Because two nodes on an IPV6 network wouldn't be stuck behind closed NAT firewalls and not able to communicate directly. Currently, if you're using Yahoo IM and want to send someone a picture, you have to relay it through Yahoo's servers, which causes a bottleneck because thousands of other people are doing the same and everything bottlenecks through Yahoo. With IPV6, both ends could have a public (possibly static) IP address, so person A could connect directly to person B and bypass the traffic jam at the server. It gets worse on something like eDonkey where if both ends are NAT'ed the transfer CANNOT happen.
As for removing the need for Skype, it wouldn't, but it would change. A Skype server would just need to know IP addresses so if person A wants to talk to person B, the Skype server just tells A "B is at 0:1:2:3:4:5" and A can then connect to B directly, instead of relaying through god-only-knows how many intermediates who may or may not be compromised or malicious. However, if you were setting up your own PBX with a static IP (And IPv6 has enough addresses that you could), then you wouldn't need a central authority (Skype server) to do the matchmaking.
Yes you'll still be firewalled, but the firewall will be at the end point, not in the middle somewhere that you have no control over (As in my case). The firewall rules would just change to a FORWARD rule instead of a "direct this port to this IP" rule.
IPV6 is about decentralizing, while NAT is essentially centralized (At the NAT server). If something can't change ports (Some VPNs, old protocols, not smart enough, etc), then you can't have more than one person behind the NAT using said protocol. NAT also causes problems because to the machine on the public-side of the NAT, only sees the address of the NAT gateway, not the address of the machine actually making the request. The implications of that are left up to the reader.
OWA wasn't like the other web-based e-mail interfaces at the time. It was the first to use the XMLHTTP object, the basis for all current AJAX and JSON RPC platforms. They added it to Outlook Express so it could communicate with Hotmail without refreshing the page all the time (Exactly what AJAX does now, but using Outlook Express for the interface instead of Javascript+DHTML). Then they didn't branch it out until other developers started using it for things like Google Maps, GMail, etc. Now their light bulbs are coming on and they're starting to think "Maybe we can use this for more than just Hotmail."
Red Hat (beginning in Fedora Core 4) has Xen RPMs for their kernels as well. Each time they release a new kernel, there are Xen versions as well. They want to get it into Linus' kernel tree so everyone can use and test it, since other distros are leery of using Red Hat stuff but fine with anything Linus has integrated into his kernel.
I must plead ignorance regarding MS Office file sharing features. Every office I've worked in has used Windows (or at least SMB) file servers to share files. I have yet to see any MS Office program share a document. What constitutes "file sharing features"?
I believe parts of Base (the DB in OO.o 2) is partially written in Java, but since I only use Writer and Calc, I was happy to save ~20MB RAM and ~10 seconds off the load time.
I switched because I was used to DOS and I was taking to UNIX like an otter in a river. I wasn't happy with Win95's problems and when I found out I could get Linux for a reasonable price, including an introduction book.
Now I stay with Linux because of the power I have over the system. It does my bidding, not Microsoft's, Apple's, Sun's, or anyone else. I can find out every process that is running on my system nearly instantly, and I can kill almost any errant program (The only exception is if it hangs while waiting on the kernel which is hung waiting on a device driver). It hasn't crashed since April, and that was my bad. I can do everything I do with a computer (browse; e-mail; IM; rip, stream, and listen to music; watch, transcode, and master video; edit images; wordprocess; work on spreadsheets; balance my accounts; and sync data between devices. And let's not forget that I can program in practically any language used by more than 50 people.
The only thing still lacking is a large selection of video games (The kind I like anyway), but I'm so busy with other projects that I haven't even had time to re-install Windows 2000 (WinXP has never touched my hardware) on my games partition since I upgraded the guts of my workstation back in June.
I'm right there with you. I'm dying to get my hands on a smartphone, but I refuse to deal with any kind of lock-in like that. I keep hoping and searching for Linux-based smartphone/PDAphone. Something like a Zaurus but with a GSM transceiver and a Bluetooth module.
Or just use the package manager that comes with your Linux distribution and knows where all of that stuff is already. It's easier than having umpteen hundred thousand symlinks thrown around the system.
