Yankee Group Survey Says Windows, Linux TCO Equal

prostoalex writes "A new survey by Yankee Group analyst Laura DiDio shows Windows and Linux are viewed as equal by U.S. businesses. In the eternal OS wars, '88 percent of respondents said that the quality, performance and reliability of Windows was equal to or better than Linux.' Companies were also asked to rank the operating systems on security. On a scale of 1 to 10 'companies rated Microsoft's security at 7.6, double the rating in a similar survey conducted last year. Linux's rating was mostly the same at 8.3.' Conclusion? 'DiDio said that most companies -- whether large or small -- rarely take the huge step of replacing one operating system with another. Instead, they usually add a mix of Windows and Linux server software to expand functionality.' Microsoft used last year's Yankee Group survey results in their Get the facts campaign."

Comments

What are they using?

[fembots]

The survey needs to take into account what OS the respondents are currently using, that's the single most important factor.

You don't use an OS that you don't like, and if that's not true (e.g. you're forced to use a pre-installed OS), then you probably wouldn't know any better alternative if you've been using only one OS.

If a Linux-only user said Windows is better, or vice versa, what does that mean? How does he come to this conclusion? The most credible answers should be from Multi-OS users.

I'm not saying this study is inaccurate, but there are simly too many things to consider, and this may well lead to a simple conclusion - software choice is more on personal preference than anything else.

Re:What are they using?

[Cruithne]

Another factor, the study says 88% said windows was equal or better - but how many said it was better versus equal? And were they given an option between the two?

More importantly, how many people said linux was better versus people that said windows was better...

This post seems suspiciously void of those kind of common-sense numbers.

Re:What are they using?

[Monkelectric]

Yep, this is fluff journalism/science, whatever they claim it is. What the study basically says is- peoples confidence in their OS is about the same no matter what OS they use.

Re:What are they using?

[Zeinfeld]

Then again, this only works with people who know what they hell they are doing.

Which goes the same for pretty much any O/S. If you have a pinhead they will configure the machine insecurely.

No matter what I would never recommend Windows as a internet-facing server. I run a Windows 2003 server here in my home but it is just to learn it and host a small site with little traffic.

You mean even if the figures say that Windows is more secure you will never choose it? Or are you only referring to the current release?

Whatever, I think that Linux advocates should take a lesson from history, it is really hard to maintain an O/S distinction in the security area. The only reason Linux is any better is that UNIX machines have been Internet connected by default for about 15 years while with windows its only about 8. Read the CERT advisories from the 90s, they are almost all reports of UNIX vulnerabilities.

UNIX got cleaned up, Windows will be cleaned up. Back in the 90s UNIX was a byword for insecurity, people still used SUID scripts and shadow passwords were only used by a minority.

What is more interesting here is the derrivative. The perception of Windows is improving rapidly, the perception of Linux is pretty static. I don't see a heck of a lot of new security action going on in the Linux world. There is a heck of a lot going on in the Windows world.

Re:What are they using?

[AstroDrabb]

Why don't we look at this rationally. The Yankee Group doesn't do "studies" for free. The Yankee Group are a for-profit company. So basically someone paid the Yankee Group to do this "study".

Now, who could it be? Could it be Red Hat, SuSE, IBM or some other pro-Linux company? I have serious doubts about that. What about Microsoft? Well, MS has certainly paid for other "studies" to be done in the past. So I don't think there would be any major reason to not count MS in on this "study". Basically we just need to find out _who_ paid for this "study" to really see where the bias lays.

I remember last year I had a phone call from some unknown company that was doing a "study" about MS. I was asked how I felt about MS as a company. How I felt about the products put out by MS and if I "trusted" MS. As soon as I answered that I "did not trust MS as a company", I was told my "interview" was over and "thank you for your time". So it seems as soon as one of these companies get a negative response about the company that are footing the bill, the interview dies.

Does anyone know who _paid_ for this "study"?

Re:What are they using?

[Coryoth]

What is more interesting here is the derrivative. The perception of Windows is improving rapidly, the perception of Linux is pretty static. I don't see a heck of a lot of new security action going on in the Linux world. There is a heck of a lot going on in the Windows world.

