, but nothing in the public domain that I could find.) However consider that the transfer of one 30-minute AVI file (say about 250 MB) is about equivalent to exchanging over 30,000 email messages averaging 8K each.
Finally many mail providers like me blacklist SMTP connections from obviously suspect hosts. In my case at least a third of all the SMTP connections that arrive here never make it past the front door because they run afoul of some blocking rule. In those cases the message itself isn't transferred at all reducing even further whatever bandwidth hit might arise from spamming.
None of what I said here should be taken as an endorsement of any policies supposedly intended to improve Internet performance. I'm simply saying I don't think spam, however annoying, plays a very large role when it comes to the global demand for bandwidth.
(I don't think DDOS matters much either. Most of the time the perpetrator is flooding the target with connection attempts that use miniscule amounts of bandwidth. Automated infection or hacking attempts like attacks against ports 1025-1027 are also quite common but again use very little bandwidth.)
I think you'll be very hard pressed to find a laptop these days that doesn't include built-in wireless.
I haven't really worried much about wireless security. I have our wifi router configured to talk to only the MAC addresses of the computers we own, so outsiders are effectively locked out. I haven't tried implementing WPA or similar encryption technologies because my experience in the past was they were a real hassle. I suppose someone could have a wireless sniffer in my neighborhood gathering passwords and the like, but I think the likelihood of that is so low as to not be worth the effort.
Hell, I can see at least a half-dozen wifi access points from the confines of my home office. Anyone wanting to jump onto a unsecured wifi will just move on to someone else's after he or she is refused a connection by my AP.
Wifi was just a much easier solution to getting a connection into my daughter's upstairs bedroom in our rented home. Sure beats pulling CAT-5 wire.
The arstechnica article reports that, according to "Fight Crime: Invest in Kids," one out of every three teenagers is the target of cyberbullying. I found this figure so implausible that I had to track down the source, which is a survey conducted on behalf of that organization by Opinion Research Corporation. Here's the actual question which I suspect is the basis for this "one in three" claim:
In the past year, how many times have any mean, threatening or embarrassing things been said about you or to you through email, instant messages, websites such as MySpace, Friendster, etc., chat rooms or text messages?
All told, 178 of the 500 (weighted) respondents, or 36%, answered affirmatively to this question. Of those, 114 (or 64% of those responding affirmatively) said this happened only once or twice in the past year. Kids have been saying "mean, threatening or embarrassing things" about one another since humans walked upright. Nowhere in the survey were these kids asked about whether "mean, threatening or embarrassing things" were said to them in any other venue besides online. Perhaps one in three teens would say they had heard similar things said to or about them in person, on paper, on blackboards, above urinals, etc. Perhaps the rate would even be higher than one in three, but we won't know the answer to that from this survey.
I'm certainly opposed to bullying, but wacko statistics like this one make it harder to take these people seriously.
I'd have to say that the arstechnica editor didn't do a very good job here either. Not only is there no link to the data supporting this claim, but the report simply cites this statistic unquestioningly.
I think a more likely outcome will be player manufacturers supporting WMA and other encumbered Microsoft formats. Sure there'll be manufacturers like Cowon http://www.cowonglobal.com/ that support Ogg, but their main concern in the longer run is having their players work with the various commercial music download services. At a minimum, that includes "Plays-for-Sure" and WMA.
I'd bet there are some inside Microsoft who aren't unhappy with this outcome. If all the MP3 manufacturers are suddenly at risk for extortion by Alcatel-Lucent, interest in other formats will only increase. I see Microsoft as a bigger beneficiary of that interest than the Ogg developers.
Take a look at products by the Korean manufacturer Cowon http://www.cowonglobal.com/. My A2 audio/video player supports Ogg, and a quick check of a couple of their audio players shows the same. Now if we could only get them to support Matroska....
The A2 is a really nice device by the way. Widescreen video, audio/video recording capability, works as a USB mass-storage device with Linux, 30 GB hard drive, all for a bit over $300 from Amazon. Bought this to replace a stolen Archos Gmini 400, which we also liked, but the A2 is definitely superior, especially its support for widescreen (704x396) formats.
Now if we could only get them to support Matroska....
I was considering a Dell laptop as this one gets more tired, but I'm thinking their website might prevent this. Oh dear.