Honestly, I don't understand how it's that difficult.
That's a feature of your Window Manager, or the program you're using. I can't speak for GNOME since I haven't played around with it, but when I use Enlightenment, and on programs that don't remember it on their own, I can specify:
-Location -Size -Shaded (rolled up) state -Stacking Layer -Opacity * -Border style -Desktop (For multiple-desktop--not multi-monitor or virtual desktops, that's Location--setup) -Sticky state -Window List Skip -Shadowing *
* Requires support from the X server.
Plus I can track changes in the future.
I use it all the time to keep Firefox on my second virtual desktop, Evolution on my third, and GAIM on my fourth. It is contained in a file inside of my Enlightenment config directory, but I find it easier to right-click the title bar and click Remember... then fill it out. And yeah, it's incredibly useful.
Also, I haven't set Remember settings in GIMP, but whenever I open it, the pallet opens on the left edge of my screen, the Layers, Undo, etc. opens on the right, and the bulk of the screen in the middle is where my image opens up. It's been the same when I open Gimp.app on my iBook. I just had to drag them to where I wanted them, size them appropriately and they've stayed there. It was the same when I had to use it on a WinXP box over a year ago. It doesn't open on a different workspace, but that's what the WM is for.
I'm curious. Are you talking about MOSIX or OpenMOSIX? You mentioned the fork, but not which branch (Or both?) you were looking at. I've got a small (much smaller than yours) cluster of computers at home and on my Project List is investigating an OpenMOSIX cluster since most of my machines sit idle and it would be nice to use the extra CPU cycles and RAM for various things.
God I hope the day of the Linux smart phones comes soon. I'm still using a Nokia Series 40 phone (Nokia 3300) because I want something that will integrate with my Linux desktop like my Sharp Zaurus does, but with a good GSM module built in. I want something that I can cross-compile on my desktop then copy over to the phone and run.
One, you're incorrect (As others have stated) in your definition of refugee. Refugee is anyone who seeks refuge. It doesn't help that the majority of refugees lately have been from political problems.
Second, and this isn't just pointed at you but everyone who parrots "they could broadcast what they want" bullshit. There is one very effective method for stopping them from bruadcasting erroneous information: The Police, National Guard, Red Cross, and other volunteers there. If KAMP did start spouting hate, inciting to riot, or anything else you're afraid of, you stop them. As you so rightly pointed out, this is a crisis (And those in charge are woefully under-manned if they're still in crisis mode this far after the problems started) and I'm sure it would be well within their power to shut down KAMP if they were deliberately broadcasting maliciously. If they were just brodacasting incorrectly, there should be someone to correct them.
There is no excuse for denying help to those in need because the people in charge can't be bothered. KAMP wants to help the people who need it. Monitor it (Which shouldn't be hard if there are radios disbursed throughought the Astrodome and surrounding areas) and shut it down if necessary! Pre-emptively shutting it down is censorship even if it's just because they can't be arsed to grab one of the numerous volunteers who they turned away, hand him/her a radio, and tell him/her to report problems with KAMP.
I call bullshit and have since day one. BnetD has nothing to do with piracy. The piracy has already happened! The game has already been cracked and installed and there are multiple different ways to play it online without a valid CD-key. BnetD was just another way to play the game, one that emulated BattleNet. They could have just as easily setup a VPN which faked a LAN connection (That was popular with several people I know) and they still wouldn't have needed a CD-key. And like others have said, direct TCP/IP connections are possible and again, no CD-key necessary.
That was just a smoke screen. Their real beef was that one of their beta testers had leaked a pre-release copy of WarCraft 3 and people were playing it online. BnetD was just a convenient target for them to take out their aggression and give their lawyers some exercise.
First, Gnome Terminal is dog slow and fat-ass. I never use it because it's such a resource drain. I use Eterm 0.9.3 actually and with the exception of it not liking some UTF-8, it works great. It takes up about 1/6 of the resources GT does.
Second, are you using the nv (Open Source) driver or the nVidia (Proprietary) driver? I've noticed the nv driver is incredibly slow compared to the nVidia driver, especially dragging windows around. I don't see pixels update, but refresh goes to about 1fps when I'm dragging Eterm windows around. With the nVidia driver, it's smooth and slick, but it randomly locks up the machine if you have RenderAccel turned on.
