The reason that installing things is so hard on Fedora is that there is a dissapointing amount of packages available for it.
Sigh. Yet another flame war over package management.
Other than packages which cannot be legally distributed with Fedora, like mp3 or dvd players, what other critical packages are missing from Fedora? I mean ones that a large number of ordinary Linux users might want to use.
I've used every Fedora version since FC1, and though I have issues with it, not being able to find the packages I need isn't typically one of them.
I've used cn-kr.blackholes.us with SpamAssassin to give extra points to messages originating in China and South Korea. Obviously it would make sense to do the same for Taiwan. However a query for tw.blackholes.us returns NXDOMAIN. I've also looked at countries.nerd.dk for a solution to this, but didn't see an obvious answer. Is anyone out there using a DNSBL to identify IPs in Taiwan? If so, what are you using?
Does it seem strange to anyone else that so many computers containing sensitive information are suddenly being reported stolen? Is it just an accident that this particular computer, containing this particular set of incredibly extensive personal information, just happened to be stolen from this person's home in Virginia?
First, how would someone know that this computer contained all this information? Perhaps this is a job for spyware. It's easy to imagine a piece of malware that looked for large personal databases and phoned home when they are found. Or, perhaps, people whose jobs gives them access to personal information are being trailed and their computers then stolen?
Yes, I know that laptop theft is pretty endemic, but the number of high-profile thefts like this one, the one's involving the auditors Ernst & Young, etc., makes one wonder if there isn't some type of sophisticated targeting going on. I realize that the pressure to disclose such thefts has risen greatly in recent months, in large part due to laws like California's that require notification. (Laws which, by the way, the Republican Congress is seeking to preempt through federal legislation.) So this could just be a result of increased reporting, but the targets involved seem to have particularly juicy caches of data.
Every time RedHat is discussed here, some bozo makes a comment like the one you just refuted. I guess the rule here is that if anyone makes any money from open-source software, they become, by definition, a "parasite" on the community. I'm sorry, but I don't object to the fact that Red Hat has managed to create a business from Linux and other OSS products; in fact, I've encouraged people to own their stock. Good for them. I've used every RedHat version from 4.0-9.0 and Fedora Core 1-5 at some time or other (and a couple of WhiteBox and CentOS respins as well). Sure there are things about these distributions that bother me (the over-emphasis of Gnome, for instance), but not the fact that Red Hat has succeeded as a business.
Usually these comments sound like sour grapes to me.
It seems hard to believe that only two out of thousands were serious enough to prosecute.
And in one of those cases, the crime involved selling an FBI agent's medical records. Wonder why the Justice Department (in which the FBI is housed) chose to prosecute that case?
I was rather puzzled by this comment, given that in the mid 1990's, Gates focused on acquiring the digital rights to lots of content around the globe. Encarta, Corbis, the acquisition of the Bettman Archive, were all parts of this strategy. Here's a good take on this I found from The Chronicle of Higher Education in 1998: http://chronicle.com/data/articles.dir/art-44.dir/ issue-33.dir/33a03301.htm.
Perhaps they just weren't capable of monopolizing sufficient content to make it a viable business strategy.
Back when I worked as a television researcher in the 80's it was becoming clear that content holders had more and more leverage over distribution channels as the networks' oligopoly was challenged. At the time I thought that content would be king. Now, I'm not so sure. Perhaps the real power these days are services like eBay or Google that repackage content in useful ways. Of course, in 1981 we couldn't see the Internet coming, so information flow was still pretty much unidirectional. The Internet enables providers like eBay to repackage content provided by its customers; essentially the audience has become the advertisers. To me, the fact that the Internet has destroyed this type of top-down information flow is one of its most revolutionary characteristics.
Of course, that's not to say that those wedded to the top-down model won't do everything they can to stop these trends, be they the MPAA, the New York Times, or the government. In the long run, though, I don't think they'll succeed.
I really didn't know; that's why I asked. I don't use Windows much anymore so I don't always keep up-to-date with every decision that comes from Redmond. Thanks for giving me the answer.
Limiting signing to 64-bit Windows will help reduce the collateral damage involving older 32-bit hardware. As for my concerns about foreclosing smaller hardware manufacturers, the article reports that Microsoft has radically lowered the barrier to entry for hardware manufacturers. MS will issue a cert for free as long as the driver writers pay an annual fee around $500 to Verisign for a "Class 3 Commercial Software Publisher Certificate." (And, might I add, what a nice little windfall for Verisign.) On the other hand, the fact that Microsoft signs a driver will no longer indicate that the driver is being certified as reliable or secure, just that it has come from someone who holds one of these free, Microsoft-issued certificates and was able to spend $500 at Verisign. This seems rather limited protection against the threat of rootkit drivers which is the claimed rationale for the requirement that drivers be signed.
