I drive an SUV. It is a 4Cyl powered machine that weighs about 3300lbs. The mileage is "acceptable" if not a bit disappointing. I have also not hit anybody nor have I been hit, outside of minor parking dings when I return to my car.
However, I have been in car accidents and the majority of them were with sedans where the driver was inattentive or downright moronic.
The trouble you have is not with SUVs but with the people who drive them. Sure, some of them may be more inclined to purchase an SUV, but trust me, they are hardly status symbols anymore. I got mine simply for the utility of it and the AWD features, as I often have no choice but to make it to work (Datacenter) and I can get a good deal of snow on the ground where I live.
Then realize that when you drive a compact or subcompact and have a mechanical failure that puts your vehicle out of your control while it is going in excess of 35 miles per hour, if you kill a pedestrian, you should be charged with manslaughter, because you knew that you were in a car.
That sounds ridiculous, and it is. Accidents happen. People who fail to realize that the world is a chaotic place outside the control of civilized or even uncivilized society will only be upset when they are shown evidence of this.
Cars do not cause accidents, guns do not cause murder, pencils do not cause spelling errors and pie does not cause obesity. The actions undertaken with the use of the "tool" is the cause and the perpetrator is to blame, not the devices. If there were no car, there would be carriage accidents. If no gun, there would be knife attacks. If no pencil, then coal would be used to misspell things on cave walls. If no pie, they would simply have to eat cake:).
Because it is impossible for a partial blowout of a tire to force a 5000lb SUV into a 1900lb compact? Why is it that when an SUV owner gets into an accident, it is because they are aggressive? You want to talk aggressive, talk to all the 530i penis compensators who drive like they are on their own personal autobahn.
HP and Intel joined up for the EPIC (Explicitly Parallel) architecture. AMD HAS innovated, by the way. Just because you choose to ignore or make light of it does not change that fact.
AMD went to on die memory controllers for x86 CPUs. They did away with the front side bus with their HT (formerly LDT) system, they have also gone on to be the first to 1Ghz in the x86 arena, using alpha technology and also, their 3DNOW! technology was one of, if not THE first SIMD architecture to handle FP calculations.
All this is NOT BAD for a company whose total worth is less than Intel spends on marketing.
I don't know where you were buying your CPUs, but I have had exactly 1 DOA processor from AMD in dealing with oh.. dozens of dozens.
Often times, what appeared like a bad CPU was actually a motherboard issue with voltage not being applied correctly. I found this on a few KT266A boards where the BIOS would be set for 1.65v but instead supply 1.58v, or barely enough to boot. The fix for this, was to set the BIOS voltage to 1.7 and usually achieve a voltage of 1.6+ which was sufficient most times.
This was not exactly AMD's fault, per se, but to the consumer, they don't care. It would be like a customer knowing if you take a Ford to Jiffy Lube, nothing goes wrong, but they can't get their wrenches in to GM cars and get the nut back on right after an oil change. Jiffy Lube doesn't get the major blame, GM does, because nobody with Fords is having the problem....
Intel was a single source supplier for CPUs. IBM wanted a second source or they would not deal with Intel. Intel then sourced production of pre 486 CPUs to AMD. However, they did not restrict AMD from selling them as their own, which they did. Then, AMD was developing their own chips based on the instruction set and specification that intel developed. Intel sued for trademark infringement, knowing that AMD had the license to produce likewise chips. The courts in the US ruled that Intel could not trademark a number, which is why there was no 586, but rather the "Pentium" with the 5 prefix Pent. This is a trade-markable name.
Being more agile than Intel, and being willing to accept thinner margins than Intel, AMD and competitors were pricing very attractively to OEMs. Intel, however, looked disfavorably on this. They punished their customers with "shortages" of their chips and chipsets, knowing it would allow their customer's competitors, also their customers, to gain an upper hand. They also offered special pricing, not for volume, but for "loyalty." They would give their customers a break if they were 100% intel customers and not "Buy 10,000 units and get 200 free, which would likely have been legal.
