Slashdot Mirror
Time To Dump XP?
An anonymous reader writes "Gartner is saying it's time to plan your migration now (if you havent already done it). I for one know my company still has loads of users still on XP, citing training costs (time and money) rather than software license fees. Is my company alone in wanting to stay in the 1990s or is Windows 7 the way forward?"
...is my company still using Windows XP SP2, but we are still using IE6. Feh...and they complained that Audacity was a security risk because it was "open source, so anyone could hack it".
Insanity.
Living With a Nerd
The main reason, in my mind, to upgrade is being able to effectively use 64-bit machines fully--and have more than 4GB of RAM.
Yes you need new machines to do this, but really, if you are buying NEW machines, you should probably upgrade. The question then becomes a matter of whether or not new machines are worthwhile. Your old machines may be still serviceable, but would newer machines result in getting work done enough faster to offset (even partially) the cost of the upgrade.
In many cases, the answer is no--a LOT of secretaries & folks that mainly do word processing are better off just staying where they are--their machines are fast enough for what they do, and additional RAM & extra cores aren't going to make a difference.
That said, if you are doing statistical analysis, engineering, graphic design, programming (and compiling), and a number of other jobs, then you should ABSOLUTELY be on a very aggressive upgrade schedule. Additionally, 8GB of RAM is more than just a good idea for many of those jobs--some of them should be stuffing as MUCH memory as they can into their machines so that they can do their jobs more efficiently.
In any work setting the bottleneck for employee performance should not be the environment or resources, but rather human capacities. That's the ideal. Obviously cost of achieving that and other considerations prevent most companies from getting to the point where that's true--but it should be the goal.
So either move to Win7-x64 OR move to another 64-bit OS with lots of power & memory in the hardware. Staying where you are only makes sense if you are doing mostly word processing.
"We don't know what we are doing, but we are doing it very carefully,..." Wherry, R.J. Personnel Psychology (1995)
Another editor writes an idiotic title??
Let's answer this simply, since the article has a simple title: "Is it finally time to dump XP?" NO. It's 2010. By your own article's admission support ends in 2014!
FTFA: "IT departments need to dump Microsoft's Windows XP operating system (OS) before the software vendor ends support for it in April 2014"
Thanks, Capt. Obvious!
Also FTFA: "the sooner the better as many new versions of applications are not expected to support XP beyond 2012."
What applications? Do these people live in the enterprise? Vendor apps are the slowest to migrate to any new OS. That's one of the major reasons why migrations happen so slowly. The other is money. In a down economy you're simply not going to see wholesale adoption of Windows 7 when there's no funding and companies can pull profits from apps that are working now! This is all fun to sit and talk about and kick up some worry but the reality is when you go back to your CIO or IT manager funding will win out. They're going to wait till they get closer to EOL and hope the economy turns around and frankly that's what they should do.
So that in a few years people don't arrive having never used XP and immediately start cursing at "this stupid system". Little things like the improved taskbar, the window snap and so on all work their way into how you interact and you suddenly feel lost without them.
Software isn't the problem, people who use 7 at home and don't want to go back to XP at work are.
That and the fact Vista and 7 don't support IE6. If the OS can't support it, IE6 is dead.
How many people can read hex if only you and dead people can read hex?
Our migration is probably similar to many other companies. Here's what we're doing in case anyone is curious how this roadmap looks in a reasonably sized company (multilocation, etc, etc):
1. We got our first Win7 system to test a few months ago. We discovered almost everything worked, but our VPN clients should be updated, our AV needed some updating, and really we should be on Office 2010. The nice thing there is we can eradicate Office 2003 once and for all.
2. So, that really prompts some server upgrades that we've been planning for a while anyway. We're going to consolidate a lot of servers onto VM'ed boxes. Most of our stuff (was) running Server 2003, with the exceptions of our domain controllers which we updated to 2008 last year. Exchange 2010 (from 2003) was planned for a while, so we pulled the trigger on that one. That also prompts an upgrade of BES (Blackberry Enterprise Server) from 4.1 to 5.0. Our asset tracking also needed some attention in order to make sure we don't populate it with garbage when new machines arrive. We're hoping to have Exchange completely migrated by the end of July using a slow migration tactic instead of cutting over in the middle of the night. The goal here is to leave some app servers on 2003 until the new version of MS's server platform comes out, then update to that on an application by application basis.