I've been running E17 from that repository for a while now (On FC3) and I really like it. I still use E16 as my primary window manager because E17 is missing a few things I use (like remembering where windows go and a few other things) but it is really nice.
It's also fun running E17 inside a nested X server under E16. I had to pick up my Mac-loving graphic-artist friend after I showed him what a fully eye-candy E17 (animated background, animated menus, animated titlebars, etc) looks like without shutting down my X session.
I hope you see this in the forest of semi-FUD that followed your comment, but I will actually try to answer the question you've asked. The Fedora Core upgrade process is identical to the later Red Hat upgrade process, namely, it checks for previous versions and then shows you a screen asking if you want to upgrade (And providing a drop-down box listing discovered previous versions, usually just one option) or if you want to install fresh.
That said, FC1 to FC4 will be quite a bit of an upgrade, and while it can be done, it may not be as smooth as you'd like.
FC1 runs the 2.4 kernel, while all the newer FCs run 2.6. This can break some server-type programs (We're still running FC1 servers in several places here because some of the JRun stuff we run only runs on 2.4 and the dev group is really nervous about trying to upgrade to a 2.6-compatible JRun server).
If everything you're running came with FC1 (i.e. you didn't install a bunch of third-party commercial software like we did), then upgrading from FC1 to FC3 or FC4 should involve running the installer, picking through/etc for.rpmsave or.rpmnew files and merging them into your regular config files, then running updates (I run this last because it can generate new.rpmsave/.rpmnew files which aren't as important as the ones in the previous step). Most all user-level programs I've run have migrated just fine.
Now, after having said all of the above, I'm going to hold off upgrading to FC4 for a month. When I upgraded to FC3 on my test rigs, some of their newest features (At the time it was udev and selinux) didn't work so hot, but they straightened those out soon enough and things ran fine after that.
Dvorak may have gotten the Apple switch to Intel correct, but he's suffering from the same mistake a lot of others I've seen talk about this lately.
Namely, he seems to believe that Apple will make OS X run on any x86 hardware.
They won't. They said they're switching to Intel for chips. They didn't say they're porting to standard x86 architecture. They didn't say you'd be able to run OS X on your current hardware. They said they'd use chips from Intel. Period.
So, going from the past CPU switches they've done, it seems more reasonable to me that Apple will either have Intel design Mac-compatible boards for them or do it themselves, using all the existing technology they use now (OpenFirmware, PCI-x, etc). They won't just start slapping together off-the-shelf hardware that will dual-boot to Windows.
They'll still be Apple. They'll still be Different. And unless they get their hardware extremely cheaply and give up their profit margin, their rigs will still be expensive compared to the cheap commodity hardware that Linux enjoys so much success on.
One of my coworkers lost his Visor to the laundry like that. Fortunately I am way too attached to my phone to leave it in pants I'm not wearing. The only way I could love this thing more is if it were running Linux (My next phone will, no matter who I have to kill, because I'm tired of this proprietary playground they keep us in).
And yeah, AT&T and Cingular are the same company, sorta kinda not really. Besides, they're both a collection of mouth-breathing asshats who I refuse to even talk to anymore. I mean how hard is it to get a damn phone shipped to the right address? And how hard is it to take off a bogus charge?!
I just got one for my cell phone called MobileSafe. It was $6 from Handango and downloaded directly to my phone. That way I always have my account numbers, CC numbers, login info, and general notes encrypted with 168-bit 3DES (IIRC) on my phone protected by my master password. It's already saved my bacon more than once.
The only down-side is that I can't sync it with anything at home, but I generally don't have to update it very often, so when I do, I also write down the passwords in an encrypted text file on my home machine.
I agree that compatibility is the longest, most time consuming and boring part of developing a website.
The problem is that any book that has a comprehensive list of browser bugs will be outdated by the time it hits the shelves. Firefox and Mozilla change quickly, Opera and Sefari, nearly as quickly. IE is the only thing that hasn't changed much recently, but with IE7 coming out soon, the only thing the book could cover are old and unsupported browsers anymore.
