I'm not going to hold my breath for it, but it would be nice if this version came with IPv6 support enabled by default. This would include the kernel, net tools, applets such as telnet, ftp, & ssh, and Apache.
The tools for a working IPv6 host are out there. It's time for the distributions to start including the support. Once the core system is up and running with IPv6, the rest of the app developers will have some incentive to get their stuff cleaned up to start running under IPv6.
Think of it this way, if there was only one major desktop environment out there (be it KDE, GNOME, or something new), it likely wouldn't be as advanced as either of these two environments. Competition pushes progress.
My biggest fear in all this is that the developers will compromise stability for new features. Then, we start to get into the old Microsoft problems.
From the e-mail: "Of
course, we are planning to have a kpart for Mozilla too so konqueror and others
can use it too if so desired. Again, we are just providing an alternative while
the core KDE team work out all the bugs and standards conformance features with
their khtml implementation."
The idea isn't to replace what Konqueror is doing but provide alternatives. This is certainly a good thing. If a Konqueror user can select which rendering engine to use, it will make it much more apparent when one engine has a problem. The programmers will have even more reason to fix any bugs since they will stick out like a wart on your nose.
I doubt the KDE team will have to rewrite a lot. The whole idea of kparts is to abstract the interface to the point where one side doesn't care about the internals of the other. What me may see are a few more methods added to the existing html widgets.
They way they can make money is to sell products. The way they can sell products is to draw attention to themselves. The way they can draw attention to temselves is to create talk in the press over projects they are supporting. This is a good thing!
The quickest way for Corel to completely fade away is to stop doing new development. It's kind of like evolution, either you advance or get overrun. A company that terminates all development efforts is usually on it's last leg.
When you consider how much money Corel is burning through, how much is it really costing them to have a couple of programmers dedicated to a project like this? $200-250K/year for a company operating on the level of $20-30M/year is not that big of deal.
Unfortunately, the OSS movement has a lot of people who are a little too deep in the socialist utopia ideal. Who cares that RedHat gives away their software for anyone to download. They are evil becuase they *gasp* want to make money.
The updates will still be available on updates.redhat.com just like they always have. RedHat is targeting the subscription services at corporations. They aren't holding back any software, just tailoring their premium services.
RedHat stated a while back that 7.0 would ship with a 2.2 kernel but that everything would be ready to go for the 2.4 kernel when it became availalble. Expect to see the kernel change as just another update package in a couple of months.
Related to this, KDE 2 is supposed to go final within the next month. That's another major applicaiton set that it would have made sense to wait for. I expect to see this updated after release as well.
Odds are, people can come up with a dozen other major packages that are "close" to having a new stable version ready. If RedHat waited for all of them, they would never get a release out.
While bringing in new features is nice, consider this: How many times has your Palm crashed? One big benefit of the "keep it simple" mentality is a VERY stable platform. That's not to even mention the speed benefits.
Personally, I don't expect to see any major advances until several hardware technologies advance such as:
Better color screens w/ higher pixel densitity
Perhaps, rollup or flexible screens
Smaller, longer lived batteries
Better wireless comm (BlueTooth?)
Much larger (yet cheap) storage capability (audio recording)
Voice recognition
{far down the road} Integrated video camera with digital recording
Rather than a "cell phone module", I'd rather have BlueTooth in both my PDA and cell phone and let them talk to each other. Let the cell phone browse the PDA's adress book. Let the PDA use the cell phone for communications. Throw in a wireless, ear-clip headset and I'd be in hog heaven doing voice queries on the PDA as well as phone conversations.
I expect the use for this cable would be somewhat limited being that it takes a mono audio patch (RCA male and female connectors) and feeds it to a 1/8" stereo input. If it was stereo clear across, I could see plenty of uses for it. Having a 20' cable available is usually a good thing.
I can honestly say that I'm not impressed with the Cue:Cat. Mine showed up a week ago, and so far it's only been able to read about one in ten barcodes. Maybe mine is just defective, but if this is common then DC has a serious problem on their hands. Either the product doesn't work as well as advertised, or they have quality control issues with the ones they are sending out.
