My personal mailbox is 1.3G on the server, and my work mailbox is 2.7G on the server. Both are accessed via IMAPS. Neither seems to register any significant CPU utilization. Additionally, since I run both clients simultaneously, one would think I'd see even worse performance, but I don't.
I run two instances of Thunderbird 3.0 here, with two separate profiles (personal and work). Overall, my CPU 96% idle at most times. that's even with Chrome and Opera sessions open. More often I find that Opera's plugin wrapper freaks out and wildly thrashes on one of my cores. Can't say I've ever seen Thunderbird consume huge amounts of CPU time other than when I've asked it to do some massive operation (some operation on hundreds to thousands of e-mails). Even searching my entire mailbox (personal or work) doesn't cause it to consume much if any CPU time.
You know what still gets my goat? That comcast has for more than a decade had an incredibly hostile AUP that banned any form of mailing list or discussion group hosting, yet you people only started screaming about your "rights" and network neutrality when they brought the hammer down on your precious porn and TV episodes.
Then switch to a business class connection with them. All these "restrictions" and problems go away. Once you've got a business class connection, they don't care if you host services, mailing lists, or throttle your uploads after X seconds/minutes/hours.
Residential service isn't intended for hosting services. It's intended for the average PC user which has nearly zero clue how to even begin hosting any kind of service. Anything beyond basic e-mail and web browsing and they are lost (if they even make it that far). If you're not a member of that group, you'd be far better off with a business class line.
taking that money out of some other part of their budget.
You assume that they have the money somewhere in their budget to do so. Now, let's say for example they don't. And without the web presence they can't bring a service to market that other companies may need or benefit from. Now the cost is not only the mythical $Z that you reference, but any benefit or utility that would have been delivered to any customer of theirs. Economics isn't as simply as dollar figures.
Minor difference here, but MS was bundling applications that they made and abusing their monopoly position in one market for an advantage in another. Perhaps the same could be said for Apple as it is bundling its own iTunes software along with a variety of other applications. However, I don't see it as the same situation with Canonical as they didn't write most of the software they are bundling. They may have added to the software but much of what they distribute is the creation of others.
Yes, traditional Xorg multi-display does require hand editing the xorg.conf. However, it is certainly possible. I configured Dual Head (two independent X displays) prior to Xinerama being widely supported using a variety of different PCI video cards.
As others have pointed out, it is certainly easier with a single card that supports multiple displays. However, this is by no means a requirement.
As a few others have indicated, how multi-monitor setups are configured largely depends on which video card you have. This isn't to say that there isn't a standard way to configure multiple displays in Xorg. However, when either an ATI or Nvidia card is used, their binary drivers tend to provide better support for the card and additional options with regard to multiple displays.
If you let us know which card you have, I'm sure that someone here can assist you with configuring it.
NTP won't run as often as it should, so my recordings are off by 20-30 secs and I have to login and manually run ntpdate
Why not have it run via cron? Also are you running an ongoing ntp daemon to continually check the time and slew the clock if needed? Doesn't sound like you are.
Has anyone been successful in prototyping a Mythbox (such that it just works for long periods of time without having to worry about tweaks and workarounds)? If so, please tell me how.
Yes, several people have. I've had two different MythTV boxes up and running for prolonged (roughly year) time frames. The only reason they didn't continue service longer was my desire to tinker with them, not need. As for how, I use Debian and the Debian packages for MythTV provided here.
All RSS feeds should look at impementing PUBSUB instead. RSS's major flaw is that it has to constantly poll sources to find updates. This is what creates the DOS effect. If they instead used a PUBSUB method (google if you'd like to know more) those that have subscribed would be notified of an update.
They are when you use the credit/business card sized CDRs. After all you're just looking for a boot device, right? 35 MB per disk sure beats the hell out of floppy.
I used to be a Red Hat user and at one time thought it was a good choice. Now, I know this was because I simply didn't know better. I figured it was normal for an upgrade from one version to another (6.0 to 6.1) to be iffy. After all there's so much software to upgrade, and how can they know about all your system's specific customizations. So most of the time an upgrade meant a complete reinstall and retweak, especially between major versions (6.x to 7.x).