I just bought my daughter an Inspiron 640m and installed Fedora Core 6 on it without much difficulty. I made sure I ordered the Intel 3945 wireless card because I knew from past experience with another laptop the Intel makes their firmware and drivers freely available. The machine also has Intel 945GM video for which Intel also publishes a Linux driver. The video RPM was included in the distro (January's "Unity" Fedora respin), but you need five RPM's for the 3945 card. They're all at atrpms.net. While you're there, pick up the 915resolution hack that enables you to set the Intel video card to the proper widescreen resolutions.
Lately the Dell configuration site has been rather slow at times. Hang in, though. I got a pretty powerful lightweight machine for $1000.
One of my clients is a community health center. We're looking into the Linux Terminal Server Project http://www.ltsp.org/ for precisely the reason that meeting HIPAA requirements for privacy and security is nearly impossible unless we can centrally control what's running on the workstations. In the next hardware tranche we're looking to go diskless with no CD writers and no USB support for mass-storage devices.
Having only one, centrally managed, desktop image has a lot of appeal as well!
I fail to understand the point of this article. Explain to me again how the world is worse-off because IBM chose to open the code for Eclipse. Hall's argument appears to be that opening code may advantage a company like IBM by forcing smaller competitors like Borland to compete against a zero-cost product. That argument seems pretty myopic to me, to say the least. It reflects an outdated view of the software marketplace that ignores the fast-growing competitive threat of free software. Borland and other proprietary software companies are no longer competing just against one another. Their competitors now span the globe and include individual developers, noncommercial cooperatives, and yes, even some commercial entities like RedHat and IBM.
I guess I no longer care whether companies like Borland survive. Their contemporary equivalents now display their wares at SourceForge. In another ten or twenty years, many more people will look to open-source repositories, and not to Microsoft, IBM, Borland, Staples, or download.com, when they want to find some new piece of software. Of course, there's already so much free software available that all or most of the programs most ordinary people need are already included for free on a Linux distribution CD.
Sure IBM has enough resources that they can develop a product like Eclipse and give it away, but what's the harm in that? Society as whole almost certainly benefits, if only in an economic sense, whenever commercial software can replaced by an effective no-cost alternative. Many of us, myself included, think that society also benefits, and perhaps benefits more greatly, when that no-cost alternative is also open-sourced and freely redistributable.
Open source has its greatest competitive advantage when it's written to fulfill some commodity function, be it serving up web pages, displaying a graphical desktop, or providing tools to develop software. These days an IDE is a commodity and not likely to be a major profit center in the years ahead.
Soon after the invention of web, a number of companies attempted to sell proprietary web-server software. Most of those companies are gone, destroyed by a bunch of ne'er-do-wells who took free software (NCSA httpd -- paid for by the US taxpayers no less) and refined it into the dominant web server on the planet. Even Sun and Tim O'Reilly probably aren't all that sad about the failure of their efforts to make money selling web-serving software. O'Reilly has no doubt made more money selling books about Apache and related software like PHP than it ever would have selling the web server software itself. Maybe that's why O'Reilly left the software business in 2001.
If having open software means that some 13-year-old who wants to learn Java can do so more easily with Eclipse, in the long run everyone benefits. If companies can't compete effectively against open-sourced software products, then that money would be better invested in some other endeavor.
I'm not sure I entirely agree with the iTunes -> iPods direction of causality. When my nieces got iPods at Christmas, they also received a number of iTunes cards as gifts. I wouldn't be surprised if the total amount spent on the cards equalled or exceeded the amount spent on the iPods themselves.
I'm sure my nieces contributed to the problems the iTunes site was experiencing in the week after Christmas.
iPods/iTunes seems like traditional razor-blade marketing to me. Every so often people buy a new razor system, but it's those blades that keep the revenues flowing at Gillette (oops, Proctor and Gamble).
You know, the next time some adware pops up on your computers, I hope you all have the same attitude toward the purveyors of that form of "guerrilla marketing" that you demonstrate here.
There are reasons why we have things like zoning laws and restrictions on billboard advertising. These guys apparently think that these days it's okay to do anything as long as it qualifies as marketing? If you're going to run a multi-city ad campaign, then go buy some billboards. Don't stick boxes around my city.
Apparently the attitude here on Slashdot is along the lines of, "wow, those dudes were like cool to think this up?" I suppose the fact that it's for Aqua Teen makes it even cooler? Personally I'm sick of marketers, and stunts like these don't do anything to improve my opinion of them.