The bigger issue with Vista is that the beta versions I've tried [don't] allow the installation of unsigned drivers.
Really? Is this simply an aspect of the product being in beta, or is this going to be the way Vista works after it's released? If the latter, I can see a bunch of unhappy users down the road. Lots of device manufacturers have decided it wasn't worth the tens of thousands of dollars to have Microsoft sign their drivers (both Linksys and Netgear come to mind here). Does that mean my WiFi card that didn't have signed XP drivers won't work in Vista, even if the XP drivers are compatible?
Maybe this is just another surveillance method. You know, design the computer to fail periodically, so when you send it back for repairs, they can remove the secret chip with all your keystrokes since the last repair and replace it with a fresh one!
Is it just me, or does the first page of Mr. Klein's statement sound like it was written by someone else? The first page is full of rhetoric about "this dangerous Orwellian project," while the rest is a lot more technical. It sounds to me like Mr. Klein stepped forward and a statement was crafted between him and his attorneys.
Don't read this comment as any sort of defense of the government's actions here. It just that I found the rhetorical parts of his statement made the more technical parts seem suspect, and I fear a jury might read it the same way.
All the comments in this thread so far have ignored the issue of software redistribution. The audience for the $100 laptop needs to be able to obtain and redistribute all its software freely. For instance, having a place (in a school, a Internet cafe, etc.) where you can connect this laptop and install anything it needs. That won't work unless all the software permits unlimited redistribution with no strings attached.
I just installed Fedora Core 4 and 5 (see below) on my daughter's Winbook W235 http://www.winbook.com/notebooks/w/w_overview.html. This is a very nice machine for $900 with a wide screen, DVD writer, 512M, and an 80G drive. It uses an Intel motherboard with the 855GM graphics adapter and 2200 wifi. As you note you'll need to install the Intel firmware, either from Livna http://rpm.livna.org/ or directly from Intel's site on Sourceforge http://ipw2200.sourceforge.net/firmware.php. Once I had the firmware installed, I could configure the wireless card with the included utilities in GNOME or KDE, and I didn't have to play with ndiswrapper or any other kludge.
The screen is more problematic. The 855 (and later 900+) adapters support a number of resolutions, but they weren't recognized by the normal drivers. The screen on this machine is 1280x768, but an out-of-the-box install forced it to 1024x768 which resulted in very ugly fonts. Luckily there's a little utility out there called 855resolution http://perso.wanadoo.fr/apoirier/ that you can load in rc.local and kick the adapter into other modes. Once I installed that, the KDE "Display" control saw the new resolution, and the Fedora desktop looked as nice as the Windows one.
We don't play games on this machine (we're both console types), so I can't speak for its graphics performance. It works great with Xine, though; movies and anime in the widescreen aspect ratio look terrific. Since I run KDE, I use the arts drivers to handle sound; they work fine.
OK, now for the big problem. FC4 installed just fine off the DVD, but FC5 would not install at all. (I've filed a bug report with the Fedora folks.) I finally installed FC5 by installing FC4 first, then running a system upgrade from the FC5 repositories http://www.fedoraproject.org/wiki/YumUpgradeFaq. This worked fine; it just added a few hours to the installation since yum had to update a couple thousand RPMs.
Finally, here's a hint if you buy a laptop like this one with Windows pre-installed. First, download and boot a copy of the Knoppix live CD or DVD http://www.knopper.net/. Once it's up and running, run "qtparted" from the command prompt. This is a nice graphical frontend to parted that will allow you to resize your partitions without having to buy something like Partition Magic. If you've used the Windows partition at all, I'd recommend running its disk optimizer to push all the Windows files to the front of the partition. Then you can lop off a chunk at the back for Linux.
I usually don't comment about statements like this, but this one is so clearly erroneous that I have to speak up. According to the Census Bureau http://www.census.gov/prod/2002pubs/censr-4.pdf, 80% of Americans lived in a "metropolitan" area in 2000. Fifty percent live in suburbs and the remaining 30% live in central cities. Even accepting the fact that some of the suburbs have rural portions, it's simply not the case that "most" Americans live in rural areas.
Over the past few months, I've seen half a dozen stories about stolen laptops. In most of these cases, the laptops had thousands of individuals' data records on them, usually including SSNs. I can't see any possible rationale for someone to be walking around with that much confidential information on a laptop. In general, these breaches arise when senior management is unwilling to put customers' privacy ahead of their employees' convenience. How happy would you be to know your insurer, accountant, etc. is out there walking around with your personal health data or tax returns on a laptop hard drive with no encryption?
especially if they are in an area that doesn't have internet / communications availability
Um, I doubt there are many places that don't have a telephone. People can use modems to connect to the main office, you know.