But as I said, this was simply because I didn't know any better. Eventaully, after trying several distributions, I found Debian (http://debian.org). After a very short while it became appearent that what I was experiencing in Red Hat (and thought to be normal) was completely unnecessary. Debian's thorough Policy helps ensure that system upgrades are almost always seamless, and your system specific customizations are maintained.
So, now knowing about Debian, I would never run my business systems (or any other system for that matter) on Red Hat.
Also, you ask why a long release cycle is important. How often do you want to upgrade (or in the case of Red Hat) rebuild your business servers? Having a long supported release is critical for business systems.
Most employment contracts specifically state that any thing or idea created, conceived, developed, etc. while employeed becomes property of the employeer
This is precisely why I made sure my employer changed the wording of that clause in my contract. I wanted to make sure that what I created was mine. They own what I create at their request only. Anything else I create is mine. This is as it should be, and anyone that does coding/development for a living should make sure they have rights to what they develop independently of the company.
In May 2000, a new CEO and his management team joined MandrakeSoft. With this new management team, the company dramatically increased expenses (by 400%), hired many new employees, and entered into many expensive long-term agreements. Under this new management, the plan was to build the company into a worldwide leader in the area of e-learning, with only a link to Linux since the underlying platform and contents were open-source.
By March 2001, the results of this strategy showed a marked decrease in income, while expenses increased by 400%. At its worst point, MandrakeSoft's "burn rate" was approximately 1.5 ME/month ($1.5M/month). As a result, it was soon decided to remove this experienced management team and to refocus the company's activities strictly toward Linux.
So, because they've made poor business decisions in the past (to include hiring bad management and throwing away money) we're supposed to support them? Give me a break! Sheez, by this logic, someone should just pay off my credit cards and give me a house (any takers?).
You didn't look hard enough. Asterisk http://asterisk.org works with most voice modems and has rather inexpensive hardware options (~$150 for single line in and out)
An important item to note, is that in the Linux sector, there isn't just ONE item. Linux has many similiar but distinct distributions. There are even several variations of the kernel. Mix and match to your hearts content.
To say that there is ONE Linux is to be just as blind as saying there is one Human. While they are all somewhat alike, they are all still quite different.
Slashdot Mirror
Lenovo's Thinkpad line still has 15" widescreens with 1920x1200.
My personal mailbox is 1.3G on the server, and my work mailbox is 2.7G on the server. Both are accessed via IMAPS. Neither seems to register any significant CPU utilization. Additionally, since I run both clients simultaneously, one would think I'd see even worse performance, but I don't.
I run two instances of Thunderbird 3.0 here, with two separate profiles (personal and work). Overall, my CPU 96% idle at most times. that's even with Chrome and Opera sessions open. More often I find that Opera's plugin wrapper freaks out and wildly thrashes on one of my cores. Can't say I've ever seen Thunderbird consume huge amounts of CPU time other than when I've asked it to do some massive operation (some operation on hundreds to thousands of e-mails). Even searching my entire mailbox (personal or work) doesn't cause it to consume much if any CPU time.
Firewall Builder does most of what the submitter is looking for already.
Time Machine alone is a huge reason to upgrade. Sure similar results can be had with shell scripting, but it really isn't nearly the same experience.
I see no reason not to use my real name.
You know what still gets my goat? That comcast has for more than a decade had an incredibly hostile AUP that banned any form of mailing list or discussion group hosting, yet you people only started screaming about your "rights" and network neutrality when they brought the hammer down on your precious porn and TV episodes.
Then switch to a business class connection with them. All these "restrictions" and problems go away. Once you've got a business class connection, they don't care if you host services, mailing lists, or throttle your uploads after X seconds/minutes/hours.
Residential service isn't intended for hosting services. It's intended for the average PC user which has nearly zero clue how to even begin hosting any kind of service. Anything beyond basic e-mail and web browsing and they are lost (if they even make it that far). If you're not a member of that group, you'd be far better off with a business class line.
taking that money out of some other part of their budget.
You assume that they have the money somewhere in their budget to do so. Now, let's say for example they don't. And without the web presence they can't bring a service to market that other companies may need or benefit from. Now the cost is not only the mythical $Z that you reference, but any benefit or utility that would have been delivered to any customer of theirs. Economics isn't as simply as dollar figures.