Slashdot Mirror
Long enough that patches should have been out for months now.
/etc/localtime has a November, 2005, timestamp. It displays the correct DST settings for 2007 when parsed with zdump. So I guess somebody was paying attention when the bill was passed. That's sixteen months if I counted correctly.
On a Fedora Core 3 machine I manage,
I have some servers running RH 7.3 for which there are no rpm-based updates that I could find (fedoralegacy having closed down). I followed the instructions in the article to update /usr/share/zoneinfo, but that alone doesn't do the trick. The file /etc/localtime on these systems is a static binary, not a link to /usr/share/zoneinfo/America/New_York or whatever's appropriate for your timezone. The fix is simply to delete /etc/localtime and create a symlink with the same name to the correct zone data in /usr/share/zoneinfo.
You seem pretty happy having made the switch. Let me toss out some advice about some of the issues you raised.
Configuring a dual-boot system took me 4-6 hours to figure out, setting up the right partitions (making sure nothing on my windows partitions got erased) took me wayy too long (screwed it up twice).
Configuring dual-boot on a single-drive system is hard. Installing a second drive makes the task much easier. Move the Windows drive to the secondary, install the new drive as the primary, install Linux. When I did this with Fedora it detected the Windows partition on the secondary drive and set up dual-boot for me without a hitch. I didn't have to play with the partitioning of the Linux drive either; I could just accept the defaults.
Figuring out how to move from firefox 1.5 to firefox 2.0 was surprisingly difficult. I don't really understand why that particular thing isn't part of the yum update process but that's just an outsider's perspective.
Most distros take a snapshot at release time and update that software throughout the life cycle of the distro. Firefox 2.0 will be in Fedora 7, but Fedora 6 will continue to stick with 1.5.x until FC6 reaches end-of-life. That's a good decision from a support perspective, but problems can arise as you discovered when you try to update something directly from the developers rather than through the distro's package-management mechanisms. Hell, the Linux kernel in things like RedHat Enterprise 4.4 is still around 2.6.9 while the current kernel release version is 2.6.19. The RHEL kernel is heavily patched to keep up with security fixes, etc., but it's still fundamentally 2.6.9 which shipped quite a few years back now.
I haven't figured out Samba yet--this seems like it should be easy but so far it's not.
I often install webmin, a web-based administrative tool. You can install this with "yum install webmin" if you have the "extras" repository enabled. Afterwards, use a browser to open https://localhost:10000/, and log in as root, and you'll have access to a very nice and ever-growing collection of graphical Linux management tools. (You might want to change the default webmin admin account or the webmin port if you're concerned about local security.)
This is much simpler way to configure Samba if you're not used to editing config files. You do have to set up separate user accounts in Samba, but after doing that, all you need do is issue two commands at the prompt (as root):
#chkconfig smb on
#service smb start
The first tells the machine to start Samba upon reboot; the second starts the server right now. Once nice feature of having separate accounts for Samba is you can use a Windows login that doesn't match your Unix login. For instance, I have the same username in both cases, but different passwords.
There are GUI tools that you can use for service management, but I prefer to use the command-line for simple tasks like this.
I'm probably going to roll out a test of the Linux Terminal Server Project for one of my healthcare clients in the next few months. We think thin desktops with no local storage and a single shared OS image is the way to go for healthcare providers that need to meet HIPAA privacy and security regulations. Having a single OS image has a lot of appeal to their support staff as well.
What distro are you using again? In Fedora, I go to the Graphics menu and select GIMP. That's it. And, I'm running KDE.
I'm out of date it appears. I know for certain that early 802.11b Linksys routers limited you to 192.168.1.0/24, and I thought that was true for competing products at the time as well. This was an obviously arbitrary limitation that I believe was intended to maintain a distinction between "consumer" and "professional" product lines. The latter allowed you to entire address space on the inside. That's probably half-a-dozen years ago now. I'm glad to see this isn't true any more.
In the legal case at issue the events took place on August 7, 2004. I obviously don't know the state of routers now, so I'm certainly not competent to comment on what was available then. Perhaps you know?
My experience with consumer routers like the ones by Linksys and Netgear is that they always assign internal addresses in the 192.168 space. Some Linksys routers only made the 192.168.1.0/24 space available; my Netgear permits 192.168.0.0/16. Perhaps there are some other consumer routers out there that let me assign addresses outside of the RFC1918 specs, but I haven't seen any.