Wow, how non-obvious is that. Of course I would expect a *graphical* file manager to open an address box when you type "/". What's so bad about having an address box at the top of the screen? Even Windows Explorer lets me type in a filepath or UNC name. Konqueror works fine for me, thank you.
The Mozilla problem has been sitting in the bugzilla queue for quite some time now, too https://bugzilla.mozilla.org/show_bug.cgi?id=30881 5. I don't understand why everyone suddenly wants to make the keyboard obsolete.
I wished that mattered. Unfortunately there's a long line of Supreme Court decisions and the like which say that if Congress votes the money for a foreign conflict, it's tantamount to declaring war. Having lived through five undeclared wars (Korea, Vietnam, Iraq/Kuwait, Afghanistan, and Iraq again) and countless "police actions" (Nicaragua, Panama, etc.), I don't expect we'll ever see a "declared" war again.
The Administration's rationale for avoiding FISA has nothing to do with the Iraq war. From TFA,
Elsewhere in the document, however, the government said President Bush had explained that after the attacks of Sept. 11, 2001, he authorized the security agency to intercept communications into and out of the United States by people linked to Al Qaeda and related organizations. The agency is ordinarily prohibited from intercepting the telephone and digital communications of American citizens without a warrant from a special intelligence court.
This was clearly a "they blew up the trade center, let's bug their phones" type of reaction.
First, I didn't have anything to do with their choice of providers. I wasn't consulted on this decision; I would have suggested a different provider. But they don't have a lot of money, and Verizon offered them a pretty good deal. As I said, they send no mail from this machine; it all comes through my servers. They just use Verizon for outbound connectivity.
As a matter of fact, I have no sympathy for Verizon. As I posted elsewhere in this thread I've been blocked repeatedly by their smtp servers for no good reason. But I also don't expect most organizations, and particularly small businesses and nonprofits, to have a clue about whether a particular provider is a "good" or "bad" Internet citizen.
If you're suggesting that I don't accept mail from the wide array of addresses you proposed, I'd be out of business. I don't accept mail from all Verizon hosts, just like I don't accept mail from all Comcast hosts, terra.com.br hosts, or countless other hosts from around the world. But, in general, I don't ban entire blocks of the IP space. I use an SMTP daemon that reverse-resolves each sending host and examines its ruleset to decide what to do with that message. So I don't accept messages sent from some-random-host.verizon.net that sends a 'MAIL FROM: xyzzy@aol.com' during the SMTP transaction. But I also don't block 70.22/16 either.
For a mail provider it's much more important to insure that legitimate mail is delivered to their paying customers than it is to "send a message" to a company like Verizon. What's more likely to hurt Verizon is its own customers suddenly being cut off from e-mail correspondents that they hitherto had no problem reaching. The more Verizon engages in reckless anti-spam policies, the more they'll annoy, or even enrage, their own subscribers.
Well, you'll be blocking a bunch of valid servers, then.
For instance, one of my clients is a largish nonprofit organization whose Verizon business DSL in is 70.22/16. As it happens, I now handle their outbound mail, but we used to send it directly from their address.
Last year my smtp server was arbitrarily blocked by Verizon at about the same time it blocked all of Europe. I dutifully filled in the whitelist form, waited a week or so, and was apparently added to the whitelist, or removed from the blacklist.
Everything was fine until a month or two ago when one of my college classmates said he wasn't receiving any of the listserver messages I handle for the class. Sure enough, he was on Verizon, so I once again dutifully filled in the whitelist form, etc., and was cleared in a couple of days.
Last night one of my clients was unable to reach a Verizon customer and forwarded to me the SMTP error reply that was posted in this article. So, once again, I have filled out the form and am waiting breathlessly for their response.
Now I'm not opposed to blacklisting IP's or IP address blocks; I do so myself. But my blocks are of the form "don't take mail allegedly From someone@aol.com unless it originates on a server in *.aol.com". I do block whole swaths of IP space that belong to clueless cable, DSL, or dialup providers like terra.com.br or tiscali.fr. But I'm also very careful to limit those blocks to subdomains that are clearly customer-premises lines. I worked hard over the years to customize the block list very carefully to avoid false positives. The result is to block about 40% of incoming messages at the smtp level.
That being said, there's no reason Verizon should be blocking my server. I'm connected over a ATT-provided business T1; I'm not on any anti-spam blacklist (see http://www.robtex.com/ for a nice one-stop RBL check); I maintain SPF records for all my domains. Yet they have arbitrarily blocked my server on three separate occasions.
