2) Want a reliable server? -Debian Stable 3) Want the latest and greatest? -Debian Unstable
I'll give you #2, but #3 is bogus. Without keeping track of the latest, popular apt.sources for X11, XMMS, mplayer, Gnome, KDE, and every other remotely large application, even unstable falls behind.
And it's just that 'unstable'. I can't count the number of times I've seen gphoto broken, just to pick one example. And the attitude of the developers? "If you want it to work, run testing or stable, it's called 'unstable' for a reason."
If you are utilizing chassie level switches, you'll run into them. The point is to have different IP subnets on the same 'switch', and have the switch route appropriately (or not as needed by the network design... without the routing piece, a host on VLAN 1, can't see a host on VLAN 2. That might be how you want it.)
If you have a chassie with 8 blades of 48 ports each, it's unlikely that you want all of them to be on the same subnet, so you have VLAN's.
We got to move buildings. Great experience if you can get the company to foot the bill.
We ended up using a distribution layer because it made the config easier, and centralized a lot of complexity, and didn't cost an arm and a leg in fiber uplink ports. We utilized Extreme Networks 7i switches as distribution (44 gig fiber links), which (compared to core chassie blades for the same fiber density) were cheep. We then uplinked each distribution layer switch (2 of them) to the 4 core chassies. This cost 8 core fiber links, as opposed to 40 (2 per floor, 20 floors), which adds up in a real hurry.
It comes down to the greenbacks. With Extreme gear, that 7i is a beautiful distribution layer for not a huge outlay in cost. It also simplifies configuration, since the floor switches are 'dumb' (configure 1 vlan over all ports, configure management IP address, done.). YMMV, especially with other venders (cisco).
You are sort of right. A 'router' is capable of working with multiple subnets, but traditionally, only has a few interfaces. A 'switch' (or hub) is traditionally only able to deal with 1 subnet, but has lots of 'interfaces' (ports).
Switches have grown up, since the advent of VLAN's, they've been able to 'route' between vlans, and have expanded to OSPF, and other high end routing protocols, while keeping the port count. These higher end switches don't usually have WAN ports (T1, T3 type), or the ability to do super high end routing (OC-16, OC-192, Terabit), which is why Cisco and Juniper still sell routers. The two terms have become quite unclear over the past decade.
BSD style is alright for a single admin, on a single box.
SYSV is worlds better when you're: dealing with many boxes or for installing init scripts from packages or for disabling 1 thing for the next boot either by hand, or by script and ssh or restarting one service without remembering it's start flags (sendmail, now is it '-q1h -d'? or '-q -r1h -d' or what? Named... now it's -t user, or is it -t user -c/var/named... oh wait, on this box, the chroot is in/home... or...).
I could keep naming things where SYSV init has made my life easier, but that's enough for now.
and how many times...
on
GCC 4.0 Preview
·
· Score: 5, Insightful
And how many times will they break ABI, API and library compatability in THIS major release? Count stands at 4 for the 3 series, maybe higher.
The biggest challenge with Binary compatability across Linux distros is the GCC release (followed by the glibc releases, who live in the same ivory tower). I realize that things have to change, but I wish that they would not break compat between versions quite so often...
I'd really like to be able to take a binary between versions, and it just work.
This is one area where Sun rocks. Any binary from any solaris2 build will just work on any later version. With some libraries, you can go back to the SunOS days (4.1.4, 4.1.3UL, etc). That's 15 years or so.
It may have already passed, but the second this comes up before a federal court, it will be struck down. The constitution forbids the states from interfering in inter-state comerce.
Since 9 times out of 10 you won't be selling to someone inside the state...
I've played Illuminati, and it's great fun. Also by Steve Jackson is Munchkin and it's off shoots. Very funny, and if you start combining them, pretty crazy.
Munchkin is also a game of Social Engineering. You never want to get onto my sisters 'mortal enemy' list... it's brutal...
Re:One of my favorites is "Scotland Yard"
on
Fun Tabletop Games?