All these things may be true, but it's unlikely that a jury would believe that this particular defendant went to all this trouble. After all, one line of argument in this case is that the defendant doesn't even use computers, much less play tricks with her IP address.
I read most, but not all, of the testimony. The most compelling argument he makes is the match between the IP address reported by KaZaa and the IP address assigned by Verizon. If the computer were behind a router, KaZaa would see that computer's router-assigned address in a space like 192.168.0.0/16. As an informed juror, I would tend to disregard the various "you can't really know it was this computer, can you?" lines of argument as a result.
Of course that says nothing about whether the defendant personally was involved.
I also find the line of argument which suggests that someone's DHCP-assigned address can arbitrarily change while connected implausible. It's highly unlikely that the address would change within a single online session, much less at the frequency you suggest through your line of questioning on p. 56. I understand that the fundamental purpose in all of this is raising doubt, but there are many more dubious things in his deposition than this.
I was really puzzled by all the dancing around the question of MAC addresses on page 60 and following. He seemed to be dodging a lot here for no obvious reason. I guess the evidentiary point is that the MAC address is somehow more "innate," though as you point out hardly immutable, while the IP address is more transitory. That's such an obvious technical point, though, that I didn't understand why there was so much bobbing-and-weaving going on.
My favorite was the one guy who actually talked about Linux, then said that Linux wasn't ready for the corporate desktop because he couldn't get OpenSuSE to play a DVD. Funny, I didn't know corporations required that their employees be able to play DVDs on the office workstations. Of course, he attributed the problem to Linux rather than to IP encumbrances, and then said that he didn't want to go the "wink, wink, nudge, nudge" route of obtaining non-free items of questionable legality. Fine, so go buy a commercial Linux with proprietary codecs already.
ISPs (or phone companies) have ligitimate (sic) reasons to be concerned about what you run. A misbehaving application can cause a computer to cause endless crap on a network
Funny how most of the large ISPs don't seem to give a damn about all those customers with compromised computers sending out thousands of spams and trying to break into my firewall every day. The fact is that ISPs don't seem to care what you're doing unless it might affect their ability to oversell capacity. Then they'll suddenly discover the need for usage caps, etc.
For the first half-dozen years that cable companies sold Internet service they never offered their customers even a cheapo firewall router to protect both the network from the customers and vice versa. They'd just happily connect your Windows PC directly onto the Internet without any firewalling and leave it to be hammered day and night. This negligence brought us years of malware and spam that we'll all be coping with for years to come.
You don't say where the Kubuntu box is located in your network. If you have a Linux box running as a firewall, I'd suggest looking at Dan's Guardian, a content filter than runs on top of Squid. You can configure iptables to push all outbound web traffic through Squid/DG so you can't avoid the proxy by changing the browser's settings. For more fine-grained control, you could write rules push traffic from some internal IPs through the proxy but let pass unfiltered traffic from other IPs like you own.
you strongly imply that Linux somehow takes more time to configure and administer than windows and that is simply false.
That may well be true when supporting first-time or novice users, the kinds of people who are going to be calling Dell when they can't figure out how to configure a printer.
Customer: "I can't figure out how to install a printer. Where's the Control Panel?"
Support: "Are you using GNOME or KDE?"
Customer: "GNOME? There's a statue of one in my mother-in-law's garden."
etc.
I'd be curious how xpunlimited.com explains that their software conforms to the Windows EULA.
While it would certainly help if American ISPs did more to control things like outbound port 25 connections, you still have the rest of the world to consider. A large fraction of the traffic that compromised machines send to my SMTP servers now originates outside North America. Bogus SMTP connections from outside the US have grown considerably faster over the past two or three years when compared to the same traffic originating on American hosts.
Though spam now makes up a very high percentage of all SMTP traffic, that traffic in toto is still not very substantial in comparison to torrents and web services that distribute images and video. Plain-text email messages, whether legitimate or spam, just aren't that large in comparison to other types of payloads. It now seems impossible to find good estimates in the public arena of the distribution of Internet traffic by type of service, so I can't point to statistics in support of my argument. (I tried a few Google searches for these figures without much luck. There are some proprietary reports that claim to include such data, but nothing in the public domain that I could find.) However consider that the transfer of one 30-minute AVI file (say about 250 MB) is about equivalent to exchanging over 30,000 email messages averaging 8K each.