Minor difference here, but MS was bundling applications that they made and abusing their monopoly position in one market for an advantage in another. Perhaps the same could be said for Apple as it is bundling its own iTunes software along with a variety of other applications. However, I don't see it as the same situation with Canonical as they didn't write most of the software they are bundling. They may have added to the software but much of what they distribute is the creation of others.
Yes, traditional Xorg multi-display does require hand editing the xorg.conf. However, it is certainly possible. I configured Dual Head (two independent X displays) prior to Xinerama being widely supported using a variety of different PCI video cards.
As others have pointed out, it is certainly easier with a single card that supports multiple displays. However, this is by no means a requirement.
As a few others have indicated, how multi-monitor setups are configured largely depends on which video card you have. This isn't to say that there isn't a standard way to configure multiple displays in Xorg. However, when either an ATI or Nvidia card is used, their binary drivers tend to provide better support for the card and additional options with regard to multiple displays.
If you let us know which card you have, I'm sure that someone here can assist you with configuring it.
NTP won't run as often as it should, so my recordings are off by 20-30 secs and I have to login and manually run ntpdate
Why not have it run via cron? Also are you running an ongoing ntp daemon to continually check the time and slew the clock if needed? Doesn't sound like you are.
Has anyone been successful in prototyping a Mythbox (such that it just works for long periods of time without having to worry about tweaks and workarounds)? If so, please tell me how.
Yes, several people have. I've had two different MythTV boxes up and running for prolonged (roughly year) time frames. The only reason they didn't continue service longer was my desire to tinker with them, not need. As for how, I use Debian and the Debian packages for MythTV provided here.
All RSS feeds should look at impementing PUBSUB instead. RSS's major flaw is that it has to constantly poll sources to find updates. This is what creates the DOS effect. If they instead used a PUBSUB method (google if you'd like to know more) those that have subscribed would be notified of an update.
All of my new systems (built in the last 6-8 months) don't even have a floppy drive in them.
They are when you use the credit/business card sized CDRs. After all you're just looking for a boot device, right? 35 MB per disk sure beats the hell out of floppy.
I used to be a Red Hat user and at one time thought it was a good choice. Now, I know this was because I simply didn't know better. I figured it was normal for an upgrade from one version to another (6.0 to 6.1) to be iffy. After all there's so much software to upgrade, and how can they know about all your system's specific customizations. So most of the time an upgrade meant a complete reinstall and retweak, especially between major versions (6.x to 7.x).
But as I said, this was simply because I didn't know any better. Eventaully, after trying several distributions, I found Debian (http://debian.org). After a very short while it became appearent that what I was experiencing in Red Hat (and thought to be normal) was completely unnecessary. Debian's thorough Policy helps ensure that system upgrades are almost always seamless, and your system specific customizations are maintained.
So, now knowing about Debian, I would never run my business systems (or any other system for that matter) on Red Hat.
Also, you ask why a long release cycle is important. How often do you want to upgrade (or in the case of Red Hat) rebuild your business servers? Having a long supported release is critical for business systems.
This is precisely why I made sure my employer changed the wording of that clause in my contract. I wanted to make sure that what I created was mine. They own what I create at their request only. Anything else I create is mine. This is as it should be, and anyone that does coding/development for a living should make sure they have rights to what they develop independently of the company.
If you keep the derivative work in house, you don't have to GPL it.
By March 2001, the results of this strategy showed a marked decrease in income, while expenses increased by 400%. At its worst point, MandrakeSoft's "burn rate" was approximately 1.5 ME/month ($1.5M/month). As a result, it was soon decided to remove this experienced management team and to refocus the company's activities strictly toward Linux.
So, because they've made poor business decisions in the past (to include hiring bad management and throwing away money) we're supposed to support them? Give me a break! Sheez, by this logic, someone should just pay off my credit cards and give me a house (any takers?).
Feel free to mark me as troll.
You didn't look hard enough. Asterisk http://asterisk.org works with most voice modems and has rather inexpensive hardware options (~$150 for single line in and out)
An important item to note, is that in the Linux sector, there isn't just ONE item. Linux has many similiar but distinct distributions. There are even several variations of the kernel. Mix and match to your hearts content. To say that there is ONE Linux is to be just as blind as saying there is one Human. While they are all somewhat alike, they are all still quite different.