My fear is that they respond to customer complaints and blacklist servers because someone received a message with a forged From address. One or another of the domains I host gets forged from time to time. This happened recently to one of my customers. Perhaps a Verizon customer got one of these forged messages, complained to Verizon, and they put my server on the block list (despite the SPF record I maintain for the domain that was forged).
Unfortunately Verizon seems to treat its anti-spam practices as some sort of trade secret. Despite the furor over blocking mail from Europe last year and the class action suit that ensued, they still provide no information about the criteria they use to block senders. And, of course, there's absolutely no information about how to contact a human being to fix any of these problems. In fact, unless Verizon's web site has been changed, there's no way to find the whitelisting form either. You have to know the secret URL or use a Google search to bring up the form.
It's especially unfortunate since I live in Massachusetts, which is Verizon country, so a number of my clients have correspondents with Verizon addresses.
Stranger grabs Billy and forces him into The Van With No Doors and No Windows
Stranger drives off.
I feel impelled to point out that events like these are very rare yet most parents seem to believe their children face perils like these every day. (Yes, I am a parent of a 14-year-old girl who, studies show, is a much more likely target for predators than younger children or boys.)
So how about some statistics? Unfortunately the data are quite spotty and the best studies cover the late 90's. The best summary I can find is http://www.ncjrs.gov/html/ojjdp/nismart/05/index.h tml which reports the following:
Key findings presented in the NISMART Bulletin National Estimates of Missing Children: An Overview include the following:
The total number of children who were missing from their caretakers in 1999 (i.e., their caretakers did not know their whereabouts and were alarmed for at least an hour while trying to locate them) is estimated to be 1,315,600.
Nearly all of the caretaker missing children (1,312,800 or 99.8 percent) were returned home alive or located by the time the study data were collected. Only a fraction of a percent (0.2 percent or 2,500) of all caretaker missing children had not returned home or been located, and the vast majority of these were runaways from institutions who had been identified in the survey of juvenile residential facilities.
The number of missing children who were reported missing in 1999 (i.e., reported to the police or missing children's agencies in order to locate them) was estimated to be 797,500, which is equivalent to a rate of 11.4 children per 1,000 in the U.S. population.
Most of the caretaker missing children became missing because they ran away (48 percent) or because of benign misunderstandings or miscommunications about where they should be (28 percent).
Children who were missing because they became lost or injured accounted for 15 percent of all caretaker missing children.
Less than one-tenth (9 percent) of caretaker missing children were abducted by family members, and only 3 percent were abducted by nonfamily perpetrators. [emphasis mine]
Unfortunately information like this doesn't drive people to watch television news shows as does the occasional high-profile kidnapping. You're not going to see many stories about how Janie was reported missing by her parents and turned out to have stayed too long at a friend's house.
Please note that I am not suggesting that crimes against children are not significant. Kidnappings, however, make up a very small fraction of crimes against children. According to http://www.ncjrs.gov/pdffiles1/ojjdp/181161.pdf, kidnapping represents about 1-2% of all violent crimes against children. From the report, "kidnapping is dwarfed by the much more common crimes of simple and aggravated assault, larceny, and sex offenses, which make up most of the crimes against juveniles."
I'm opposed to letting ISPs have *any* "control over the content they deliver." The Internet was designed from the outset as a structure with the intelligence in the nodes and a neutral transport mechanism in between. Over the past decade we've seen how successful a model this really is. Any system which gives the telecom providers the ability to discriminate against certain types of traffic will almost certainly retard innovation. Where would Skype and Vonage be if the Internet had been designed to allow the telecom providers to discriminate? VOIP would be a service that the telcos used to move traffic around their networks, not a service that endusers can employ to reduce the cost of voice telecommunications.
It almost makes you wish that TOS field didn't exist in the IP headers!
We're consistently running into questions about where "content" begins and "carriage" ends. For instance, suppose a cable ISP systematically degraded the carriage of packets originating with IP source addresses in a competitor's network block? Is there anything in current regulation that forbids such activities? Can supposed quality-of-service prioritization rely on information in the data packet itself, or just the headers, or just certain parts of the header? At what point does prioritizing service run afoul of discriminating on the basis of the content being delivered?
How about because the market for PCs in advanced industrial economies is reaching the saturation point? If you're Intel or Microsoft, where does your future lie? Of course they're hoping to convince all of us to upgrade to dual-core CPUs and Vista, but I'd bet there are well-placed executives in both companies who see the developing world as the true next market for computing technologies.
I don't much altruism at work here, just hard-headed business logic.
I think part of the problem remains that people don't understand email technology. They look at the From: header and think it's meaningful. I agree that they also don't understand the "social implications" of the technology, but the technology itself is still opaque.