Finally many mail providers like me blacklist SMTP connections from obviously suspect hosts. In my case at least a third of all the SMTP connections that arrive here never make it past the front door because they run afoul of some blocking rule. In those cases the message itself isn't transferred at all reducing even further whatever bandwidth hit might arise from spamming.
None of what I said here should be taken as an endorsement of any policies supposedly intended to improve Internet performance. I'm simply saying I don't think spam, however annoying, plays a very large role when it comes to the global demand for bandwidth.
(I don't think DDOS matters much either. Most of the time the perpetrator is flooding the target with connection attempts that use miniscule amounts of bandwidth. Automated infection or hacking attempts like attacks against ports 1025-1027 are also quite common but again use very little bandwidth.)
I can't help but wonder what would happen to an NBA or NFL player, or even a politician or corporate CEO, if he talked like Fuzzy.
You mean someone like Tim Hardaway?
I think you'll be very hard pressed to find a laptop these days that doesn't include built-in wireless.
I haven't really worried much about wireless security. I have our wifi router configured to talk to only the MAC addresses of the computers we own, so outsiders are effectively locked out. I haven't tried implementing WPA or similar encryption technologies because my experience in the past was they were a real hassle. I suppose someone could have a wireless sniffer in my neighborhood gathering passwords and the like, but I think the likelihood of that is so low as to not be worth the effort.
Hell, I can see at least a half-dozen wifi access points from the confines of my home office. Anyone wanting to jump onto a unsecured wifi will just move on to someone else's after he or she is refused a connection by my AP.
Wifi was just a much easier solution to getting a connection into my daughter's upstairs bedroom in our rented home. Sure beats pulling CAT-5 wire.
The arstechnica article reports that, according to "Fight Crime: Invest in Kids," one out of every three teenagers is the target of cyberbullying. I found this figure so implausible that I had to track down the source, which is a survey conducted on behalf of that organization by Opinion Research Corporation. Here's the actual question which I suspect is the basis for this "one in three" claim:
In the past year, how many times have any mean, threatening or embarrassing things been said about you or to you through email, instant messages, websites such as MySpace, Friendster, etc., chat rooms or text messages?
All told, 178 of the 500 (weighted) respondents, or 36%, answered affirmatively to this question. Of those, 114 (or 64% of those responding affirmatively) said this happened only once or twice in the past year. Kids have been saying "mean, threatening or embarrassing things" about one another since humans walked upright. Nowhere in the survey were these kids asked about whether "mean, threatening or embarrassing things" were said to them in any other venue besides online. Perhaps one in three teens would say they had heard similar things said to or about them in person, on paper, on blackboards, above urinals, etc. Perhaps the rate would even be higher than one in three, but we won't know the answer to that from this survey.
I'm certainly opposed to bullying, but wacko statistics like this one make it harder to take these people seriously.
I'd have to say that the arstechnica editor didn't do a very good job here either. Not only is there no link to the data supporting this claim, but the report simply cites this statistic unquestioningly.
I think a more likely outcome will be player manufacturers supporting WMA and other encumbered Microsoft formats. Sure there'll be manufacturers like Cowon http://www.cowonglobal.com/ that support Ogg, but their main concern in the longer run is having their players work with the various commercial music download services. At a minimum, that includes "Plays-for-Sure" and WMA.
I'd bet there are some inside Microsoft who aren't unhappy with this outcome. If all the MP3 manufacturers are suddenly at risk for extortion by Alcatel-Lucent, interest in other formats will only increase. I see Microsoft as a bigger beneficiary of that interest than the Ogg developers.
Take a look at products by the Korean manufacturer Cowon http://www.cowonglobal.com/. My A2 audio/video player supports Ogg, and a quick check of a couple of their audio players shows the same. Now if we could only get them to support Matroska....
The A2 is a really nice device by the way. Widescreen video, audio/video recording capability, works as a USB mass-storage device with Linux, 30 GB hard drive, all for a bit over $300 from Amazon. Bought this to replace a stolen Archos Gmini 400, which we also liked, but the A2 is definitely superior, especially its support for widescreen (704x396) formats.
Now if we could only get them to support Matroska....
I was considering a Dell laptop as this one gets more tired, but I'm thinking their website might prevent this. Oh dear.