The issue is not substandardness nor the inability to compete. Instead, it was that after the original Athlon, AMD was able to out maneuver the challenges that intel through in its way and was able to out innovate them in many areas. The FSB that intel still uses can cause memory bottlenecks as well as poor scaling to multiple sockets and cores, but that is a topic for another discussion. Intel was abusing their customers, their competitors, and consumers with their methods of market manipulation. But, to quell your intent to show that AMD et al were simply riding on Intel's coat tails, ask yourself "Who built the spec to extend x86 to 64bits with extended register counts?"
Monopolies, in markets like this, are not meant in such that there are 0 competitors. However, when a company becomes so large that it can sway the market on its whims, then it becomes abusive, and therefor detrimental to consumers and competition.
Intel has been using their size, money and influence to keep competitors out of use in their customer's systems. This is anti competitive, and when on a scale of this size, is considered monopolistic. Intel owns over 80% of the microprocessor market, plus they design specs for systems, such as their PCI spec.
If Intel is guilty of keeping other processors out of machines by being anti competitive, they are going to see some sanctions and fines.
Windows XP? Try using it Pre SP1 where many drivers cards, especially 802.11b cards were new. Windows 2000, still a major player in the corporate market, does the same thing.
How lazy have we become when "I have to download a driver for my sound/video/network?" is a dealbreaker? Most often, the incorporated driver on MS platforms is older, feature reduced and possibly buggy, due to it being 2 or more revisions older than release drivers.
the NWN games themselves sucked. The Baldur's Gate predecessors were far more involving from a storyline standpoint. The good thing about NWN, however, was that you could use the engine and build tools to create your own games. I HIGHLY recommend the Adam & Jamie games (No, nothing about Mythbusters)
The cost to make something is not always the only cost associated with that. Sure, the materials may cost $19 and then the assembly and electricity to product it, say another $7. However, who pays the engineers who come up with the design? Who pays testers or programmers who make sure the phones work?
They are selling a distribution that actually does seem to "get along better" with a windows environment right out of the box. Sure, you can do the same things with Red Hat, but it is just a tad easier with the features in YaST to get things done.
They are not MS' bitch, Trojan horse, or anything else. I may be wrong on this, but Novell contribute a "Metric Shit Ton" towards F/OSS software and are champions of it, in general. The deal with MS is being blown way out of proportion.
You could always give him the benefit of the doubt and assume that management heard his concerns and said "We'll assume all risks" which means "You take the fall when my manager asks why shit went wrong."
I have been looking at this, but have heard from some colleagues that Xen performance is not up to par with solutions like VMware when you don't use the Xensource drivers. Have you had any other experiences than theirs? How is I/O performance compared to native machine performance? Does it abstract the hardware the same way (relatively) VMware does?
The indemnification is basically MS' promise to not go after Novell's customers. I agree they sold them "magic beans" but I have to assume A customer or customers wanted this. This was done under the umbrella of SCO's lawsuits, etc. Some customers probably did not want the hassle and wanted to have somebody shield them. Maybe it was no better than a policy against alien abduction, but shit, Lloyds of London will sell you one if you want them to.
So Novell Pays MS and Novell gets paid MORE than that BY MS.
I realized it was not an elegant solution, but I will definitely have to look at that write up. I have bounced around at companies a bit lately and will likely stay where I am because I finally kinda like it here. However they are RH and not even RH and SLES.
Regardless, The last place I worked was a big shop that did hosting and client services (large UK company's presence in the US). They had a client with a RH cluster that they kept on complaining about and having it fence off nodes, etc. Of course they "hated" SUSE for some inexplicable reason. I tried to point them to Heartbeat and even http://www.drbd.org/ but they thought they were "Not big enough to use here" and "Red Hat Cluster is a real product" meanwhile GFS has poor locking, in my opinion.
From my various experiences with both, Novell's support IS better.
In getting somebody on the phone, they are about a tie in time, but Novell edges them out in hold music.
In getting solutions, RH loves to point you to articles, Novell likes to get lots of logs, but goes a bit further sometimes. I have 2 stories to share on that.