I'm sure many people imagine that email works more or less like postal mail. If a letter arrives in the mail in a Citibank envelope on Citibank stationery most of us assume that it's really from Citibank. So why shouldn't they think that an email on Citibank "stationery" (logos, etc.) claiming to be "From: security@citibank.com" isn't really from Citibank?
I also agree with the earlier poster who observed that the banks themselves have not done enough to educate their users. I'm sure they hope as well that some technological magic bullet will be invented to make this problem go away.
Slashdot Mirror
The reason that installing things is so hard on Fedora is that there is a dissapointing amount of packages available for it.
Sigh. Yet another flame war over package management.
Other than packages which cannot be legally distributed with Fedora, like mp3 or dvd players, what other critical packages are missing from Fedora? I mean ones that a large number of ordinary Linux users might want to use.
I've used every Fedora version since FC1, and though I have issues with it, not being able to find the packages I need isn't typically one of them.
I've used cn-kr.blackholes.us with SpamAssassin to give extra points to messages originating in China and South Korea. Obviously it would make sense to do the same for Taiwan. However a query for tw.blackholes.us returns NXDOMAIN. I've also looked at countries.nerd.dk for a solution to this, but didn't see an obvious answer. Is anyone out there using a DNSBL to identify IPs in Taiwan? If so, what are you using?
Does it seem strange to anyone else that so many computers containing sensitive information are suddenly being reported stolen? Is it just an accident that this particular computer, containing this particular set of incredibly extensive personal information, just happened to be stolen from this person's home in Virginia?
First, how would someone know that this computer contained all this information? Perhaps this is a job for spyware. It's easy to imagine a piece of malware that looked for large personal databases and phoned home when they are found. Or, perhaps, people whose jobs gives them access to personal information are being trailed and their computers then stolen?
Yes, I know that laptop theft is pretty endemic, but the number of high-profile thefts like this one, the one's involving the auditors Ernst & Young, etc., makes one wonder if there isn't some type of sophisticated targeting going on. I realize that the pressure to disclose such thefts has risen greatly in recent months, in large part due to laws like California's that require notification. (Laws which, by the way, the Republican Congress is seeking to preempt through federal legislation.) So this could just be a result of increased reporting, but the targets involved seem to have particularly juicy caches of data.
Am I being paranoid?
Every time RedHat is discussed here, some bozo makes a comment like the one you just refuted. I guess the rule here is that if anyone makes any money from open-source software, they become, by definition, a "parasite" on the community. I'm sorry, but I don't object to the fact that Red Hat has managed to create a business from Linux and other OSS products; in fact, I've encouraged people to own their stock. Good for them. I've used every RedHat version from 4.0-9.0 and Fedora Core 1-5 at some time or other (and a couple of WhiteBox and CentOS respins as well). Sure there are things about these distributions that bother me (the over-emphasis of Gnome, for instance), but not the fact that Red Hat has succeeded as a business.
Usually these comments sound like sour grapes to me.
It seems hard to believe that only two out of thousands were serious enough to prosecute.
And in one of those cases, the crime involved selling an FBI agent's medical records. Wonder why the Justice Department (in which the FBI is housed) chose to prosecute that case?
I was rather puzzled by this comment, given that in the mid 1990's, Gates focused on acquiring the digital rights to lots of content around the globe. Encarta, Corbis, the acquisition of the Bettman Archive, were all parts of this strategy. Here's a good take on this I found from The Chronicle of Higher Education in 1998:/ issue-33.dir/33a03301.htm.
http://chronicle.com/data/articles.dir/art-44.dir
Perhaps they just weren't capable of monopolizing sufficient content to make it a viable business strategy.
Back when I worked as a television researcher in the 80's it was becoming clear that content holders had more and more leverage over distribution channels as the networks' oligopoly was challenged. At the time I thought that content would be king. Now, I'm not so sure. Perhaps the real power these days are services like eBay or Google that repackage content in useful ways. Of course, in 1981 we couldn't see the Internet coming, so information flow was still pretty much unidirectional. The Internet enables providers like eBay to repackage content provided by its customers; essentially the audience has become the advertisers. To me, the fact that the Internet has destroyed this type of top-down information flow is one of its most revolutionary characteristics.
Of course, that's not to say that those wedded to the top-down model won't do everything they can to stop these trends, be they the MPAA, the New York Times, or the government. In the long run, though, I don't think they'll succeed.
I really didn't know; that's why I asked. I don't use Windows much anymore so I don't always keep up-to-date with every decision that comes from Redmond. Thanks for giving me the answer.
