When general support for Windows NT 4.0 ends Friday, it may open up opportunities for Linux, and other operating systems. In other news, new critial flaws in Windows have yet to be patched.
NT’s Retirement Opens Windows for Linux
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
I’ve worked in NT/Windows shops for years, and I’ve seen a little Linux here and there in some of them. I’ve set up many an Apache/MySQL server on Windows-dominated networks, in fact, and think Apache on Linux is a great combination. People running NT4 at this point may be unhappy about the cost and effort involved in upgrading to another MS product, but I’m willing to bet that not one in a thousand has any intention of shifting away from the Windows domain model and single sign-on security architecture they’ve been accustomed to all these years. Linux advocates would like to think that Windows shops can cost-effectively migrate to a Unix-like operating system for their core computing needs, but I’ve yet to see a place where this made sense or was actually achieved. This is a silly article.
Samba 3 can replace (nearly perfectly) a full network of file and printing NT servers including Windows domain model and single sign-on security architecture. You can even replace servers so your users will never know the difference apart from better uptime and speed and no virus problem.
You only need to read and understand at least a part of this docs http://us2.samba.org/samba/docs/man/Samba-Guide/ or hire a good Samba admin.
Probably we will see not 0.1% of servers moved to Samba but at least 10%
This is too good to be true. I can’t see anyone switching. I’ve been using linux primarily for the last 5 years and while it’s more than fine for me, Joe Business likes his NT4. Even if that means finding external support, I believe most businesses will stick with NT4 just because its more business-like to use a proprietary operating system and its too warm in Microsoft’s bed. Sorry if this comment sounds silly but thats how it seems to be. Silly!
I work for a large company in the UK, and they’ve arranged a special paid support deal with MS, rather than upgrade/replace the NT systems running on the network.
The associated costs of having someone learn samba, or get in a team of samba professionals, and then migrate all the services is obviously more expensive than just paying for extended support until the systems are upgraded to a newer windows version.
And herein lies the problem – MS shops will always get a better deal from MS. If they mention the word “RedHat”, “Samba” or “Linux” in general in front of their MS rep, they get a hefty discount.
Linux will only become commonplace in these sorts of scenarios when it has the home user adoption, so existing infrastructure staff know how to support the day-to-day running of the file/print/authentication services on non-windows platforms.
Does it mean, that on dec 31 they (MS) will remove all the NT4 hotfixes/service packs from Windows Update? Or it means they only dont release new ones?
Changing systems is not free, but it is a one time cost. The initial cost of just upgrading to a new version of Windows may be less, but the long term cost may be more. You need to evaluate the options and then determine what is the best solution for your application. If you lock yourself into whatever one provider offers, then you will definately pay more in the long run.
/Linux advocates would like to think that Windows shops can cost-effectively migrate to a Unix-like operating system for their core computing needs, but I’ve yet to see a place where this made sense or was actually achieved. This is a silly article./
Sure they can migrate,
but they are no longer called a “Windows shop”.
How could you state that headline and take yourself seriously?
Seriously, NT is alive and well, in the form of Windows 2000 (NT 5.0) and XP (NT 6.0) only the name it is sold under has changed. Granted, they’ve grown quite a bit from NT 4.0 which is being retired officially by Microsoft, but they are every bit NT (all 32 or 64 of them, depending on your platform).
I’m not stating which OS is superior here, I’m only stating reality that NT is NOT retiring as the headline claims: only an older revision of it is losing official support. Would you waste your time on stating that nobody is officially supporting a Linux kernel of 1.x????
Just because msft isn’t actively supporting NT4 doesn’t mean that your NT4 will blow-up. NT4, with a full GUI, runs fine on anything from a 486 on up.
I think the total cost of converting from NT4 to Linux would be far higher than the cost of converting form NT4 to Win2003. You would have to change all of you applications, re-train everybody etc. Also, the Linux that is widely trusted in business in RedHat – and RedHat is expensive.
They mean NT4. It’s amazing how many people still use it. Anyway, over here we’re replacing old NT servers left and right with Linux, and we won’t look back. Web servers and mail servers are the easiest, file servers and domain controllers are a bit trickier, but still very doable.