At one prior company that switched from MS infrastructure servers to Linux, they moved from RHEL 3 to SLES 9 and 10 on VMware. I had an issue with logging into their servers via SSH using the company approved terminal program (not free, but oru version was not in support, being about 2 versions out of date. PuTTY worked fine.) After calling their support and escalating, their 2nd level guy said he'd call me back. A few hours later, having downloaded and installed the trial for the terminal program, he gave me the settings I had to change in the sshd_config to make it work.
At the same company as above, we had issues using SAMBA/Winbind to authenticate users to the server. It kept losing kerberos tickets in our environment. We sent various logs to them and finally were sent to 3d level support. They shortly sent us to engineering support and issued us a patch for "our" environment and told us to use this version of SAMBA and to email them when the next version alert for it was sent to us with the reference to this case so they could check the change logs and backport the fixes they had implemented when/if we wanted to upgrade.
Hell, I love their cool solutions pages and even use the novell docs sometimes to get things done on Redhat, due to their being more informative.
"They can wish in one hand and take a crap in the other, and see which gets filled first"
I don't think MS is trying to play nice. Hedging your bets is not about "Ok guys, I can't stack the deck anymore, lets play fair."
It is simply "How can we make money off this thing we can't control while we figure out how to fsck them in some way?"
I think Novell got smarter to the ways of the beasts within MS. They are not some small group of Doofus faced chuds (Ok, Miguel is). I have faith that MS is not the dragon they once were, but not from their lack of trying.
You may also want to look at the free VMware Server edition. Having multiple servers that require say >300MB of RAM on a nice 4GB or greater new server is nice. If you have any sort of shared storage like a SAN or even a fast NAS, you can even have the LUNs/Shares mapped to multiple machines. That way, while not "elegant" you can have a hardware failure and then launch the VM on the 2nd box.
Speaking of security, I kinda like 2 things about Apparmor vs SELinux (which oddly, Canonical chose for Ubuntu, over SELinux). One, the control follows the relative path to the file as opposed to the inode. Now, some may like it the other way around, but if you update the executable, as long as the path stays the same, no changes have to be made. This increases the chance of an administrator NOT forgetting to update the settings for the restrictions during an update or for a patch where they have not gotten all the details. The other reason...It is not written by the NSA. Call me nutty, but I don't exactly like the involvement of government in my software.
I understand the paradigm but until that network access is faster and ubiquitous, I will want full control over MY system:). The benign overlords will always be benign? I like the ability to say "Hey, Google is pissing me off, let me try Ask or Yahoo instead. I prefer the power being in my hands.
Slashdot Mirror
Wow, an anecdotal account. I see yours and recount with a person who had 2 blowouts at different times and ended up in the divider both times.
I drive an SUV. It is a 4Cyl powered machine that weighs about 3300lbs. The mileage is "acceptable" if not a bit disappointing. I have also not hit anybody nor have I been hit, outside of minor parking dings when I return to my car.
However, I have been in car accidents and the majority of them were with sedans where the driver was inattentive or downright moronic.
The trouble you have is not with SUVs but with the people who drive them. Sure, some of them may be more inclined to purchase an SUV, but trust me, they are hardly status symbols anymore. I got mine simply for the utility of it and the AWD features, as I often have no choice but to make it to work (Datacenter) and I can get a good deal of snow on the ground where I live.
That sounds ridiculous, and it is. Accidents happen. People who fail to realize that the world is a chaotic place outside the control of civilized or even uncivilized society will only be upset when they are shown evidence of this.
Cars do not cause accidents, guns do not cause murder, pencils do not cause spelling errors and pie does not cause obesity. The actions undertaken with the use of the "tool" is the cause and the perpetrator is to blame, not the devices. If there were no car, there would be carriage accidents. If no gun, there would be knife attacks. If no pencil, then coal would be used to misspell things on cave walls. If no pie, they would simply have to eat cake
Because it is impossible for a partial blowout of a tire to force a 5000lb SUV into a 1900lb compact? Why is it that when an SUV owner gets into an accident, it is because they are aggressive? You want to talk aggressive, talk to all the 530i penis compensators who drive like they are on their own personal autobahn.