Oh, their "Convergence Cable" is nothing to write home about either (except to tell your family to avoid it). Now, I can't get my TV audio to stop going through the speakers on the computer every time I reboot. Also, the only time a show used an audio que (NBC Olympics), the web browser errored trying to find the supposedly passed link.
I'm a big fan of these mixed media tools (computer/TV/print integration), but if this is an example of what we can expect, you had better watch out.
"minimal" means that you still have to exercise some due diligence to avoid having your software exported to a few "bad" countries such as Iraq and Libya. I'm not sure that this restriction even applies to Open Source software. When you think about it, the idea is pretty futile. All you need is some sort of FTP proxy server outside of the US and anywhere in the world can then get to your file archive.
I'm one of those sick people who actually watches NASA TV during shuttle missions. It's kind of a out of this world reality TV show. Anyway, the laptops onboard do lockup from time to time. They just reboot and try again. As was said earlier, nothing mission critical is run on them.
I'm pretty sure the Shuttle uses IBM ThinkPads. I'm very sure the ISS does. This probably means that the track point on the keyboard is the preferred method of control. Of course, the astronaut would have to be strapped down or they would drift away every time they touch the keyboard.
There have been articles published in the past about Linux being used to operate experiments on the Space Shuttle. Some of the experiments have self contained computer systems for managment and data collection. Linux was ideal here.
Now, I'd like to see a Linux/Apache web server in orbit. Even if it feeds nothing but text telemetry data, it would be impressive. Hmm, is there a TLD for LEO yet?
I should have qualified my statement about ATM to say "ATM in the LAN". Most people I've talked to trying to use ATM in a building have been very unhappy and wanted nothing more than to rip it out and go to a combination of 1Gb and 100Mb Ethernet.
If you want to see how expensive FC equipment can be. Go to Dell's website and look at some of their "PowerVault" 630F and 650F storage equipment. This is all FC based, and can give you a good idea of how expensive a good array costs.
One thing you learn working around the gov't is that if you collect enough nonclassified information together you may eventually end up with a product that ends up being classified. It is quite possible that the people being interviewed did not individually expose any company secrets, but when everything was corelated together a bigger picture was revealed.
Cost. Fibre Channel equipment costs considerably more than SCSI. The idea of NAS (Network Attached Storage) is to reduce cost. This box probably has four 80GB IDE disk drives inside of it. To build something simliar using Fibre Channel would likely cost $10,000 or more.
Fibre Channel is good if you need some of the features like several servers sharing one array or joined arrays in differnet buildings. Otherwise, you can match it for performance and far beat it for cost using old fashioned SCSI. Remember, the new Ultra3 runs 160MB/s vs only 100MB/s for most Fibre Channel implementations.
I fear that if we don't see some major good developments in the Fibre Channel arena soon, it could end up going the way of ATM.
I thought the fish tail swimmer was quite novel. Give it time. The biggest constraint right now is computing power to test enough permutations. The site already talks about adding additional base components in. Once enough base components are combined into a composite object, you may have something new and never before seen.
It will be interesting to see if their software evolves to use federations of computers (like distributed.net) or clusters. This would allow testing of a much wider variety of base components and much larger composite constructs.
Make sure you download the simulation software at http://www.demo.cs.brandeis. edu/pr/golem/download.html. It only runs under Windows, but it's curious to watch the process running. I've got it running on a couple of computers overnight. It will be interesting to see what is crawling around at the office tomorrow morning.
Hmm, "GOSIP", there's something I haven't heard about in a few years (thank goodness).
Remember, this is the same group who also wanted ALL programming done in ADA for about five years. If it's good enough to fly a missle, it's good enough to handle an accounting system.
This is also the same group that mandated POSIX (UNIX and X-Windows) for all desktop systems. In hindsight, this might not have been too bad. At the time, however, the industry was moving in a decidely different direction.
The US Govt, and DoD in particular, has a long history of heading down the wrong path with IT solutions.
It's visible over just as much of the southern hemisphere as the northern one. It crosses the equator at (I believe) 54 degrees. That sends it up as far north as 54 lattitude and as far south as 54 lattitude. This is something like 80% of the Earth's land surface.
Oh, and there are souther hemisphere countries invovled as well. Brazil signed on several years ago.