Limiting signing to 64-bit Windows will help reduce the collateral damage involving older 32-bit hardware. As for my concerns about foreclosing smaller hardware manufacturers, the article reports that Microsoft has radically lowered the barrier to entry for hardware manufacturers. MS will issue a cert for free as long as the driver writers pay an annual fee around $500 to Verisign for a "Class 3 Commercial Software Publisher Certificate." (And, might I add, what a nice little windfall for Verisign.) On the other hand, the fact that Microsoft signs a driver will no longer indicate that the driver is being certified as reliable or secure, just that it has come from someone who holds one of these free, Microsoft-issued certificates and was able to spend $500 at Verisign. This seems rather limited protection against the threat of rootkit drivers which is the claimed rationale for the requirement that drivers be signed.
The bigger issue with Vista is that the beta versions I've tried [don't] allow the installation of unsigned drivers.
Really? Is this simply an aspect of the product being in beta, or is this going to be the way Vista works after it's released? If the latter, I can see a bunch of unhappy users down the road. Lots of device manufacturers have decided it wasn't worth the tens of thousands of dollars to have Microsoft sign their drivers (both Linksys and Netgear come to mind here). Does that mean my WiFi card that didn't have signed XP drivers won't work in Vista, even if the XP drivers are compatible?
Maybe this is just another surveillance method. You know, design the computer to fail periodically, so when you send it back for repairs, they can remove the secret chip with all your keystrokes since the last repair and replace it with a fresh one!
Tinfoil hat is definitely on.
Is it just me, or does the first page of Mr. Klein's statement sound like it was written by someone else? The first page is full of rhetoric about "this dangerous Orwellian project," while the rest is a lot more technical. It sounds to me like Mr. Klein stepped forward and a statement was crafted between him and his attorneys.
Don't read this comment as any sort of defense of the government's actions here. It just that I found the rhetorical parts of his statement made the more technical parts seem suspect, and I fear a jury might read it the same way.
No, they'll be playing on a fifty-yard field.
This will be the new "Mini" series to complement the "Street" games.
All the comments in this thread so far have ignored the issue of software redistribution. The audience for the $100 laptop needs to be able to obtain and redistribute all its software freely. For instance, having a place (in a school, a Internet cafe, etc.) where you can connect this laptop and install anything it needs. That won't work unless all the software permits unlimited redistribution with no strings attached.
I just installed Fedora Core 4 and 5 (see below) on my daughter's Winbook W235 http://www.winbook.com/notebooks/w/w_overview.html . This is a very nice machine for $900 with a wide screen, DVD writer, 512M, and an 80G drive. It uses an Intel motherboard with the 855GM graphics adapter and 2200 wifi. As you note you'll need to install the Intel firmware, either from Livna http://rpm.livna.org/ or directly from Intel's site on Sourceforge http://ipw2200.sourceforge.net/firmware.php. Once I had the firmware installed, I could configure the wireless card with the included utilities in GNOME or KDE, and I didn't have to play with ndiswrapper or any other kludge.
The screen is more problematic. The 855 (and later 900+) adapters support a number of resolutions, but they weren't recognized by the normal drivers. The screen on this machine is 1280x768, but an out-of-the-box install forced it to 1024x768 which resulted in very ugly fonts. Luckily there's a little utility out there called 855resolution http://perso.wanadoo.fr/apoirier/ that you can load in rc.local and kick the adapter into other modes. Once I installed that, the KDE "Display" control saw the new resolution, and the Fedora desktop looked as nice as the Windows one.
We don't play games on this machine (we're both console types), so I can't speak for its graphics performance. It works great with Xine, though; movies and anime in the widescreen aspect ratio look terrific. Since I run KDE, I use the arts drivers to handle sound; they work fine.
OK, now for the big problem. FC4 installed just fine off the DVD, but FC5 would not install at all. (I've filed a bug report with the Fedora folks.) I finally installed FC5 by installing FC4 first, then running a system upgrade from the FC5 repositories http://www.fedoraproject.org/wiki/YumUpgradeFaq. This worked fine; it just added a few hours to the installation since yum had to update a couple thousand RPMs.
Finally, here's a hint if you buy a laptop like this one with Windows pre-installed. First, download and boot a copy of the Knoppix live CD or DVD http://www.knopper.net/. Once it's up and running, run "qtparted" from the command prompt. This is a nice graphical frontend to parted that will allow you to resize your partitions without having to buy something like Partition Magic. If you've used the Windows partition at all, I'd recommend running its disk optimizer to push all the Windows files to the front of the partition. Then you can lop off a chunk at the back for Linux.