AMD went to on die memory controllers for x86 CPUs. They did away with the front side bus with their HT (formerly LDT) system, they have also gone on to be the first to 1Ghz in the x86 arena, using alpha technology and also, their 3DNOW! technology was one of, if not THE first SIMD architecture to handle FP calculations.
All this is NOT BAD for a company whose total worth is less than Intel spends on marketing.
Often times, what appeared like a bad CPU was actually a motherboard issue with voltage not being applied correctly. I found this on a few KT266A boards where the BIOS would be set for 1.65v but instead supply 1.58v, or barely enough to boot. The fix for this, was to set the BIOS voltage to 1.7 and usually achieve a voltage of 1.6+ which was sufficient most times.
This was not exactly AMD's fault, per se, but to the consumer, they don't care. It would be like a customer knowing if you take a Ford to Jiffy Lube, nothing goes wrong, but they can't get their wrenches in to GM cars and get the nut back on right after an oil change. Jiffy Lube doesn't get the major blame, GM does, because nobody with Fords is having the problem....
http://h20219.www2.hp.com/integrity/cache/342254-0-0-0-121.html
or http://www.sun.com/servers/x64/x4600/specs.xml
PCs may have higher shipping volume, but servers are no slouch either, and produce higher margins.
Intel was a single source supplier for CPUs. IBM wanted a second source or they would not deal with Intel. Intel then sourced production of pre 486 CPUs to AMD. However, they did not restrict AMD from selling them as their own, which they did. Then, AMD was developing their own chips based on the instruction set and specification that intel developed. Intel sued for trademark infringement, knowing that AMD had the license to produce likewise chips. The courts in the US ruled that Intel could not trademark a number, which is why there was no 586, but rather the "Pentium" with the 5 prefix Pent. This is a trade-markable name.
Being more agile than Intel, and being willing to accept thinner margins than Intel, AMD and competitors were pricing very attractively to OEMs. Intel, however, looked disfavorably on this. They punished their customers with "shortages" of their chips and chipsets, knowing it would allow their customer's competitors, also their customers, to gain an upper hand. They also offered special pricing, not for volume, but for "loyalty." They would give their customers a break if they were 100% intel customers and not "Buy 10,000 units and get 200 free, which would likely have been legal.
The issue is not substandardness nor the inability to compete. Instead, it was that after the original Athlon, AMD was able to out maneuver the challenges that intel through in its way and was able to out innovate them in many areas. The FSB that intel still uses can cause memory bottlenecks as well as poor scaling to multiple sockets and cores, but that is a topic for another discussion. Intel was abusing their customers, their competitors, and consumers with their methods of market manipulation. But, to quell your intent to show that AMD et al were simply riding on Intel's coat tails, ask yourself "Who built the spec to extend x86 to 64bits with extended register counts?"
Yeah. They would be shown some new backdoors and have their interfaces expanded to accept all sorts of new peripherals.
Intel has been using their size, money and influence to keep competitors out of use in their customer's systems. This is anti competitive, and when on a scale of this size, is considered monopolistic. Intel owns over 80% of the microprocessor market, plus they design specs for systems, such as their PCI spec.
If Intel is guilty of keeping other processors out of machines by being anti competitive, they are going to see some sanctions and fines.
You get to play as an aging Han Solo who finds an artifact and meets aliens..........wait........redundant?
How lazy have we become when "I have to download a driver for my sound/video/network?" is a dealbreaker? Most often, the incorporated driver on MS platforms is older, feature reduced and possibly buggy, due to it being 2 or more revisions older than release drivers.
http://adamandjamie.com/nwn/
R&D actually costs money.
They are selling a distribution that actually does seem to "get along better" with a windows environment right out of the box. Sure, you can do the same things with Red Hat, but it is just a tad easier with the features in YaST to get things done.
They are not MS' bitch, Trojan horse, or anything else. I may be wrong on this, but Novell contribute a "Metric Shit Ton" towards F/OSS software and are champions of it, in general. The deal with MS is being blown way out of proportion.