Well, it sad to see Iridium go. I've followed is progress over the last several years and thought the concept really had a chance. Unfortunately, market tides and marketing foulups shifted under Iridum's feet, and they fell on their face. Let me tackle a few of the questions here.
The time hasn't come yet.
The time came five years ago even more than now. Iridium could have been used to bootstrap phone networks up in developing countries until regular cellular towers were available. One of the concepts was also a form of "village phone" that was basicly a phone booth with a sat antenna on top.
Expensive bulky phones that didn't work indoors?
The phones did have problems indoors. They really needed a line of site in order to connect up to the sats. It would have worked fine for a roof mounted antenna on a truck, ship, or plane. For some reason, the marketing brain power at Iridium decided to target mobile executives rather than commercial industry. Instead of trying to get a Fortune 500 CEO to carry one in a briefcase, they could have targeted trucking companies who do cross country runs, shipping that is in the middle of the ocean, and airlines who could use a cost effective replacement for those "Airphones" they try and charge $3/minute for. Iridium failed to target the tech to the market is was sufficient for.
Why not just auction the suckers?
Won't work. First, there is a lot of ground support involved. I believe the cost is at somewhere around $1M/day to operate the sats. Next, you have to send up replacements too often. This is not a geosync sat that just hangs out. This is five dozen plus sats in low orbit experiencing constant drag. Within a few years, the first generation sats will start coming home on their own. With a controlled deorbit, you can at least make sure they all end up in the ocean instead of having chunks of metal land in New York and Tokyo.
Iridium completely missed the boat on data service. The system is designed around voice and low-bandwidth pager data. This was a major design flaw with the move to an information society over the last few years. If Teledesic gets off the ground, maybe my faith in these sat clusters will be renewed, but it will take a lot.
The failure of these first generation sat clusters has hurt more than just the sat companies themselves. Several companies were developing new low cost launching technologies intended to support this market. You can write off Rotary Rocket and serverl other companies because they saw their potential customer go away before they were even out the door.
L1:The 1GHz Xeon chip offers 256KB of Level 2 cache and a 133MHz bus, he said. L2:Wow so it's got twice the cache, a little over twice the clockspeed and a slighly higher bus speed than my OC'd celeron300 that I bought for pennies over a year ago.
The Xeon is a different animal. The cache is bound closer to the procesoor core to move data in and out even faster. However, most of my work only uses the 2MB cache version of the processor. That is where the performance really goes up, and the cost goes up even more.
One of my greatest personal accomplishments a couple of years ago was getting Linux to boot on a quad Xeon 400. Now there was a DES cracking monster.
Slashdot Mirror
I'm not going to hold my breath for it, but it would be nice if this version came with IPv6 support enabled by default. This would include the kernel, net tools, applets such as telnet, ftp, & ssh, and Apache.
The tools for a working IPv6 host are out there. It's time for the distributions to start including the support. Once the core system is up and running with IPv6, the rest of the app developers will have some incentive to get their stuff cleaned up to start running under IPv6.
Think of it this way, if there was only one major desktop environment out there (be it KDE, GNOME, or something new), it likely wouldn't be as advanced as either of these two environments. Competition pushes progress.
My biggest fear in all this is that the developers will compromise stability for new features. Then, we start to get into the old Microsoft problems.
From the e-mail: "Of course, we are planning to have a kpart for Mozilla too so konqueror and others can use it too if so desired. Again, we are just providing an alternative while the core KDE team work out all the bugs and standards conformance features with their khtml implementation."
The idea isn't to replace what Konqueror is doing but provide alternatives. This is certainly a good thing. If a Konqueror user can select which rendering engine to use, it will make it much more apparent when one engine has a problem. The programmers will have even more reason to fix any bugs since they will stick out like a wart on your nose.
I doubt the KDE team will have to rewrite a lot. The whole idea of kparts is to abstract the interface to the point where one side doesn't care about the internals of the other. What me may see are a few more methods added to the existing html widgets.
They way they can make money is to sell products. The way they can sell products is to draw attention to themselves. The way they can draw attention to temselves is to create talk in the press over projects they are supporting. This is a good thing!
The quickest way for Corel to completely fade away is to stop doing new development. It's kind of like evolution, either you advance or get overrun. A company that terminates all development efforts is usually on it's last leg.