I just bought my daughter an Inspiron 640m and installed Fedora Core 6 on it without much difficulty. I made sure I ordered the Intel 3945 wireless card because I knew from past experience with another laptop the Intel makes their firmware and drivers freely available. The machine also has Intel 945GM video for which Intel also publishes a Linux driver. The video RPM was included in the distro (January's "Unity" Fedora respin), but you need five RPM's for the 3945 card. They're all at atrpms.net. While you're there, pick up the 915resolution hack that enables you to set the Intel video card to the proper widescreen resolutions.
Lately the Dell configuration site has been rather slow at times. Hang in, though. I got a pretty powerful lightweight machine for $1000.
One of my clients is a community health center. We're looking into the Linux Terminal Server Project http://www.ltsp.org/ for precisely the reason that meeting HIPAA requirements for privacy and security is nearly impossible unless we can centrally control what's running on the workstations. In the next hardware tranche we're looking to go diskless with no CD writers and no USB support for mass-storage devices.
Having only one, centrally managed, desktop image has a lot of appeal as well!
I fail to understand the point of this article. Explain to me again how the world is worse-off because IBM chose to open the code for Eclipse. Hall's argument appears to be that opening code may advantage a company like IBM by forcing smaller competitors like Borland to compete against a zero-cost product. That argument seems pretty myopic to me, to say the least. It reflects an outdated view of the software marketplace that ignores the fast-growing competitive threat of free software. Borland and other proprietary software companies are no longer competing just against one another. Their competitors now span the globe and include individual developers, noncommercial cooperatives, and yes, even some commercial entities like RedHat and IBM.
I guess I no longer care whether companies like Borland survive. Their contemporary equivalents now display their wares at SourceForge. In another ten or twenty years, many more people will look to open-source repositories, and not to Microsoft, IBM, Borland, Staples, or download.com, when they want to find some new piece of software. Of course, there's already so much free software available that all or most of the programs most ordinary people need are already included for free on a Linux distribution CD.
Sure IBM has enough resources that they can develop a product like Eclipse and give it away, but what's the harm in that? Society as whole almost certainly benefits, if only in an economic sense, whenever commercial software can replaced by an effective no-cost alternative. Many of us, myself included, think that society also benefits, and perhaps benefits more greatly, when that no-cost alternative is also open-sourced and freely redistributable.
Open source has its greatest competitive advantage when it's written to fulfill some commodity function, be it serving up web pages, displaying a graphical desktop, or providing tools to develop software. These days an IDE is a commodity and not likely to be a major profit center in the years ahead.
Soon after the invention of web, a number of companies attempted to sell proprietary web-server software. Most of those companies are gone, destroyed by a bunch of ne'er-do-wells who took free software (NCSA httpd -- paid for by the US taxpayers no less) and refined it into the dominant web server on the planet. Even Sun and Tim O'Reilly probably aren't all that sad about the failure of their efforts to make money selling web-serving software. O'Reilly has no doubt made more money selling books about Apache and related software like PHP than it ever would have selling the web server software itself. Maybe that's why O'Reilly left the software business in 2001.
If having open software means that some 13-year-old who wants to learn Java can do so more easily with Eclipse, in the long run everyone benefits. If companies can't compete effectively against open-sourced software products, then that money would be better invested in some other endeavor.
I'm not sure I entirely agree with the iTunes -> iPods direction of causality. When my nieces got iPods at Christmas, they also received a number of iTunes cards as gifts. I wouldn't be surprised if the total amount spent on the cards equalled or exceeded the amount spent on the iPods themselves.
I'm sure my nieces contributed to the problems the iTunes site was experiencing in the week after Christmas.
iPods/iTunes seems like traditional razor-blade marketing to me. Every so often people buy a new razor system, but it's those blades that keep the revenues flowing at Gillette (oops, Proctor and Gamble).
You know, the next time some adware pops up on your computers, I hope you all have the same attitude toward the purveyors of that form of "guerrilla marketing" that you demonstrate here.
There are reasons why we have things like zoning laws and restrictions on billboard advertising. These guys apparently think that these days it's okay to do anything as long as it qualifies as marketing? If you're going to run a multi-city ad campaign, then go buy some billboards. Don't stick boxes around my city.
Apparently the attitude here on Slashdot is along the lines of, "wow, those dudes were like cool to think this up?" I suppose the fact that it's for Aqua Teen makes it even cooler? Personally I'm sick of marketers, and stunts like these don't do anything to improve my opinion of them.