You could always give him the benefit of the doubt and assume that management heard his concerns and said "We'll assume all risks" which means "You take the fall when my manager asks why shit went wrong."
I have been looking at this, but have heard from some colleagues that Xen performance is not up to par with solutions like VMware when you don't use the Xensource drivers. Have you had any other experiences than theirs? How is I/O performance compared to native machine performance? Does it abstract the hardware the same way (relatively) VMware does?
So Novell Pays MS and Novell gets paid MORE than that BY MS.
Regardless, The last place I worked was a big shop that did hosting and client services (large UK company's presence in the US). They had a client with a RH cluster that they kept on complaining about and having it fence off nodes, etc. Of course they "hated" SUSE for some inexplicable reason. I tried to point them to Heartbeat and even http://www.drbd.org/ but they thought they were "Not big enough to use here" and "Red Hat Cluster is a real product" meanwhile GFS has poor locking, in my opinion.
In getting somebody on the phone, they are about a tie in time, but Novell edges them out in hold music.
In getting solutions, RH loves to point you to articles, Novell likes to get lots of logs, but goes a bit further sometimes. I have 2 stories to share on that.
At one prior company that switched from MS infrastructure servers to Linux, they moved from RHEL 3 to SLES 9 and 10 on VMware. I had an issue with logging into their servers via SSH using the company approved terminal program (not free, but oru version was not in support, being about 2 versions out of date. PuTTY worked fine.) After calling their support and escalating, their 2nd level guy said he'd call me back. A few hours later, having downloaded and installed the trial for the terminal program, he gave me the settings I had to change in the sshd_config to make it work.
At the same company as above, we had issues using SAMBA/Winbind to authenticate users to the server. It kept losing kerberos tickets in our environment. We sent various logs to them and finally were sent to 3d level support. They shortly sent us to engineering support and issued us a patch for "our" environment and told us to use this version of SAMBA and to email them when the next version alert for it was sent to us with the reference to this case so they could check the change logs and backport the fixes they had implemented when/if we wanted to upgrade.
Hell, I love their cool solutions pages and even use the novell docs sometimes to get things done on Redhat, due to their being more informative.
"They can wish in one hand and take a crap in the other, and see which gets filled first"
I don't think MS is trying to play nice. Hedging your bets is not about "Ok guys, I can't stack the deck anymore, lets play fair." It is simply "How can we make money off this thing we can't control while we figure out how to fsck them in some way?"
I think Novell got smarter to the ways of the beasts within MS. They are not some small group of Doofus faced chuds (Ok, Miguel is). I have faith that MS is not the dragon they once were, but not from their lack of trying.
http://www.novell.com/linux/microsoft/faq_opensource.html
http://news.cnet.com/Microsoft,-Novell-spar-over-Linux-agreement/2100-7344_3-6137444.html
Microsoft got Novell to agree to very very little :). This is a collaboration effort and customer indemnification.
You may also want to look at the free VMware Server edition. Having multiple servers that require say >300MB of RAM on a nice 4GB or greater new server is nice. If you have any sort of shared storage like a SAN or even a fast NAS, you can even have the LUNs/Shares mapped to multiple machines. That way, while not "elegant" you can have a hardware failure and then launch the VM on the 2nd box.
Speaking of security, I kinda like 2 things about Apparmor vs SELinux (which oddly, Canonical chose for Ubuntu, over SELinux). One, the control follows the relative path to the file as opposed to the inode. Now, some may like it the other way around, but if you update the executable, as long as the path stays the same, no changes have to be made. This increases the chance of an administrator NOT forgetting to update the settings for the restrictions during an update or for a patch where they have not gotten all the details. The other reason...It is not written by the NSA. Call me nutty, but I don't exactly like the involvement of government in my software.
I understand the paradigm but until that network access is faster and ubiquitous, I will want full control over MY system :). The benign overlords will always be benign? I like the ability to say "Hey, Google is pissing me off, let me try Ask or Yahoo instead. I prefer the power being in my hands.