If you don't see much happening with regard to security in the Linux and UNIX world, then you simply aren't really paying enough attention. UNIX is getting fitted with a new, significant, very powerful, security architecture. The difference is aking the the difference between a single-user and a multi-user os. It's coming to Linux via SELinux (though there are other implementations of the basic concept such as RSBAC). The BSDs have it in TrustedBSD, and the new (open source) Solaris 10 has it (Trusted Solaris has been integrated into the main branch). Does Windows have anything even close anywhere on the horizon? No.

Sure, for all of these systems the security architecture is new, and by default it is often either off, or in a relatively minimal configuration. The point is that it is already developed, and implemented, and in the respective kernels. From here it's a matter of educating users and developers, getting better application support allowing for stronger/stricter policies by default, and building better tools to configure and administer the system. For Windows any level of Mandatory Access Controls is still in the hazy future, to be implemented, at best, in the release after Longhorn. By the time Windows secures all its holes UNIX may well have moved a quantum leap ahead.

Jedidiah.

Re:What are they using?

[FreeBSD evangelist]

What is more interesting here is the derrivative. The perception of Windows is improving rapidly, the perception of Linux is pretty static. I don't see a heck of a lot of new security action going on in the Linux world. There is a heck of a lot going on in the Windows world.

There's a lot of "security action" going on in Windows because there's a lot that needs to be done. If they were to get close to the security of the average FreeBSD box (like, never) that activity would slow down too.

Re:What are they using?

[Zeinfeld]

For Windows any level of Mandatory Access Controls is still in the hazy future, to be implemented, at best, in the release after Longhorn. By the time Windows secures all its holes UNIX may well have moved a quantum leap ahead.

You obviously do not know what Mandatory Access Controls are. Butler Lampson certainly does since he invented the idea. so does David Cutler since he implemented them in VMS long before they arrived in any Unix variant. They have both worked for Microsoft for over a decade.

Windows NT had MACs built in from day one. They are not quite the same as the VMS implementation - and for very good reason, the VMS implementation of ACLs was too complex for most people to grasp, particularly when you got into the propagation rules. But they certainly are there and are built into the O/S at a much more fundamental level than they are in Linux.

The problem with Windows security has absolutely nothing to do with lack of security features. The problem is the exact reverse, the problem is too many damn features and applications that can't make use of them.

Re:What are they using?

[setagllib]

When Windows can use algorithms other than DES and 3DES for IPSec, then we'll talk. And where are the static keys? It wouldn't be so bad if its twisted way of setting up key exchanges was workable, but it's only barely so. And IPSec, if you'd ever used it properly, you would know to be a VERY valuable security and privacy feature, yet in Windows it's on barely workable levels. Linux got a not-too-bad IPSec in kernel 2.6, and while it's still a lot flakier than any BSD (like having to use -m tunnel), it actually works and can use algorithms that aren't yet obsolete. And it can use them properly. Static keys, IKE, tunnel, transport, whatever, it works. Windows is excruciating and crippled in this respect, and it's a huge lacking in its security capabilities.

That's why many *nix's don't need to make huge progress: they're already on the bleeding edge. With Windows so far behind in respects like these, it need all the improvement it can get.

DiDio = Shill

[NiceGeek]

Did this bimbo ever have any credibility?

Re:DiDio = Shill

[gnasher719]

Does she have credibility? About as much as Ken Brown ("A swedish student named Linux Torvald copied Linus from Minux which his professor Tannenbaum copied from Unix"), Rob PretEnderle (the one with the Ferrari Notebook that makes Vroom Vroom noises) and Maureen O' Gara (Linux is completely stolen from SCO) together. Minus infinity + Minus infinity + Minus infinity = Minus Infinity.

Opinion Based

[Anonymous Coward]

'88 percent of respondents said that the quality, performance and reliability of Windows was equal to or better than Linux.' Companies were also asked to rank the operating systems on security. On a scale of 1 to 10 'companies rated Microsoft's security at 7.6, double the rating in a similar survey conducted last year. Linux's rating was mostly the same at 8.3.'

Notice, it doesn't say security professionals for security, it doesn't say economists for TCO, it says companies. I'm sorry, but the first thing to enter my mind in this situation is a "Pointy Haired Boss" filling these things out. It's basically an opinion survey, pointless in anything but spreading FUD.

Windows XP and 2000 arn't bad, just expensive

[Anonymous Coward]

Actually Windows XP and 2000 are both pretty reliable products. I haven't had any problems with XP/2000 reliability (unlike Windows95/98 crash randomly).