“assist customers upgrading from Windows NT Server 4.0 to the Windows 2000 Server and Windows Server 2003 families,”
Doesn’t 2000 Server expire in 5 years? Well, there might be some maintance mode happening after that, not sure. Can any one comment on the time frame for 2000 server?
“”If a Windows NT Server is doing what’s supposed to do, the cheapest thing is to let it be and continue to realize ROI [return on investment] on the original investment,””
This logic is flawed, period. Your going to have to upgrade no matter, what. In a year, or two of extended life, you will have to migrate, so not only are you paying for premium extended support, you still have to upgrade. Companies should have switch a long time ago to:
1) 2003 Server
2) Another Operating System
This could be a last ditch effort for most companies too stiff MS in two upgrade plans. Buy two years of support and then hit the new Longhorn Server platform and bypass 2000, 2003 server. In this case, extended support might be cost justified.
A corporation will prefer to pay redmond, instead to put this money on his staff to give them tool they want. Even is samba4 is way better, that does not really matter. With MS they have good support, or at least the sentiment they are VIP. The persons who take decisions is a business man, not a technological _geek_. the first question will be – but if we have a problem, who can we call? and Will that compagny still exist in 2-3-5 years? You have an answer, too late, they already decide that it is too risky.
COST is NOT a point as long it’s does not make them uncompetitive.
And anyway, how a programmer who work for free can make a good program? I often have to deal with that, how crazy it may sound, but people in other field than science, cannot really understand why someone can put the best of them for ideological reason. To chanlenge themself for the pure pleasure of being proud of want they do/did. I rarely acheive that when I code at work, to many restriction, no technological foresight. I would say that 90% of my client are really stunt to that and even if I make them understand, they most of the time decline because of insecurity. ( I code internet/intranet service on my pass time for a couple of client, and what I try to sell is mostly Linux/Apache-ftp server )
Anyway, my point is, what will make linux gain ground, is the creation of totaly new compagny chanlenging the old one who are scared to innovate. If the cost/performance of linux show up, those new business will make switching innevitable for the viability of the other.
“If we make peaceful revolution impossible, we make violent revolution inevitable.” — John F. Kennedy
amazon prove that, they completly trow competition away.
Doh! I put “@Jonathan Thompson” in the email address field of the previous comment. Please disregard my mistake!
To make this post worthwhile:
The last two Windows servers we have are a domain controller/file server and a database server for our current pseudoERP. We’re getting a new ERP soon, and one of the requirements is that it should use a database that runs on Linux. So that’s going away soon. There file server will be a bit trickier, since it serves as a domain controller, so we have to be extra careful when migrating all the stuff over to Samba, and making the changes so smooth that the users don’t notice.
I find filename conversions a major annoyance: a few people have Linux desktops and access the Windows file server. On the command line, it’s quite a pain to have to type “/Important Files/Important Spreadsheet Blah Blah”. I wonder how other people have approached this other than just sucking it up.
“NT’s Retirement Opens Windows for Linux”
Uh, no it doesn’t. What it does is open the need to upgrade to 2000 or 2003. You think some corporate entity will decide to chuck the NT bandwagon jsut to go the Linux/Samba route? You’re sadly mistaken. Companies fall into the “devil you know to the devil you don’t” mentality far more than you may like to believe.
This will only cause more 2000 and 2003 installations. It’s about time too, because NT was asking for the plug to be pulled a long time ago.
What you say has some truth in it, although it’s expressed a little pedantically. Let’s just say Windows NT(r) 4 is retiring then. Furthermore, Windows XP is 5.1, not 6.0 (5.1.2600 indicating that the 2600th build went to retail, to be precise).
Back onto the topic, I use Samba and like it very much (having once used Win2K and AD to power te site). However, there are two fundamental weaknesses that I would like to see addressed:
* The inherent complexity in integrating and managing it via LDAP. Simply, it needs to be better documented, more consistent and the procedures more accessible. One has to be a veritable geek to handle this aspect at present, IMHO.