·
· Score: 1
I'm really supprised to see Scotland Yard recommended in a computer geek forum... If the inspectors are smart and rigorus, then there is no way that Mr. X escapes.
All they have to do is spread out until the first showing of Mr. X, and then start converging. They'll have him surrounded by about move 26 or 27.
Until this technique is put into the field, we won't know how good this 'one number' is. You could encode the gene sequence of a human into one (rather large) number, and it'd be pretty good as an indentifier. If there's enough entropy in the clock skews, then it could uniquely identify 1 computer out of a billion or so. But that's an 'if'.
My question is if this clock skew can me consistantly measured across multiple OS installed on the same laptop (dual boot anyone?).
The best advice is to go to a store, and try them. The one you're more comfortable with is the safer knife. This is pretty much universal in knives, from multi-tools, to wood carving, to kitchen. If you're comfortable with it, you'll use it more, and more safely.
That said, I've had more experience with Leatherman's products. The 'SuperTool' is nice, but the 'Wave' is greatly improved: rounded handles, blades available on the outside (one handed). Both of these offer locking blades.
Gerber is very nice too. If you need one handed access to plyers more than anything else, then they're the clear winner (though I can one hand the leatherman butterfly knife style).
For years, you've had a duality between gerber and leatherman, but recently some newcomers have entered. There's a new Titanium leatherman that's expensive, looks cool, and I have not had the oppertunity to handle. Vitronox (sp) (makers of swiss army knives) has a multitool out that has some good reviews as well.
Like it or not, these machines will be rooted or get seriously fouled up at some point. This is actually one area where Linux really shines. You can set up a net boot environment (or live cd) that brings the box to a known good state. Don't keep any real data on these boxes. You don't even HAVE to keep a desktop image. You can NFS mount / if you really wanted too (though it's probably better to have an OS image local that can be over written easily).
This means you'll probably need a more beefey (at least in hard drive space) server that this lab will live off of, but I assume you already knew that.
Out of curiosity, what did you have on the other end of those firewire ports? Anything that could really push 800 Megabits? [1]
3 firewire2 ports is only 300 megabytes/s, which is less than the scsi3 360 cards, that fit happily within the bandwidth of a 64 bit, 66mhz PCI slot. It may be a PCI-X card, but I still contend that it doesn't NEED to be. I can't speak to the High Speed Color scanner. That'd probably be a pretty cool application. Though would you really need 16 Gb of ram?
[1] Now, I admit, I had to dig a little bit to find out if firewire2 was in megabits or megabytes. According to this article it's megabits)
Whether to use a cluster or not depends heavily on the problem domain. Until very recently, clusters didn't work so well with large databases, etc.
You also seem to be shopping specs rather than throughput. Your mention of 16x PCIexpress is what gives this away. The only cards that support this now are high, high end graphics cards, and these cards don't even need it. There's no real difference between the AGP 8x and PCIex versions of these cards.
That said, you're not going to find what you're looking for in the beige box world. You're looking (realistically) at about 4 different venders: Windows: Dell, IBM, and HP. UNIX: IBM, HP, and Sun.
You're also only looking at servers (not desktop or towers).
My experience is with Sun, and a little Dell and IBM. So I'm going to speak to those. Sun makes magnificient hardware. Their support organization has had problems recently, but the hardware is good enough that we don't need it often. Sun's V880 servers are amazing. up to 8 CPU's and up to 32 gigs of ram, with great growth potential (12 PCI slots, several of them 64 bit, 66 MHz).
We've had lots of problems with our Dell hradware. Whole lines of their servers have been crap, and dell replaced thier 16xx line with their 17xx line for us for free. Our exchange server runs on a 6550 IIRC, which has at least 8 gigs of ram. This model probably can go higher in ram, but I'm not sure.
We've been really impressed with the IBM hardware we've started to purchase. It's been pretty stable, fun to work with, etc. IBM has a long history of making great servers. They probably have several models that will help.
1) The success of Netflix. People are willing to hold any 3 movites for as long as they want. Delete the old one, and get a new one. Charge a subscription fee, and done.