Debian uses apt-get, which can be scripted to feed off a (group of) particular server(s). I don't use Debian, so I can't speak specifically to its strengths and weaknesses, but I'm sure someone else will.
Fedora/Red Hat systems have RPM and yum, both of which are network aware (Though no one uses the network functionality in RPM that I can find). I have smaller networks (40 machines in one, and 12 machines in another) that feed off of one yum server, which is a box that would otherwise be re-purposed into a boat anchor. I've even rolled my own RPMs (and signed them; none of my machines will install any un-signed package) and had those updated on all of my servers by the next morning, and created my own repositories (They can even be on different machines if you like). You can have different repositories setup for different groups, as well as share network-wide repos. It's really easy once you (or one of your geeks) can truck through the learning curve. And if they're a Linux or UNIX geek, they're probably part-way through already.
Ximian has Red Carpet, which I believe they sell "big boy" licenses to (Or did before Novell bought them). I haven't touched this in a while, but it seems like a cross between what I've outlined above and Windows Update Services / Software Update Services.
With all of the above (And I'm sure there are more such as Mandriva's urpmi or whatever they're using now) can be customized to only have the software you need, though I would recommend keeping mirrors of the base OS (for network installs, and installs of software later), and updates. Especially with > 200 desktops, I wouldn't want to tap-dance on any of the mirrors' nerves by beating on their servers nightly running my updates.
As for the rest of the question (upgrades, CALs, Office), faggedaboudit. Once you standardize on a distro, you can upgrade as often as you want; some distros will even let you roll out whole OS upgrades through the above-mentioned software updaters. You may want to think carefully about that, since most office workers I know don't like arriving in the morning to find their OS is different. However, it is easy to walk around with (several) net boot CDs and upgrade a bunch of systems over the network. And no Free or Open Source software requires anything as asinine as CALs.
Slashdot Mirror
Because two nodes on an IPV6 network wouldn't be stuck behind closed NAT firewalls and not able to communicate directly. Currently, if you're using Yahoo IM and want to send someone a picture, you have to relay it through Yahoo's servers, which causes a bottleneck because thousands of other people are doing the same and everything bottlenecks through Yahoo. With IPV6, both ends could have a public (possibly static) IP address, so person A could connect directly to person B and bypass the traffic jam at the server. It gets worse on something like eDonkey where if both ends are NAT'ed the transfer CANNOT happen.
As for removing the need for Skype, it wouldn't, but it would change. A Skype server would just need to know IP addresses so if person A wants to talk to person B, the Skype server just tells A "B is at 0:1:2:3:4:5" and A can then connect to B directly, instead of relaying through god-only-knows how many intermediates who may or may not be compromised or malicious. However, if you were setting up your own PBX with a static IP (And IPv6 has enough addresses that you could), then you wouldn't need a central authority (Skype server) to do the matchmaking.
Yes you'll still be firewalled, but the firewall will be at the end point, not in the middle somewhere that you have no control over (As in my case). The firewall rules would just change to a FORWARD rule instead of a "direct this port to this IP" rule.
IPV6 is about decentralizing, while NAT is essentially centralized (At the NAT server). If something can't change ports (Some VPNs, old protocols, not smart enough, etc), then you can't have more than one person behind the NAT using said protocol. NAT also causes problems because to the machine on the public-side of the NAT, only sees the address of the NAT gateway, not the address of the machine actually making the request. The implications of that are left up to the reader.
OWA wasn't like the other web-based e-mail interfaces at the time. It was the first to use the XMLHTTP object, the basis for all current AJAX and JSON RPC platforms. They added it to Outlook Express so it could communicate with Hotmail without refreshing the page all the time (Exactly what AJAX does now, but using Outlook Express for the interface instead of Javascript+DHTML). Then they didn't branch it out until other developers started using it for things like Google Maps, GMail, etc. Now their light bulbs are coming on and they're starting to think "Maybe we can use this for more than just Hotmail."
Red Hat (beginning in Fedora Core 4) has Xen RPMs for their kernels as well. Each time they release a new kernel, there are Xen versions as well. They want to get it into Linus' kernel tree so everyone can use and test it, since other distros are leery of using Red Hat stuff but fine with anything Linus has integrated into his kernel.