* While the NT4 domain based model is probably good enough for most shops, the inability to easily apply system policies is a real problem. Currently, one must hunt around for the old poledit.exe tool, import some 3rd party hacked-up .ADM files (in order to support the features of newer system such as Win2K/XP), run the policy editor on a Windows box then dump the policy files in the netlogon share. Compared to the current implementation in Win2K AD this is very primitive and an administrative hassle.
I suppose my point is that we can speculate until the cows come home about corporate policy but the notion of “switching” will be an uphill struggle for as long as there are genuinely useful features that are either missing, poorly implemented or where the parts just don’t fit together perfectly and/or lack easy-to-use administration tools.
As I said, I love Samba and will continue to use it; of course, it has some unique tricks up its sleeve and is a marvellous piece of engineering. And it makes for a great domain controller in a fair few situations. But the fact is that some features that many Windows administrators may require (and rightly expect) are either missing or incredibly difficult to implement and administer. In light of this, why would we expect them to switch?
I’ve no idea as to how prevalent Samba is in “enterprise” environments. One thing I have noticed with the open-source crowd in general is that the vast majority of folks I run into who use Samba seem to be using it for trivial tasks (like as not, in a home context) and don’t really understand SMB/CIFS/NetBIOS and the Windows networking model at all. Conversely, the hard-core Unix admins seem to be mostly using Linux to different ends i.e. they don’t use Samba at all. In fact, I hang out in a server-oriented Linux IRC channel and I’m the only person that I know for sure that uses it in that channel.
My feeling is (again, IMHO) that we are generally unsympathetic to the practices and requirements of administrators who manage Windows networks and that it is fuelled, in no small part, by a combination of disinterest and animosity towards things that are associated with Microsoft in general. Perhaps this is something that could be addressed before we realistically expect people to change their practices – for whatever reason.
This might open opportunities for a small to mid-size business, and even then it would be to replace servers. Most businesses I’ve seen are using 2000/2003 Server and XP or 2000 Pro desktops already. I don’t know of any businesses that are still running NT 4 servers. Granted, I don’t know the inner workings of but a few businesses.
At one point, I had a NT4 server box running Apache for our departments unofficial help page at the MS internal helpdesk. The Nimda storm required me to clear the log files once in a while.
It would be fun from a hobbyist standpoint to fold, spindle, and gene-splice NT 4.0 with open-source software to get a fast, clean server or desktop OS.
Why would a company already investing money in software spend countless thousands re-writing all their applications to run on Linux?
They don’t, they just upgrade or purchase new equipment as need basis.
When a company is running NT 4 Server, MS gives deep discounts and help to make the upgrade.
They will be upgrading hardware/software to Windows Server 2003 if they have their present software tuned for their needs.
We upgraded all of our Workstations to Windows XP Professional about a year ago, locked down the permissions to all users are listed as ‘user’ and ‘power user’ with selective rights. Using Active Directory implementing these solutions have enabled us to have a secure network. We also have our Network locked down on IP addressing and virus/firewall software also.
You can have a Windows XP Professional clients running all the applications, you have to be able to understand permissions to a granular level and setup the accounts correctly. It takes time, but after the project was completed, we have had ZERO cases of spyware, viruses or any other problems. We could have possible switched to Linux but the cost of Redhat and re-writing applications and making them ‘work’ for all of our users was sky high.
We are very pleased with Microsoft and their support, kudo’s to them.
Damon Riptkin
Maybe, a lot of people think that moving to linux from microsoft products is stupid. Well, if its like that than i am afraid the number of stupid people is increasing. You need proof? I do have that…
http://www.crn-india.com/breakingnews/stories/55927.html
It opens the door for Windows, since Linux commands a whopping 3% of user desktops, and most of that is for hobbyists and advocates. Businesses in general have no desire to migrate to Linux, and anything touted as contrary to this is simply wishful thinking.