2) My main worry with this very nice dream: Bandwidth caps. The noise around these has died down, but if you have a couple of people downloading multiple gigs of data, someone is going to run into a bandwidth cap somewhere, and scream really loudly.
The right to property isn't in the Constitution, but it is in the Declaration of Independance. In the Bill of Rights, the fifth ammendment has this:
"nor shall private property be taken for public use, without just compensation."
Basically, the Founding Fathers knew that people would claim land that the Government would find too useful to pass up. So they put this piece into the Bill of Rights. This is called Eminent Domain. The government decides that it needs a piece of land, determines a fair value for it, and gives you the money, and you have to leave.
Now, this is is probematic on occasion because 'Just Compensation" isn't defined in the constitution, and it is up to the government to decide what is 'just'. You (sometimes) can sue for more money, but it's a real challenge in the courts.
Eminent Domain is something that governments need. The problem is balance.
BanthaPoodoo wrote: Just curious: whats the feasibility of sending antiwaves into oncoming tsunamis to reduce their effect?
None. Think about it. The Tsunamis are circular, originating from a central point. There's no way to set up your 'antiwave' to blunt it.
Even if that wern't a problem, the other problem is raw force. These things have the power that's measured in megatons. The only things we have that can produce that force are nukes. How do you think people would feel if there was a nuke parked beside any volcano, fault line, etc? And also one that had to be detonated instantly, without human oversight and control?
If you're company is large enough to have a dedicated helpdesk, that helps a lot. So empower them. Give them limited rights to do the silly stuff that wastes your time, but is required for the smooth running of any IT shop, like changing passwords, releasing emails from spam quarantine, etc.
While I applaud the notion behind Freer distribution (as in speech) it's also highly probable that Intel doesn't have much ground make them freer
Intel can't make them more free, not because of licences from other people, but because of the law. These devices are general purpose RF transmitters. Arbitrary frequencies, listen to anything, etc.
The FCC has mandated that only certain spectrums can be used by non-licensed individuals. Some others can be used by licenseed individuals. Some can only be used by the police, etc.
Intel knows what would happen if the firmware was given to OSS community. Someone, somewhere would think a general purpose RF transmiter/reciever would be cool, and make it. And it's 100% illegal.
You can argue that it SHOULDN'T be illegal, but Intel is following the law, and generally shouldn't be faulted for that.
One thing that I always have a problem with, when teaching my mom anything about the computer is to only show her one thing (or change) at a time. In your example, the FIRST thing to show them is that they can copy and paste more than one cell at a time. Tell them how to highlight all the cells, and let them copy with the menu option. Then have them paste them into the same sheet to prove to them it works. Give them several practice runs on this procedure, only letting them highlight multiple cells one way (either by click/drag, or control or shift which ever they're used to... don't show them the others)
After this, show them that they can do multiple cells between sheets, still using menu copy/paste, and still highlighting the same way, and flipping sheets the same way. Let them practice this one change a few times.
Then show them ctrl-c and ctrl-v, and let them practice several times.
Then show them other ways to highlight, and let them practice.
My problem is that I must resist taking the mouse from her. I must let her practice each thing several times so that SHE understands, and so that I don't have to do it for her every time.
We all have learning curves, and if we have good teachers, they are easier. A good example is worth infinatly more than an RTFM.
Well, Take a look at the 'Steves Digicam' review linked above. At the end of the review are some full rez pictures from the camera... I have to say that I'm not that impressed, however I am a self admitted camera snob.
If you're taking these pictures at a party, and just going to post them to the web or something, you're fine. The images will do great. However, if you're planning on printing out these images, you're probably in for some disappointment.
Basically, it comes down to the usual mantra: Right tool for the right job. I have to admit that the form factor on this thing is great... But I'm still going to buy a Nikon D70 next year.
Though I might sell my canon G3 and purchase this after that, just to have a true, pocket camera... hrm.
Slashdot Mirror
2) Want a reliable server? -Debian Stable 3) Want the latest and greatest? -Debian Unstable
I'll give you #2, but #3 is bogus. Without keeping track of the latest, popular apt.sources for X11, XMMS, mplayer, Gnome, KDE, and every other remotely large application, even unstable falls behind.