Try it. If nothing blows up, it must not have been important. If something does, turn it back on.
(I've got a good idea what it is, but I'm not going to tell)
I must plead ignorance regarding MS Office file sharing features. Every office I've worked in has used Windows (or at least SMB) file servers to share files. I have yet to see any MS Office program share a document. What constitutes "file sharing features"?
I believe parts of Base (the DB in OO.o 2) is partially written in Java, but since I only use Writer and Calc, I was happy to save ~20MB RAM and ~10 seconds off the load time.
I switched because I was used to DOS and I was taking to UNIX like an otter in a river. I wasn't happy with Win95's problems and when I found out I could get Linux for a reasonable price, including an introduction book.
Now I stay with Linux because of the power I have over the system. It does my bidding, not Microsoft's, Apple's, Sun's, or anyone else. I can find out every process that is running on my system nearly instantly, and I can kill almost any errant program (The only exception is if it hangs while waiting on the kernel which is hung waiting on a device driver). It hasn't crashed since April, and that was my bad. I can do everything I do with a computer (browse; e-mail; IM; rip, stream, and listen to music; watch, transcode, and master video; edit images; wordprocess; work on spreadsheets; balance my accounts; and sync data between devices. And let's not forget that I can program in practically any language used by more than 50 people.
The only thing still lacking is a large selection of video games (The kind I like anyway), but I'm so busy with other projects that I haven't even had time to re-install Windows 2000 (WinXP has never touched my hardware) on my games partition since I upgraded the guts of my workstation back in June.
I'm right there with you. I'm dying to get my hands on a smartphone, but I refuse to deal with any kind of lock-in like that. I keep hoping and searching for Linux-based smartphone/PDAphone. Something like a Zaurus but with a GSM transceiver and a Bluetooth module.
Or just use the package manager that comes with your Linux distribution and knows where all of that stuff is already. It's easier than having umpteen hundred thousand symlinks thrown around the system.
Honestly, I don't understand how it's that difficult.
That's a feature of your Window Manager, or the program you're using. I can't speak for GNOME since I haven't played around with it, but when I use Enlightenment, and on programs that don't remember it on their own, I can specify:
... then fill it out. And yeah, it's incredibly useful.
-Location
-Size
-Shaded (rolled up) state
-Stacking Layer
-Opacity *
-Border style
-Desktop (For multiple-desktop--not multi-monitor or virtual desktops, that's Location--setup)
-Sticky state
-Window List Skip
-Shadowing *
* Requires support from the X server.
Plus I can track changes in the future.
I use it all the time to keep Firefox on my second virtual desktop, Evolution on my third, and GAIM on my fourth. It is contained in a file inside of my Enlightenment config directory, but I find it easier to right-click the title bar and click Remember
Also, I haven't set Remember settings in GIMP, but whenever I open it, the pallet opens on the left edge of my screen, the Layers, Undo, etc. opens on the right, and the bulk of the screen in the middle is where my image opens up. It's been the same when I open Gimp.app on my iBook. I just had to drag them to where I wanted them, size them appropriately and they've stayed there. It was the same when I had to use it on a WinXP box over a year ago. It doesn't open on a different workspace, but that's what the WM is for.
And never return to the US, since now you're a criminal there.
I'm curious. Are you talking about MOSIX or OpenMOSIX? You mentioned the fork, but not which branch (Or both?) you were looking at. I've got a small (much smaller than yours) cluster of computers at home and on my Project List is investigating an OpenMOSIX cluster since most of my machines sit idle and it would be nice to use the extra CPU cycles and RAM for various things.
God I hope the day of the Linux smart phones comes soon. I'm still using a Nokia Series 40 phone (Nokia 3300) because I want something that will integrate with my Linux desktop like my Sharp Zaurus does, but with a good GSM module built in. I want something that I can cross-compile on my desktop then copy over to the phone and run.
Two things:
One, you're incorrect (As others have stated) in your definition of refugee. Refugee is anyone who seeks refuge. It doesn't help that the majority of refugees lately have been from political problems.