The fact that you think migrating to a new platform is a “one time cost” illuminates your ignorance regarding the costs involved with running an IT shop. After all the planning and implementation costs, there are the costs to maintain the technology. This involves training of both support staff and end users, constructing test labs, developer costs (read: most likely rewriting millions of lines of code for your average NT shop), etc etc. Your statement rings of generic Linux zealot-ism, as it is shallowly anchored in reality and reflects the opposite of what most of the suits already know: That any migration to Linux either in part or in whole, would quickly consume even the most generous IT budget.
“This could be a last ditch effort for most companies too stiff MS in two upgrade plans. Buy two years of support and then hit the new Longhorn Server platform and bypass 2000, 2003 server. In this case, extended support might be cost justified.”
companies want their older software and os to be compatible. who knows what software situation each company has and win 2003 didnt have sp1 released yet and it wont be until march 2005 why didnt ms wait then kill ms support. its only three more months.
Your comments smell of troll, but what the hell. Migration is simple or not depending on what you are using your servers for. We migrated email and web first. We were very dissatisfied with Windows. Too expensive, and not reliable enough, at least NT4. We were rebooting our email server a few times per month. Rather than to shell out a few grand for some W2k3 licenses, Linux was the best solution for email and web servers. No pain to switch whatsoever. That was a year and a half ago, and I’ve rebooted the web server twice, once to upgrade a kernel, and the other to physically move the server to another room. Not so with NT. Granted, email and web are the easiest applications to migrate, since they have basically no impact on the end user. But chances are, any custom server app that is as old as NT is probably running on some crusty old Unix box. I’d say most NT installations out there is running web, email, domain and file services, for which it makes no financial sense to shell out that kind of license money, especially if you have fairly competent staff on board.
It opens the door for Windows, since Linux commands a whopping 3% of user desktops
I think the article is mainly concerned with servers, not desktops. I tend to agree: replacing older servers, especially mail, file and print servers with Linux makes sense, especially considering the costs of moving to Win2K or Win2K3 Server.
Businesses in general have no desire to migrate to Linux, and anything touted as contrary to this is simply wishful thinking.
Lots of businesses already use Linux in the server room. IBM, RedHat and Novell have been quite aggressive in pushing Linux servers, and the results are quite encouraging: growth in Linux server numbers is unprecendented, according to IDC and the Gartner group.
The fact that you think migrating to a new platform is a “one time cost” illuminates your ignorance regarding the costs involved with running an IT shop. After all the planning and implementation costs, there are the costs to maintain the technology.
This is one of the reasons why Linux makes sense in the server room, as it actually requires a smaller number of admins to manage the same number of servers. Simply put, Linux servers require less maintenance.
This involves training of both support staff and end users
End users do not need training when servers are being replaced…
developer costs (read: most likely rewriting millions of lines of code for your average NT shop)
Rewriting millions of lines? That seems a bit exaggerated, though I guess in some cases it could happen. Well, if there are a few legacy NT apps which run off the servers that can’t be easily replaced or ported, it’s always possible to use virtualization software such as VMWare for those apps. It’s a small investment to make, and works very well for nearly all apps.
Your statement rings of generic Linux zealot-ism, as it is shallowly anchored in reality and reflects the opposite of what most of the suits already know:
Well, your statement rings of the usual anti-Linux FUD, and unless you provide data to support your claims you shouldn’t speak for “the suits”. Now that trusted IT names such as IBM and Novell are behind Linux, lots of suits are ready to take another look at it – hence the dramatic advances made by Linux in the server room.
That any migration to Linux either in part or in whole, would quickly consume even the most generous IT budget.
That is an unsubstantiated affirmation. In fact, most TCO studies shows that switching to Linux in the data center can save a LOT of money.
In other words: FUD, FUD, FUD.
I ask once more (it disappered among these nonsense posts silently)
Does it mean, that on dec 31 they (MS) will remove all the NT4 hotfixes/service packs from Windows Update?
Or it means they only dont release new ones???
Or it means they only dont release new ones???
They won’t release any new ones to none paid-for-support customers
“We have both a NT4 SP6 Print Server and a Linux Print Server and guess which one constantly needs attention?
Yep NT.”
Well, maybe you need to understand the operating system. Plus, if it is having numerous problems, it is the hardware not the operating system. The fact you listed that it has problems, should raise a red flag to any decent system admin.