Most US citizens are in rural areas
I usually don't comment about statements like this, but this one is so clearly erroneous that I have to speak up. According to the Census Bureau http://www.census.gov/prod/2002pubs/censr-4.pdf, 80% of Americans lived in a "metropolitan" area in 2000. Fifty percent live in suburbs and the remaining 30% live in central cities. Even accepting the fact that some of the suburbs have rural portions, it's simply not the case that "most" Americans live in rural areas.
Over the past few months, I've seen half a dozen stories about stolen laptops. In most of these cases, the laptops had thousands of individuals' data records on them, usually including SSNs. I can't see any possible rationale for someone to be walking around with that much confidential information on a laptop. In general, these breaches arise when senior management is unwilling to put customers' privacy ahead of their employees' convenience. How happy would you be to know your insurer, accountant, etc. is out there walking around with your personal health data or tax returns on a laptop hard drive with no encryption?
especially if they are in an area that doesn't have internet / communications availability
Um, I doubt there are many places that don't have a telephone. People can use modems to connect to the main office, you know.
Wow, how non-obvious is that. Of course I would expect a *graphical* file manager to open an address box when you type "/". What's so bad about having an address box at the top of the screen? Even Windows Explorer lets me type in a filepath or UNC name. Konqueror works fine for me, thank you.
1 5. I don't understand why everyone suddenly wants to make the keyboard obsolete.
The Mozilla problem has been sitting in the bugzilla queue for quite some time now, too https://bugzilla.mozilla.org/show_bug.cgi?id=3088
I wished that mattered. Unfortunately there's a long line of Supreme Court decisions and the like which say that if Congress votes the money for a foreign conflict, it's tantamount to declaring war. Having lived through five undeclared wars (Korea, Vietnam, Iraq/Kuwait, Afghanistan, and Iraq again) and countless "police actions" (Nicaragua, Panama, etc.), I don't expect we'll ever see a "declared" war again.
The Administration's rationale for avoiding FISA has nothing to do with the Iraq war. From TFA,
Elsewhere in the document, however, the government said President Bush had explained that after the attacks of Sept. 11, 2001, he authorized the security agency to intercept communications into and out of the United States by people linked to Al Qaeda and related organizations. The agency is ordinarily prohibited from intercepting the telephone and digital communications of American citizens without a warrant from a special intelligence court.
This was clearly a "they blew up the trade center, let's bug their phones" type of reaction.
First, I didn't have anything to do with their choice of providers. I wasn't consulted on this decision; I would have suggested a different provider. But they don't have a lot of money, and Verizon offered them a pretty good deal. As I said, they send no mail from this machine; it all comes through my servers. They just use Verizon for outbound connectivity.
As a matter of fact, I have no sympathy for Verizon. As I posted elsewhere in this thread I've been blocked repeatedly by their smtp servers for no good reason. But I also don't expect most organizations, and particularly small businesses and nonprofits, to have a clue about whether a particular provider is a "good" or "bad" Internet citizen.
If you're suggesting that I don't accept mail from the wide array of addresses you proposed, I'd be out of business. I don't accept mail from all Verizon hosts, just like I don't accept mail from all Comcast hosts, terra.com.br hosts, or countless other hosts from around the world. But, in general, I don't ban entire blocks of the IP space. I use an SMTP daemon that reverse-resolves each sending host and examines its ruleset to decide what to do with that message. So I don't accept messages sent from some-random-host.verizon.net that sends a 'MAIL FROM: xyzzy@aol.com' during the SMTP transaction. But I also don't block 70.22/16 either.
For a mail provider it's much more important to insure that legitimate mail is delivered to their paying customers than it is to "send a message" to a company like Verizon. What's more likely to hurt Verizon is its own customers suddenly being cut off from e-mail correspondents that they hitherto had no problem reaching. The more Verizon engages in reckless anti-spam policies, the more they'll annoy, or even enrage, their own subscribers.
Well, you'll be blocking a bunch of valid servers, then.
For instance, one of my clients is a largish nonprofit organization whose Verizon business DSL in is 70.22/16. As it happens, I now handle their outbound mail, but we used to send it directly from their address.
Last year my smtp server was arbitrarily blocked by Verizon at about the same time it blocked all of Europe. I dutifully filled in the whitelist form, waited a week or so, and was apparently added to the whitelist, or removed from the blacklist.
Everything was fine until a month or two ago when one of my college classmates said he wasn't receiving any of the listserver messages I handle for the class. Sure enough, he was on Verizon, so I once again dutifully filled in the whitelist form, etc., and was cleared in a couple of days.
Last night one of my clients was unable to reach a Verizon customer and forwarded to me the SMTP error reply that was posted in this article. So, once again, I have filled out the form and am waiting breathlessly for their response.