When you consider how much money Corel is burning through, how much is it really costing them to have a couple of programmers dedicated to a project like this? $200-250K/year for a company operating on the level of $20-30M/year is not that big of deal.
Unfortunately, the OSS movement has a lot of people who are a little too deep in the socialist utopia ideal. Who cares that RedHat gives away their software for anyone to download. They are evil becuase they *gasp* want to make money.
The updates will still be available on updates.redhat.com just like they always have. RedHat is targeting the subscription services at corporations. They aren't holding back any software, just tailoring their premium services.
RedHat stated a while back that 7.0 would ship with a 2.2 kernel but that everything would be ready to go for the 2.4 kernel when it became availalble. Expect to see the kernel change as just another update package in a couple of months.
Related to this, KDE 2 is supposed to go final within the next month. That's another major applicaiton set that it would have made sense to wait for. I expect to see this updated after release as well.
Odds are, people can come up with a dozen other major packages that are "close" to having a new stable version ready. If RedHat waited for all of them, they would never get a release out.
While bringing in new features is nice, consider this: How many times has your Palm crashed? One big benefit of the "keep it simple" mentality is a VERY stable platform. That's not to even mention the speed benefits.
Personally, I don't expect to see any major advances until several hardware technologies advance such as:
There is still a lot to go into this platform!
Rather than a "cell phone module", I'd rather have BlueTooth in both my PDA and cell phone and let them talk to each other. Let the cell phone browse the PDA's adress book. Let the PDA use the cell phone for communications. Throw in a wireless, ear-clip headset and I'd be in hog heaven doing voice queries on the PDA as well as phone conversations.
I expect the use for this cable would be somewhat limited being that it takes a mono audio patch (RCA male and female connectors) and feeds it to a 1/8" stereo input. If it was stereo clear across, I could see plenty of uses for it. Having a 20' cable available is usually a good thing.
I can honestly say that I'm not impressed with the Cue:Cat. Mine showed up a week ago, and so far it's only been able to read about one in ten barcodes. Maybe mine is just defective, but if this is common then DC has a serious problem on their hands. Either the product doesn't work as well as advertised, or they have quality control issues with the ones they are sending out.
Oh, their "Convergence Cable" is nothing to write home about either (except to tell your family to avoid it). Now, I can't get my TV audio to stop going through the speakers on the computer every time I reboot. Also, the only time a show used an audio que (NBC Olympics), the web browser errored trying to find the supposedly passed link.
I'm a big fan of these mixed media tools (computer/TV/print integration), but if this is an example of what we can expect, you had better watch out.
"minimal" means that you still have to exercise some due diligence to avoid having your software exported to a few "bad" countries such as Iraq and Libya. I'm not sure that this restriction even applies to Open Source software. When you think about it, the idea is pretty futile. All you need is some sort of FTP proxy server outside of the US and anywhere in the world can then get to your file archive.
I'm one of those sick people who actually watches NASA TV during shuttle missions. It's kind of a out of this world reality TV show. Anyway, the laptops onboard do lockup from time to time. They just reboot and try again. As was said earlier, nothing mission critical is run on them.
I'm pretty sure the Shuttle uses IBM ThinkPads. I'm very sure the ISS does. This probably means that the track point on the keyboard is the preferred method of control. Of course, the astronaut would have to be strapped down or they would drift away every time they touch the keyboard.
There have been articles published in the past about Linux being used to operate experiments on the Space Shuttle. Some of the experiments have self contained computer systems for managment and data collection. Linux was ideal here.
Now, I'd like to see a Linux/Apache web server in orbit. Even if it feeds nothing but text telemetry data, it would be impressive. Hmm, is there a TLD for LEO yet?
I should have qualified my statement about ATM to say "ATM in the LAN". Most people I've talked to trying to use ATM in a building have been very unhappy and wanted nothing more than to rip it out and go to a combination of 1Gb and 100Mb Ethernet.
If you want to see how expensive FC equipment can be. Go to Dell's website and look at some of their "PowerVault" 630F and 650F storage equipment. This is all FC based, and can give you a good idea of how expensive a good array costs.