3. So.. that means there's a fair amount of work to do before we want to consider replacing the user machines. I suspect most companies are in that boat. I think most companies are itching to replace XP - it's getting pretty tough to maintain these days and pretty outdated. Plus, no (sane) company actually upgrades machines from XP to Win7 - you transition to Win7 when your leases expire or you need to purchase a new desktop/laptop. Upgrading is in no way cost effective. Therefore, based on a lifecycle of 3 - 4 years per machine, we'll see XP still being used for 2 - 3 years at least for light duty.
Now, the really crazy part? Most suppliers are pushing 32-bit Win7. That means the 32-bit legacy is going to continue to haunt us when we could have transitioned to Win64.
----- obSig
The cynic might observe that in many companies the employees are there because they are unemployable elsewhere...
Fanatically anti-fanatical
While I agree that most of the marketing and Aero-fluff is useless, there are a few very big improvements I have to point out with Windows 7: -Security is MUCH better than XP. I'm not calling it *nix level or anything, but it is much better. I've seen a huge decrease in spyware infections on Win 7. -Bitlocker is secure, fast and accessible for most users. Again, Truecrypt may be better, but this is a good thing for the OS to have native. -The systems management functions (i.e. power settings controlled by GPO, proper grouping of the event logs, etc) are far superior in Windows 7. I would much rather manage 1000 Win 7 desktops than 250 XP.
Grammer Nazis - I mod you "troll" unless you actually add something on-topic. Yes, I know I have mispellings in my sig.
Really? OK, I posted earlier about all the reasons that moving to *nix from Windows is hard, but, well, these are none of them. Every problem you list has at least one often more solutions in the Unix world:
File sharing: Several options. Assuming all Unix machines, NFS is by far the easiest. As long as all users are authenticating off the same directory their UIDs will match between systems. This is the "go to" Unix file sharing system, but there's other options. You can use Samba of course, and there's a few nifty distributed file systems out there that are starting to get mature. The first two options will work on any Unix system including Macs. The distributed solutions are spottier in what they support, being often new.
Centralized Login: Two major solutions. LDAP and NIS+. LDAP is by far the more modern and and scalable, though it can be slightly tricky to set up. Very slightly, nothing any half competent admin can't figure out. Original NIS is also an option, but is getting long in the tooth and has some security problems. Macs are perfectly capable of using LDAP, and I assume NIS as well, though I've never tried
Policy management: This is a little less defined in the Unix world than it is in Windows, but still manageable. Most of these policies are managed by various text files in Unix, so what I typically do is run a script when I first install them to set everything the way I want it. In the unlikely event I need to make a change I just change one system and propagate it to all the others. I have a script that copies a file where ever it needs to go on every machine in the network. You can also automate this through rsync though I've never personally bothered to set this up. I've never run a network complicated enough to really need it. The hard part here is if you have a heterogeneous Unix environment, since nearly all Unix's insist on using different files and different syntax to manage this stuff. I'll admit this is a slightly weak area, but definitely manageable.
Update Management: It's trivial to setup a local repository for any *nix repository system I am aware of. Setup you client to update to the local repository and test updates before you put them on the repo server.
Mostly this stuff is trivial in Unix/Mac environments. When I manage heterogeneous networks my problem is usually getting the Windows boxes to play nice with everyone else. Unix and Mac machines will all happily share files and directory data with each other, even across different OS's and hardware platforms, while the Windows boxes insist on playing their own little game. Samba helps with file sharing, but getting everyone to log in against the same network shared directory is an undertaking and a half.
I don't need a million points of light, just two points of multi-mode fiber and a 10 Gig-E router.
In addition, in specialty environments (e.g., some manufacturing shops), you're often constrained by what your other software vendors and equipment providers will support. A number of our key tools (e.g., 3D CAD) support Windows 7, but we have many legacy tools that only run on XP (or earlier environments!). In some cases, vendors are only supporting newer OSes if we also upgrade the machines that are tethered to the workstations--that means it's not as simple as buying a new PC and a new version of software, but instead could mean a $200,000.00 investment in a manufacturing device that will again be tethered to a specific build of Windows.
I use irony whenever I can, but my shirts are still wrinkled...