Second, and this isn't just pointed at you but everyone who parrots "they could broadcast what they want" bullshit. There is one very effective method for stopping them from bruadcasting erroneous information: The Police, National Guard, Red Cross, and other volunteers there. If KAMP did start spouting hate, inciting to riot, or anything else you're afraid of, you stop them. As you so rightly pointed out, this is a crisis (And those in charge are woefully under-manned if they're still in crisis mode this far after the problems started) and I'm sure it would be well within their power to shut down KAMP if they were deliberately broadcasting maliciously. If they were just brodacasting incorrectly, there should be someone to correct them.
There is no excuse for denying help to those in need because the people in charge can't be bothered. KAMP wants to help the people who need it. Monitor it (Which shouldn't be hard if there are radios disbursed throughought the Astrodome and surrounding areas) and shut it down if necessary! Pre-emptively shutting it down is censorship even if it's just because they can't be arsed to grab one of the numerous volunteers who they turned away, hand him/her a radio, and tell him/her to report problems with KAMP.
I call bullshit and have since day one. BnetD has nothing to do with piracy. The piracy has already happened! The game has already been cracked and installed and there are multiple different ways to play it online without a valid CD-key. BnetD was just another way to play the game, one that emulated BattleNet. They could have just as easily setup a VPN which faked a LAN connection (That was popular with several people I know) and they still wouldn't have needed a CD-key. And like others have said, direct TCP/IP connections are possible and again, no CD-key necessary.
That was just a smoke screen. Their real beef was that one of their beta testers had leaked a pre-release copy of WarCraft 3 and people were playing it online. BnetD was just a convenient target for them to take out their aggression and give their lawyers some exercise.
A couple of things:
First, Gnome Terminal is dog slow and fat-ass. I never use it because it's such a resource drain. I use Eterm 0.9.3 actually and with the exception of it not liking some UTF-8, it works great. It takes up about 1/6 of the resources GT does.
Second, are you using the nv (Open Source) driver or the nVidia (Proprietary) driver? I've noticed the nv driver is incredibly slow compared to the nVidia driver, especially dragging windows around. I don't see pixels update, but refresh goes to about 1fps when I'm dragging Eterm windows around. With the nVidia driver, it's smooth and slick, but it randomly locks up the machine if you have RenderAccel turned on.
I've been running E17 from that repository for a while now (On FC3) and I really like it. I still use E16 as my primary window manager because E17 is missing a few things I use (like remembering where windows go and a few other things) but it is really nice.
It's also fun running E17 inside a nested X server under E16. I had to pick up my Mac-loving graphic-artist friend after I showed him what a fully eye-candy E17 (animated background, animated menus, animated titlebars, etc) looks like without shutting down my X session.
More likely he lost his Linux team (Yes, one guy can be a "team" in managerspeak) and hired a bunch of MSCEs to replace the site.
Or farmed it out to a Microsoft-based development and hosting company.
Was that before or after you de-neutered it?
I hope you see this in the forest of semi-FUD that followed your comment, but I will actually try to answer the question you've asked. The Fedora Core upgrade process is identical to the later Red Hat upgrade process, namely, it checks for previous versions and then shows you a screen asking if you want to upgrade (And providing a drop-down box listing discovered previous versions, usually just one option) or if you want to install fresh.
/etc for .rpmsave or .rpmnew files and merging them into your regular config files, then running updates (I run this last because it can generate new .rpmsave/.rpmnew files which aren't as important as the ones in the previous step). Most all user-level programs I've run have migrated just fine.
That said, FC1 to FC4 will be quite a bit of an upgrade, and while it can be done, it may not be as smooth as you'd like.
FC1 runs the 2.4 kernel, while all the newer FCs run 2.6. This can break some server-type programs (We're still running FC1 servers in several places here because some of the JRun stuff we run only runs on 2.4 and the dev group is really nervous about trying to upgrade to a 2.6-compatible JRun server).
If everything you're running came with FC1 (i.e. you didn't install a bunch of third-party commercial software like we did), then upgrading from FC1 to FC3 or FC4 should involve running the installer, picking through
Now, after having said all of the above, I'm going to hold off upgrading to FC4 for a month. When I upgraded to FC3 on my test rigs, some of their newest features (At the time it was udev and selinux) didn't work so hot, but they straightened those out soon enough and things ran fine after that.