Blaming the Operating System for your lack of understanding is downright foolish.
Richard Weeks
I do not think so.
According to Netcraft, Windows has increased it’s marketshare greatly. Linux has as well. It seems Linux & Windows are the fastest growing operating systems. Mac OS X is also growing with the introduction of the Xserve G5.
And you know where that marketshare for windows is coming from? Legacy versions of UNIX (HP-UX, AIX, UnixWare, etc.) Solaris seems to be pretty much stablised.
I really do not see Linux cutting in on Windows at this time.
I think that one thing that people are forgetting about here is the human aspect. I work as an admin for a school, and we have dumped several netware/NT4 machines in favor of linux, but have not done so for most of our critical database servers. This is not because it can’t be done in linux, but because we, the admins, myself included, don’t know enough to be absolutely certain about setting up linux right to make completely stable. This is defintely not a limitation of linux, but rather of us the admins, there are many admins that can run linux wonderfully but many of us (coming from windows/netware) that just are sure about it. I think that this is what hurts linux the most, not the technical aspects of it.
In the second to last sentence the are should be an aren’t, sorry about that.
This logic is flawed, period. Your going to have to upgrade no matter, what. In a year, or two of extended life, you will have to migrate, so not only are you paying for premium extended support, you still have to upgrade. Companies should have switch a long time ago to:
1) 2003 Server
2) Another Operating System
Flawed how? If NT4 worked why upgrade. Will an invoice printed on a later flavor of windows bring you more money, or will it be paid faster by your customer? I don’t think so.
In the 80’s and the early 90’s software ususally got substantially better from one version to the next. It performed tasks that previoulsly had to be handled manually, so people upgraded as soon as a new version became available.
After a while sofware upgrade became almost a Pavlovian reflex. As soon as new software was shown to a sysadmin he upgraded, benefits or no benefits. Now days software does most of what we need, and the differences between versions are mostly cosmetic.
”
It opens the door for Windows, since Linux commands a whopping 3% of user desktops, and most of that is for hobbyists and advocates. Businesses in general have no desire to migrate to Linux, and anything touted as contrary to this is simply wishful thinking.
”
You do realize this article was about servers right? Linux has 25% market share on servers, approximately 9% installed base and is growing very quickly, more quickly than Windows. Too bad there wasn’t some kind of technology that forced people to actually read articles before commenting.
Samba doesn’t have a great security record.
I’m sure a worm could be written to exploit it nicely.
”
Samba doesn’t have a great security record.
I’m sure a worm could be written to exploit it nicely.
”
And windows does? I don’t think it matters whether you use a Windows server or a Samba server, the security on both sucks. I personally don’t even use Samba at home.
> Flawed how? If NT4 worked why upgrade.
That’s good in theory, but if general support is gone, you’re at the mercy of the critical next security that comes along. Successful migrations take time, so the last thing you need is to be forced into a quick migration because the next Super-Nimba comes along that renders your OS useless.
The great thing about open source is that if worse comes to worse, you or a consultant you hire can support the software even if no one else cares about it. Fortunately, there are many people who like like you. NT4 was released back in 1996, around the time that Linux kernel 2.0 was released. AFAIK, the Linux 2.0 kernel is still being supported so companies that used even this archiac version of Linux can feel comfortable that there is at least some level of support. Since Linux 2.0 had a small install base, support for Linux 2.4 and 2.6 will likely last longer.
Another great attribute of Linux (and Unix in general that runs open source) is that it’s possible to upgrade only the parts that you need. If your Linux 2.0 environment works okay but you want to take advantage of the latest version of SAMBA, you’re free to do so without changing your entire OS, because as you say, you know everything else works. In many cases, it’s possible to run two versions in parallel for an extended period of time while your performing a test migration (although you generally have to compile from source to get this flexibility). You have much less such flexibility with NT 4.0 since everything is integrated.