And it's just that 'unstable'. I can't count the number of times I've seen gphoto broken, just to pick one example. And the attitude of the developers? "If you want it to work, run testing or stable, it's called 'unstable' for a reason."
Debian is becoming the slackware of this decade.
> What is the point of a VLAN?
If you are utilizing chassie level switches, you'll run into them. The point is to have different IP subnets on the same 'switch', and have the switch route appropriately (or not as needed by the network design... without the routing piece, a host on VLAN 1, can't see a host on VLAN 2. That might be how you want it.)
If you have a chassie with 8 blades of 48 ports each, it's unlikely that you want all of them to be on the same subnet, so you have VLAN's.
We got to move buildings. Great experience if you can get the company to foot the bill.
We ended up using a distribution layer because it made the config easier, and centralized a lot of complexity, and didn't cost an arm and a leg in fiber uplink ports. We utilized Extreme Networks 7i switches as distribution (44 gig fiber links), which (compared to core chassie blades for the same fiber density) were cheep. We then uplinked each distribution layer switch (2 of them) to the 4 core chassies. This cost 8 core fiber links, as opposed to 40 (2 per floor, 20 floors), which adds up in a real hurry.
It comes down to the greenbacks. With Extreme gear, that 7i is a beautiful distribution layer for not a huge outlay in cost. It also simplifies configuration, since the floor switches are 'dumb' (configure 1 vlan over all ports, configure management IP address, done.). YMMV, especially with other venders (cisco).
You are sort of right. A 'router' is capable of working with multiple subnets, but traditionally, only has a few interfaces. A 'switch' (or hub) is traditionally only able to deal with 1 subnet, but has lots of 'interfaces' (ports).
Switches have grown up, since the advent of VLAN's, they've been able to 'route' between vlans, and have expanded to OSPF, and other high end routing protocols, while keeping the port count. These higher end switches don't usually have WAN ports (T1, T3 type), or the ability to do super high end routing (OC-16, OC-192, Terabit), which is why Cisco and Juniper still sell routers. The two terms have become quite unclear over the past decade.
The sad thing is that someone will probably use your post as a design document...
BSD style is alright for a single admin, on a single box.
/var/named... oh wait, on this box, the chroot is in /home... or...).
SYSV is worlds better when you're:
dealing with many boxes
or for installing init scripts from packages
or for disabling 1 thing for the next boot either by hand, or by script and ssh
or restarting one service without remembering it's start flags (sendmail, now is it '-q1h -d'? or '-q -r1h -d' or what? Named... now it's -t user, or is it -t user -c
I could keep naming things where SYSV init has made my life easier, but that's enough for now.
And how many times will they break ABI, API and library compatability in THIS major release? Count stands at 4 for the 3 series, maybe higher.
The biggest challenge with Binary compatability across Linux distros is the GCC release (followed by the glibc releases, who live in the same ivory tower). I realize that things have to change, but I wish that they would not break compat between versions quite so often...
I'd really like to be able to take a binary between versions, and it just work.
This is one area where Sun rocks. Any binary from any solaris2 build will just work on any later version. With some libraries, you can go back to the SunOS days (4.1.4, 4.1.3UL, etc). That's 15 years or so.
There are only two downsides:
1) It'd take 10 years for him to finish it.
2) It'd take 10 years away from the AoCP work...
It may have already passed, but the second this comes up before a federal court, it will be struck down. The constitution forbids the states from interfering in inter-state comerce.
Since 9 times out of 10 you won't be selling to someone inside the state...
I've played Illuminati, and it's great fun. Also by Steve Jackson is Munchkin and it's off shoots. Very funny, and if you start combining them, pretty crazy.
Munchkin is also a game of Social Engineering. You never want to get onto my sisters 'mortal enemy' list... it's brutal...
I'm really supprised to see Scotland Yard recommended in a computer geek forum... If the inspectors are smart and rigorus, then there is no way that Mr. X escapes.