Dvorak may have gotten the Apple switch to Intel correct, but he's suffering from the same mistake a lot of others I've seen talk about this lately.
Namely, he seems to believe that Apple will make OS X run on any x86 hardware.
They won't. They said they're switching to Intel for chips. They didn't say they're porting to standard x86 architecture. They didn't say you'd be able to run OS X on your current hardware. They said they'd use chips from Intel. Period.
So, going from the past CPU switches they've done, it seems more reasonable to me that Apple will either have Intel design Mac-compatible boards for them or do it themselves, using all the existing technology they use now (OpenFirmware, PCI-x, etc). They won't just start slapping together off-the-shelf hardware that will dual-boot to Windows.
They'll still be Apple. They'll still be Different. And unless they get their hardware extremely cheaply and give up their profit margin, their rigs will still be expensive compared to the cheap commodity hardware that Linux enjoys so much success on.
One of my coworkers lost his Visor to the laundry like that. Fortunately I am way too attached to my phone to leave it in pants I'm not wearing. The only way I could love this thing more is if it were running Linux (My next phone will, no matter who I have to kill, because I'm tired of this proprietary playground they keep us in).
And yeah, AT&T and Cingular are the same company, sorta kinda not really. Besides, they're both a collection of mouth-breathing asshats who I refuse to even talk to anymore. I mean how hard is it to get a damn phone shipped to the right address? And how hard is it to take off a bogus charge?!
(breathe, just breathe)
I just got one for my cell phone called MobileSafe. It was $6 from Handango and downloaded directly to my phone. That way I always have my account numbers, CC numbers, login info, and general notes encrypted with 168-bit 3DES (IIRC) on my phone protected by my master password. It's already saved my bacon more than once.
The only down-side is that I can't sync it with anything at home, but I generally don't have to update it very often, so when I do, I also write down the passwords in an encrypted text file on my home machine.
I agree that compatibility is the longest, most time consuming and boring part of developing a website.
The problem is that any book that has a comprehensive list of browser bugs will be outdated by the time it hits the shelves. Firefox and Mozilla change quickly, Opera and Sefari, nearly as quickly. IE is the only thing that hasn't changed much recently, but with IE7 coming out soon, the only thing the book could cover are old and unsupported browsers anymore.
This has been solved so many ways...
Debian uses apt-get, which can be scripted to feed off a (group of) particular server(s). I don't use Debian, so I can't speak specifically to its strengths and weaknesses, but I'm sure someone else will.
Fedora/Red Hat systems have RPM and yum, both of which are network aware (Though no one uses the network functionality in RPM that I can find). I have smaller networks (40 machines in one, and 12 machines in another) that feed off of one yum server, which is a box that would otherwise be re-purposed into a boat anchor. I've even rolled my own RPMs (and signed them; none of my machines will install any un-signed package) and had those updated on all of my servers by the next morning, and created my own repositories (They can even be on different machines if you like). You can have different repositories setup for different groups, as well as share network-wide repos. It's really easy once you (or one of your geeks) can truck through the learning curve. And if they're a Linux or UNIX geek, they're probably part-way through already.
Ximian has Red Carpet, which I believe they sell "big boy" licenses to (Or did before Novell bought them). I haven't touched this in a while, but it seems like a cross between what I've outlined above and Windows Update Services / Software Update Services.
With all of the above (And I'm sure there are more such as Mandriva's urpmi or whatever they're using now) can be customized to only have the software you need, though I would recommend keeping mirrors of the base OS (for network installs, and installs of software later), and updates. Especially with > 200 desktops, I wouldn't want to tap-dance on any of the mirrors' nerves by beating on their servers nightly running my updates.
As for the rest of the question (upgrades, CALs, Office), faggedaboudit. Once you standardize on a distro, you can upgrade as often as you want; some distros will even let you roll out whole OS upgrades through the above-mentioned software updaters. You may want to think carefully about that, since most office workers I know don't like arriving in the morning to find their OS is different. However, it is easy to walk around with (several) net boot CDs and upgrade a bunch of systems over the network. And no Free or Open Source software requires anything as asinine as CALs.