Please indicate to me where Samba has successfully replaced a native Windows Active Directory topology of more than, say, three servers and 25 users. I’ve played with Samba on a network of roughly 300 users and saw quickly that I would be fired for even suggesting it on a production network. Would this be a solution for a cash-strapped office of a dozen users, with a Linux admin in house? Maybe. But for a healthy enterprise with a realistic IT budget, you’ve got to be kidding. Samba is a remarkable open source project, but it is not ready for hard-banging production in a Windows AD setting.
I’ve seen schools move to Samba. Samba is much faster than Windows Directory.
havent read rest of comments so this may have already been said but… XP is not NT6.0 its NT5.1
When it’s all said and done I think NT 4 and NT 5 (Windows 2000) were the last good operating systems to ever come out of Redmond. I used NT 4 for many years, and still do when I need a reasonably secure “Windows” machine for run of the mill stuff. But I had to eventually move to NT 5 (Windows 2000) for USB support and the plug and play features of Windows 98 with the “security” of NT.
I have tried XP and absolutely hated the default desktop, it looked like Microsoft had contracted with Fisher Price to redesign their desktop, quite frankly it insulted my intelligence. I always preferred the look they “borrowed” form VMS, not to mention the fact that Windows 2000 is much faster on identical hardware. As it stands now the next primary OS I’ll go to is SuSE linux, I have been running SuSE since 6.3. Once they get more than mediocre 3D support and the ability to play DVDs out of the box it will become my primary OS instead of Windows.
…that, when Red Hat discontinues support for distros older than two years old, it will open up opportunities for Windows? This logic door apparently swings both ways, if you buy its basic premise (which I don’t)
This logic door apparently swings both ways, if you buy its basic premise (which I don’t)
Well, considering where you’re posting from, I’m not that surprised! 😉
…that, when Red Hat discontinues support for distros older than two years old, it will open up opportunities for Windows? This logic door apparently swings both ways, if you buy its basic premise (which I don’t)
The difference is that, you at least in theory could buy support elsewhere if you don’t want to upgrade your Red Hat. Who will give you bugfixes for Micrsoft discontinued code?
If you upgrade, then there is the matter of education. If you upgrade your Red Hat it will not have changed much in the next version. It will be slightly faster, slightly smaller in footprint and a couple of bugs less. Meaning your educational costs for an upgrade will be next to none.
If you upgrade a Microsoft system, much will have changed. It will have changed the way how to enter user information, how to enter the IP address, how to install… As a result you send your staff on education, that probably is much more expensive than an overprised Red Hat upgrade.
– New Users buying Windows
– Legacy UNIX Users buying Windows
– Windows Users leaving to New Unix & Linux
The 2 top ones outgrow the bottom one (in overall users). I believe that users will be forced to buy support and/or upgrade.
This will generate cash to Microsoft. It generated plenty of cash to Red Hat when they did this. But you know, red hat support and patches and upgrades could be gotten somewhere else other than red hat..
So I believe this will amplify Microsoft’s sales.. more so than how redhat’s sales were boosted.
Well, considering where you’re posting from, I’m not that surprised! 😉
Don’t shoot the messenger. I think his basic premise was sound. If you expect users to move away from Windows to Linux simply because NT4 isn’t supported, the same could be said of people moving away from unsupported Linux distros to Windows running POSIX subsystem.
The difference is that, you at least in theory could buy support elsewhere if you don’t want to upgrade your Red Hat. Who will give you bugfixes for Micrsoft discontinued code?
At what cost? In the long run, most check-signing corporate types prefer to get their support from the same vendor supplying their OS and/or hardware. I know. I work with them every day. Obsolescence is factored into their long-range budgets. Nobody is going to give you money to spend today unless they know how you’re planning on retiring the assets that you already have.
At this stage, most of the troublesome issues have been worked out of NT4. It’s a stable platform. And, even if that weren’t the case and some security problem were found, it would probably still be cheaper to stick your NT4 boxes behind a firewall in a closet — and do filtering to prevent them from being exposed to the vast majority of port vulnerabilities.
If you upgrade a Microsoft system, much will have changed. It will have changed the way how to enter user information, how to enter the IP address, how to install… As a result you send your staff on education, that probably is much more expensive than an overprised Red Hat upgrade.