Now I'm not opposed to blacklisting IP's or IP address blocks; I do so myself. But my blocks are of the form "don't take mail allegedly From someone@aol.com unless it originates on a server in *.aol.com". I do block whole swaths of IP space that belong to clueless cable, DSL, or dialup providers like terra.com.br or tiscali.fr. But I'm also very careful to limit those blocks to subdomains that are clearly customer-premises lines. I worked hard over the years to customize the block list very carefully to avoid false positives. The result is to block about 40% of incoming messages at the smtp level.
That being said, there's no reason Verizon should be blocking my server. I'm connected over a ATT-provided business T1; I'm not on any anti-spam blacklist (see http://www.robtex.com/ for a nice one-stop RBL check); I maintain SPF records for all my domains. Yet they have arbitrarily blocked my server on three separate occasions.
My fear is that they respond to customer complaints and blacklist servers because someone received a message with a forged From address. One or another of the domains I host gets forged from time to time. This happened recently to one of my customers. Perhaps a Verizon customer got one of these forged messages, complained to Verizon, and they put my server on the block list (despite the SPF record I maintain for the domain that was forged).
Unfortunately Verizon seems to treat its anti-spam practices as some sort of trade secret. Despite the furor over blocking mail from Europe last year and the class action suit that ensued, they still provide no information about the criteria they use to block senders. And, of course, there's absolutely no information about how to contact a human being to fix any of these problems. In fact, unless Verizon's web site has been changed, there's no way to find the whitelisting form either. You have to know the secret URL or use a Google search to bring up the form.
It's especially unfortunate since I live in Massachusetts, which is Verizon country, so a number of my clients have correspondents with Verizon addresses.
I feel impelled to point out that events like these are very rare yet most parents seem to believe their children face perils like these every day. (Yes, I am a parent of a 14-year-old girl who, studies show, is a much more likely target for predators than younger children or boys.)
So how about some statistics? Unfortunately the data are quite spotty and the best studies cover the late 90's. The best summary I can find is http://www.ncjrs.gov/html/ojjdp/nismart/05/index.h tml which reports the following:
Key findings presented in the NISMART Bulletin National Estimates of Missing Children: An Overview include the following:
Unfortunately information like this doesn't drive people to watch television news shows as does the occasional high-profile kidnapping. You're not going to see many stories about how Janie was reported missing by her parents and turned out to have stayed too long at a friend's house.
Please note that I am not suggesting that crimes against children are not significant. Kidnappings, however, make up a very small fraction of crimes against children. According to http://www.ncjrs.gov/pdffiles1/ojjdp/181161.pdf, kidnapping represents about 1-2% of all violent crimes against children. From the report, "kidnapping is dwarfed by the much more common crimes of simple and aggravated assault, larceny, and sex offenses, which make up most of the crimes against juveniles."
I'm opposed to letting ISPs have *any* "control over the content they deliver." The Internet was designed from the outset as a structure with the intelligence in the nodes and a neutral transport mechanism in between. Over the past decade we've seen how successful a model this really is. Any system which gives the telecom providers the ability to discriminate against certain types of traffic will almost certainly retard innovation. Where would Skype and Vonage be if the Internet had been designed to allow the telecom providers to discriminate? VOIP would be a service that the telcos used to move traffic around their networks, not a service that endusers can employ to reduce the cost of voice telecommunications.
It almost makes you wish that TOS field didn't exist in the IP headers!
We're consistently running into questions about where "content" begins and "carriage" ends. For instance, suppose a cable ISP systematically degraded the carriage of packets originating with IP source addresses in a competitor's network block? Is there anything in current regulation that forbids such activities? Can supposed quality-of-service prioritization rely on information in the data packet itself, or just the headers, or just certain parts of the header? At what point does prioritizing service run afoul of discriminating on the basis of the content being delivered?
How about because the market for PCs in advanced industrial economies is reaching the saturation point? If you're Intel or Microsoft, where does your future lie? Of course they're hoping to convince all of us to upgrade to dual-core CPUs and Vista, but I'd bet there are well-placed executives in both companies who see the developing world as the true next market for computing technologies.
I don't much altruism at work here, just hard-headed business logic.
I think part of the problem remains that people don't understand email technology. They look at the From: header and think it's meaningful. I agree that they also don't understand the "social implications" of the technology, but the technology itself is still opaque.
I'm sure many people imagine that email works more or less like postal mail. If a letter arrives in the mail in a Citibank envelope on Citibank stationery most of us assume that it's really from Citibank. So why shouldn't they think that an email on Citibank "stationery" (logos, etc.) claiming to be "From: security@citibank.com" isn't really from Citibank?
I also agree with the earlier poster who observed that the banks themselves have not done enough to educate their users. I'm sure they hope as well that some technological magic bullet will be invented to make this problem go away.