All they have to do is spread out until the first showing of Mr. X, and then start converging. They'll have him surrounded by about move 26 or 27.
Until this technique is put into the field, we won't know how good this 'one number' is. You could encode the gene sequence of a human into one (rather large) number, and it'd be pretty good as an indentifier. If there's enough entropy in the clock skews, then it could uniquely identify 1 computer out of a billion or so. But that's an 'if'.
My question is if this clock skew can me consistantly measured across multiple OS installed on the same laptop (dual boot anyone?).
The best advice is to go to a store, and try them. The one you're more comfortable with is the safer knife. This is pretty much universal in knives, from multi-tools, to wood carving, to kitchen. If you're comfortable with it, you'll use it more, and more safely.
That said, I've had more experience with Leatherman's products. The 'SuperTool' is nice, but the 'Wave' is greatly improved: rounded handles, blades available on the outside (one handed). Both of these offer locking blades.
Gerber is very nice too. If you need one handed access to plyers more than anything else, then they're the clear winner (though I can one hand the leatherman butterfly knife style).
For years, you've had a duality between gerber and leatherman, but recently some newcomers have entered. There's a new Titanium leatherman that's expensive, looks cool, and I have not had the oppertunity to handle. Vitronox (sp) (makers of swiss army knives) has a multitool out that has some good reviews as well.
Like it or not, these machines will be rooted or get seriously fouled up at some point. This is actually one area where Linux really shines. You can set up a net boot environment (or live cd) that brings the box to a known good state. Don't keep any real data on these boxes. You don't even HAVE to keep a desktop image. You can NFS mount / if you really wanted too (though it's probably better to have an OS image local that can be over written easily).
This means you'll probably need a more beefey (at least in hard drive space) server that this lab will live off of, but I assume you already knew that.
Out of curiosity, what did you have on the other end of those firewire ports? Anything that could really push 800 Megabits? [1]
3 firewire2 ports is only 300 megabytes/s, which is less than the scsi3 360 cards, that fit happily within the bandwidth of a 64 bit, 66mhz PCI slot. It may be a PCI-X card, but I still contend that it doesn't NEED to be. I can't speak to the High Speed Color scanner. That'd probably be a pretty cool application. Though would you really need 16 Gb of ram?
[1] Now, I admit, I had to dig a little bit to find out if firewire2 was in megabits or megabytes. According to this article it's megabits)
Whether to use a cluster or not depends heavily on the problem domain. Until very recently, clusters didn't work so well with large databases, etc.
You also seem to be shopping specs rather than throughput. Your mention of 16x PCIexpress is what gives this away. The only cards that support this now are high, high end graphics cards, and these cards don't even need it. There's no real difference between the AGP 8x and PCIex versions of these cards.
That said, you're not going to find what you're looking for in the beige box world. You're looking (realistically) at about 4 different venders: Windows: Dell, IBM, and HP. UNIX: IBM, HP, and Sun.
You're also only looking at servers (not desktop or towers).
My experience is with Sun, and a little Dell and IBM. So I'm going to speak to those. Sun makes magnificient hardware. Their support organization has had problems recently, but the hardware is good enough that we don't need it often. Sun's V880 servers are amazing. up to 8 CPU's and up to 32 gigs of ram, with great growth potential (12 PCI slots, several of them 64 bit, 66 MHz).
We've had lots of problems with our Dell hradware. Whole lines of their servers have been crap, and dell replaced thier 16xx line with their 17xx line for us for free. Our exchange server runs on a 6550 IIRC, which has at least 8 gigs of ram. This model probably can go higher in ram, but I'm not sure.
We've been really impressed with the IBM hardware we've started to purchase. It's been pretty stable, fun to work with, etc. IBM has a long history of making great servers. They probably have several models that will help.
There are two things that this brings up.
1) The success of Netflix. People are willing to hold any 3 movites for as long as they want. Delete the old one, and get a new one. Charge a subscription fee, and done.
2) My main worry with this very nice dream: Bandwidth caps. The noise around these has died down, but if you have a couple of people downloading multiple gigs of data, someone is going to run into a bandwidth cap somewhere, and scream really loudly.