Not necessarily. First, many people don’t realize this but Windows applications are but one subsystem that will run atop the Windows kernel. Windows also contains a POSIX-compliant subsystem that will run Unix apps that have been compiled for x86. In some cases, you may not even need to recompile if you’re already running x86 apps. Second, Microsoft corporate license agreements are very competitive with Red Hat. More so than you realize. Sure, the OSS bigots think that you should be able to install ‘nix for free — but that’s just not realistic or reasonable — Microsoft’s real competition in the Linux space is Red Hat. And, on that score, they are competitive.
– New Users buying Windows
Probably true. But the interesting thing, is how and why they buy it. Only about 70% of the windows users have upgraded to XP after 3 years of its existance. This is about the rate of hardware replacement in many companies.
Meaning that a large part of the users using XP today do so as they got it with new hardware. This is a sign of a stagnated market. Nobody is buying it because they really need it, or because it is an amazingly good peace of technology. They buy it because they have to.
As most needs of the users are fullfilled with the current version of windows or even the now end of lifed NT4, Microsoft will have to dream up new needs for their users.
This takes a lot of money both in development and marketing. It is also risky if you change anything on a product that satisfy most users you may piss some of them off if you do it wrong. This costs get worse the closer you get to the perfect OS.
The same naturally holds true for Linux, but Linux have more to gain and a smaller userbase to keep happy. Imaginge Microsoft is successful and really invents something useful that makes 1% of the Linux users to switch to Windows. This would probably be hidden as a rounding error Microsoft sales figures. On the other hand if Microsoft does something that makes 1% of their users to switch to Linux. The Linux market value will expand significantly.
Given that Linux is slightly behind both Apple and Microsoft in desktop design and functionality for the average joe. The Linux developer can watch and stay clear mistakes made by Apple and Microsoft and improving their system at a lower cost.
At some point, Microsoft will be forced to lower their prices or even give away their software for free. to keep a valuable support market. At the same time they need to increase their efforts in the R&D department.
This will not be good news for Microsoft share holders. Not that Microsoft is going to disappear any time soon, but they have a very high risk of getting a significanly lower market share than today. This is what happened to European and US car industry 30-40 years ago, when cheap but sufficiently good japanese cars fludded the market.
For one, WinXP’s interface is completely skinnable.
Two, you can go back to the standard interface.
Three, perhaps you need your intelligence insulted if you find a freaking UI insulting to your intelligence.
Also, all you need is some tweaks here and there and XP will in fact run at least the same as 2000 on the same hardware. Not to mention XP doesn’t have the various different hangups in certain areas that 2000 had.
Not siding with Windows here, just quit make dumb remarks about things that just simply are not true, and then go on later to say you are going to make your primary OS Linux.
This happens at least 3 or 4 times with every Windows or Linux news article… same exact format.
I believe http://www.reactos.com is more promising alternative than linux as companies look more for compatibiltly and small learning curve rather than cost of software. They want their current software to just work rather develope new software for a new platform. Linux is definalty an alternative for server’s but not workstations.
This is the same reason why, when go to the bank or travel agent you see them still using a 20 year old telnet system.
Reactos isnt ready for mass use yet but in maybe 2 years with the right investment, you could get a dell pc with a optional windows compatiable os preinstalled rather than ms windows.
Don’t shoot the messenger.
I didn’t shoot anyone, just remarked that the guy was posting from Microsoft, and that he seemed to have a pro-MS bias. I just thought it was funny.
I think his basic premise was sound. If you expect users to move away from Windows to Linux simply because NT4 isn’t supported, the same could be said of people moving away from unsupported Linux distros to Windows running POSIX subsystem.
Okay, let’s imagine that this is true (even though the analogies are faulty, because with Linux you’re not stuck with a single vendor, and you don’t have to upgrade all of your system at once).
This would likely happen in similar proportions – i.e. one out of ten NT4 server switches to Linux, and one out of ten no-longer-supported Linux server switches to some newer version of Windows.
Even with that scenario, lots more servers would pass into the Linux side, because the installed base of NT4 servers is larger than that of a no-longer-supported Linux.