You just had to throw the glowing bunny at us, didn't you? :)
One thing you learn working around the gov't is that if you collect enough nonclassified information together you may eventually end up with a product that ends up being classified. It is quite possible that the people being interviewed did not individually expose any company secrets, but when everything was corelated together a bigger picture was revealed.
Cost. Fibre Channel equipment costs considerably more than SCSI. The idea of NAS (Network Attached Storage) is to reduce cost. This box probably has four 80GB IDE disk drives inside of it. To build something simliar using Fibre Channel would likely cost $10,000 or more.
Fibre Channel is good if you need some of the features like several servers sharing one array or joined arrays in differnet buildings. Otherwise, you can match it for performance and far beat it for cost using old fashioned SCSI. Remember, the new Ultra3 runs 160MB/s vs only 100MB/s for most Fibre Channel implementations.
I fear that if we don't see some major good developments in the Fibre Channel arena soon, it could end up going the way of ATM.
It will be interesting to see if their software evolves to use federations of computers (like distributed.net) or clusters. This would allow testing of a much wider variety of base components and much larger composite constructs.
Make sure you download the simulation software at http://www.demo.cs.brandeis. edu/pr/golem/download.html. It only runs under Windows, but it's curious to watch the process running. I've got it running on a couple of computers overnight. It will be interesting to see what is crawling around at the office tomorrow morning.
Remember, this is the same group who also wanted ALL programming done in ADA for about five years. If it's good enough to fly a missle, it's good enough to handle an accounting system.
This is also the same group that mandated POSIX (UNIX and X-Windows) for all desktop systems. In hindsight, this might not have been too bad. At the time, however, the industry was moving in a decidely different direction.
The US Govt, and DoD in particular, has a long history of heading down the wrong path with IT solutions.
Oh, and there are souther hemisphere countries invovled as well. Brazil signed on several years ago.
Well, it sad to see Iridium go. I've followed is progress over the last several years and thought the concept really had a chance. Unfortunately, market tides and marketing foulups shifted under Iridum's feet, and they fell on their face. Let me tackle a few of the questions here.
The time hasn't come yet.
The time came five years ago even more than now. Iridium could have been used to bootstrap phone networks up in developing countries until regular cellular towers were available. One of the concepts was also a form of "village phone" that was basicly a phone booth with a sat antenna on top.
Expensive bulky phones that didn't work indoors?
The phones did have problems indoors. They really needed a line of site in order to connect up to the sats. It would have worked fine for a roof mounted antenna on a truck, ship, or plane. For some reason, the marketing brain power at Iridium decided to target mobile executives rather than commercial industry. Instead of trying to get a Fortune 500 CEO to carry one in a briefcase, they could have targeted trucking companies who do cross country runs, shipping that is in the middle of the ocean, and airlines who could use a cost effective replacement for those "Airphones" they try and charge $3/minute for. Iridium failed to target the tech to the market is was sufficient for.
Why not just auction the suckers?
Won't work. First, there is a lot of ground support involved. I believe the cost is at somewhere around $1M/day to operate the sats. Next, you have to send up replacements too often. This is not a geosync sat that just hangs out. This is five dozen plus sats in low orbit experiencing constant drag. Within a few years, the first generation sats will start coming home on their own. With a controlled deorbit, you can at least make sure they all end up in the ocean instead of having chunks of metal land in New York and Tokyo.
Iridium completely missed the boat on data service. The system is designed around voice and low-bandwidth pager data. This was a major design flaw with the move to an information society over the last few years. If Teledesic gets off the ground, maybe my faith in these sat clusters will be renewed, but it will take a lot.
The failure of these first generation sat clusters has hurt more than just the sat companies themselves. Several companies were developing new low cost launching technologies intended to support this market. You can write off Rotary Rocket and serverl other companies because they saw their potential customer go away before they were even out the door.
Such is life...
L2: Wow so it's got twice the cache, a little over twice the clockspeed and a slighly higher bus speed than my OC'd celeron300 that I bought for pennies over a year ago.
The Xeon is a different animal. The cache is bound closer to the procesoor core to move data in and out even faster. However, most of my work only uses the 2MB cache version of the processor. That is where the performance really goes up, and the cost goes up even more.
One of my greatest personal accomplishments a couple of years ago was getting Linux to boot on a quad Xeon 400. Now there was a DES cracking monster.