The right to property isn't in the Constitution, but it is in the Declaration of Independance. In the Bill of Rights, the fifth ammendment has this:
"nor shall private property be taken for public use, without just compensation."
Basically, the Founding Fathers knew that people would claim land that the Government would find too useful to pass up. So they put this piece into the Bill of Rights. This is called Eminent Domain. The government decides that it needs a piece of land, determines a fair value for it, and gives you the money, and you have to leave.
Now, this is is probematic on occasion because 'Just Compensation" isn't defined in the constitution, and it is up to the government to decide what is 'just'. You (sometimes) can sue for more money, but it's a real challenge in the courts.
Eminent Domain is something that governments need. The problem is balance.
BanthaPoodoo wrote:
Just curious: whats the feasibility of sending antiwaves into oncoming tsunamis to reduce their effect?
None. Think about it. The Tsunamis are circular, originating from a central point. There's no way to set up your 'antiwave' to blunt it.
Even if that wern't a problem, the other problem is raw force. These things have the power that's measured in megatons. The only things we have that can produce that force are nukes. How do you think people would feel if there was a nuke parked beside any volcano, fault line, etc? And also one that had to be detonated instantly, without human oversight and control?
If you're company is large enough to have a dedicated helpdesk, that helps a lot. So empower them. Give them limited rights to do the silly stuff that wastes your time, but is required for the smooth running of any IT shop, like changing passwords, releasing emails from spam quarantine, etc.
--Jason
Out of curriosity, how does one verify that a subpoena is served properly? I assume that you read such very carefully, and call it a day.
While I applaud the notion behind Freer distribution (as in speech) it's also highly probable that Intel doesn't have much ground make them freer
Intel can't make them more free, not because of licences from other people, but because of the law. These devices are general purpose RF transmitters. Arbitrary frequencies, listen to anything, etc.
The FCC has mandated that only certain spectrums can be used by non-licensed individuals. Some others can be used by licenseed individuals. Some can only be used by the police, etc.
Intel knows what would happen if the firmware was given to OSS community. Someone, somewhere would think a general purpose RF transmiter/reciever would be cool, and make it. And it's 100% illegal.
You can argue that it SHOULDN'T be illegal, but Intel is following the law, and generally shouldn't be faulted for that.
One thing that I always have a problem with, when teaching my mom anything about the computer is to only show her one thing (or change) at a time. In your example, the FIRST thing to show them is that they can copy and paste more than one cell at a time. Tell them how to highlight all the cells, and let them copy with the menu option. Then have them paste them into the same sheet to prove to them it works. Give them several practice runs on this procedure, only letting them highlight multiple cells one way (either by click/drag, or control or shift which ever they're used to... don't show them the others)
After this, show them that they can do multiple cells between sheets, still using menu copy/paste, and still highlighting the same way, and flipping sheets the same way. Let them practice this one change a few times.
Then show them ctrl-c and ctrl-v, and let them practice several times.
Then show them other ways to highlight, and let them practice.
My problem is that I must resist taking the mouse from her. I must let her practice each thing several times so that SHE understands, and so that I don't have to do it for her every time.
We all have learning curves, and if we have good teachers, they are easier. A good example is worth infinatly more than an RTFM.
They were /.'ed yesterday just from being posted to dpreview.com... Heaven forbid what the real /. will do...
Though when I invest in a DSLR early next year, I think I'll take a long look at their offerings...
Well, Take a look at the 'Steves Digicam' review linked above. At the end of the review are some full rez pictures from the camera... I have to say that I'm not that impressed, however I am a self admitted camera snob.
If you're taking these pictures at a party, and just going to post them to the web or something, you're fine. The images will do great. However, if you're planning on printing out these images, you're probably in for some disappointment.
Basically, it comes down to the usual mantra: Right tool for the right job. I have to admit that the form factor on this thing is great... But I'm still going to buy a Nikon D70 next year.
Though I might sell my canon G3 and purchase this after that, just to have a true, pocket camera... hrm.