So even if making a straight analogy was correct (and it isn’t), Linux would still come out ahead.
I didn’t shoot anyone, just remarked that the guy was posting from Microsoft, and that he seemed to have a pro-MS bias. I just thought it was funny.
I don’t think bias had anything to do with his post. He was merely skeptical of the leap in logic that ceasing support for NT4 means Linux gains. And I happen to agree with him: Shops that run Windows are more likely to upgrade because the move to an upgrade is a helluva lot easier than rewriting all of your Windows apps to work on Linux.
Even with that scenario, lots more servers would pass into the Linux side, because the installed base of NT4 servers is larger than that of a no-longer-supported Linux.
As I just pointed out, that requires you to accept that companies are willing to rewrite all of their Windows applications to run under Linux. Now, it probably isn’t a big deal if you’re only moving across file/print servers or rudimentary web servers within a small network. But doing so within a large network is a completely different matter. Large corporations don’t run Samba. It isn’t suitable for deployment in corporate America. No matter what anybody here on OS News tells you, Samba doesn’t scale to the same level as ActiveDirectory. So, any move from Windows to Linux requires not just moving a few servers but a massive rethinking of corporate network topology. I will grant you that IT guys are looking at Linux but, in my experience, they’re only doing so in the most narrow of circumstances — and usually only in shops which were Unix to begin with.
So even if making a straight analogy was correct (and it isn’t), Linux would still come out ahead.
Linux is taking market share away from Unix (Solaris, AIX, etc), not Windows. There are simply too many obstacles to moving corporations from Windows to Linux — and the numbers that I’ve seen prove that every time.
The difference is that, you at least in theory could buy support elsewhere if you don’t want to upgrade your Red Hat.
You mean, in practice, one can find affordable support for any version of any Linux distro ever provided? Or you meant it in theory, like “in theory, Open Source software does not have bugs?”
Who will give you bugfixes for Micrsoft discontinued code?
You, at least in theory, should be able to make Microsoft provide you extended support, assuming you are willing to pay for it enough to make Microsoft interested.
Also, in theory, you could buy Windows NT from Microsoft and support it itself.
If you upgrade your Red Hat it will not have changed much in the next version.
That statement is not supported by the comparison between Red Hat Enterprise Server version 2.1 and version 3.0.
It will be slightly faster, slightly smaller in footprint and a couple of bugs less.
That statement is not supported by the comparison between Red Hat Enterprise Server version 2.1 and version 3.0.
If you upgrade a Microsoft system, much will have changed.
True.
As a result you send your staff on education, that probably is much more expensive than an overprised Red Hat upgrade.
If you have Windows NT to retire and consider replacing it with Linux or UNIX, the price of re-education of your staff for the new version of Windows can not be higher than for the version of Linux or UNIX.
It is obvious that in many places NT retirement will result in urgent need to fill Microsoft cash registers, because upgrade is now imminent. People will stay with Windows, and upgrade.
In few places where Linux or UNIX trained admin was forced against his will to support Windows NT there will be transition to Linux or UNIX. I am sure these transitions will be heralded in OSNews and other places like “Microsoft dies as we speak.” I am also sure that next FY Microsoft will publish healthy profits, again.
the numbers that I’ve seen prove that every time.
Please provide these numbers, or I shall be forced to believe that you made them up. Thanks.
…replacing older servers, especially mail, file and print servers with Linux makes sense, especially considering the costs of moving to Win2K or Win2K3 Server.
Please provide numbers that show high costs of moving from Win NT to Win2K, or I shall be forced to believe that you made it up. Thanks.
Please stop being such a MS cheerleader. Thanks.
When you’re through playing the smar@ss, you can look up the price of a Win2K3 Server license yourself.
http://www.theage.com.au/news/Breaking/TCO-study-Linux-wins-again/2…
http://techupdate.zdnet.com/techupdate/stories/main/0,14179,2907876…
http://www.ameinfo.com/news/Detailed/48918.html
Of course you’ll find studies that point to the contrary…however, these studies were funded by Microsoft and their methodology has been questioned.
But don’t let facts stand in the way of good FUD…