As for performance and driver support, Linux wins on performance but windows wins on support.

I'd say they are equal if you discount price, which this survey did.

Re:Windows XP and 2000 arn't bad, just expensive

[rikkards]

I think that is part of the problem. Linux is not that hard to figure out. The documentation has improved tremendously in the last 5 years.

Most Linux people like to think that Windows only shops don't want to go to Linux for fear of the unknown as well as inexperience. I don't think it is that at all. I think it is laziness. If it is working fine, you have proper usage policies, all the workstations and servers are patched properly, antivirus updated regularly. Why fix what isn't broken?

Back in the 95 and early NT days, there was a definite window but Linux wasn't ready. Now with the stability of 2000, 2003 and XP and Microsoft appearing to actually be moving towards a more stable platform as has been displayed with the default settings in Windows 2003 and Exchange 2003, the road in for Linux has become a little more difficult.

Personally I think Linux has definitely got to the point where it is a viable workstation OS as well as a more than suitable server but Microsoft has adapted.

Will Microsoft continue to have dominance? Probably not, even Billy has said that but when that will be? Not likely anytime soon.

From TFA...

[Suhas]

.... "Server operating systems are largely commoditized," DiDio said, adding that many companies were not tracking their operating costs closely enough to base their decisions on total cost of ownership, or TCO, the main cost metric when comparing Linux and Windows.

If they are not tracking operating costs, then that means they are only tracking the initial cost of acquisition, which for Linux, is ..umm...zero. So how exactly is the windows TCO equal to that of Linux? What a fucking troll of an article.

Irrelevant

[pegasustonans]

Yeah, so a bunch of people are asked for their opinion about which OS is better. How is this even remotely relevant to anyone other than social scientists and marketers?

Re:No comment...

[wasted]

From the Article: "Server operating systems are largely commoditized," DiDio said, adding that many companies were not tracking their operating costs closely enough to base their decisions on total cost of ownership, or TCO, the main cost metric when comparing Linux and Windows.

So, they ask the bosses "What is the TCO for Windows-based servers?"
"I don't know"

Then, they ask the bosses "What is the TCO for Linux-based servers?"
"I don't know"

Since "I don't know" equals "I don't know", the conclusion is that the operating systems have equal TCOs, at least in the eyes of the business managers.

That's the point

[jerometremblay]

If your intent is to measure their PERCEPTION, this is exactly what you need to do.

Instead of taking it as FUD and discarding it, consider it as a TODO list to increase your favorite OS acceptance (whatever that may be).

Re:That's the point

[menace3society]

If you read the article, it is not being portrayed as perception.

"Study shows Microsoft and Linux Neck-and-Neck"
Not "Perceived as Roughly Equivalent"

"Most U.S. businesses say there is very little difference between the cost of maintaining a Windows versus a Linux-based corporate computing environment."
Many of whom, the article goes on to say, don't really bother to keep track of the costs in the first place.

"In the independent study, 88 percent of respondents said that the quality, performance and reliability of Windows was equal to or better than Linux."
The article does not say that the quality, performance, and reliability seems to be equal or better, but that it is.

It's also worth noting that it never says a) whether the company has actually used both OSs recently, and who precisely is doing the response. If I were the Director of the IT department or some such at a corporation, the last thing I'd do is admit to anyone, even if they said they were taking a survey (they might be a spy from President's office) that the system we had in place was anything less than the best possible. Even if I knew it for a fact to be otherwise.

Re:That's the point

[endofoctober]

But in this case, it's /not/ a question of what one needs to do - it's asking the proper question to the person who is /qualified/ to answer it. That would be the difference between a good survey ('...we asked network security experts about network security issues and got the following results...') versus creating a good marketing tool ('...we asked marketing experts about network security issues, and got the following results...').

In the first instance, you've provided real information that decisionmakers can use when choosing an OS for their business. In the second instance you've provided tainted fluff that uninformed bean counters will /perceive/ as real information. Sounds like FUD to me.

Not to say that the Linux community can't learn from it, but it's important to recognize that groups like this aren't assessing quality. I don't know about you, but when I have to make decisions about software, I require more than just knowing how someone 'feels' about it. Teasing the facts from these so-called 'objective' studies lately is /not/ worth my time.

Not FUD!

[CaymanIslandCarpedie]

I don't see how this survey can be considered FUD. They aren't saying anything either is better or worse than the other. They simply relay feelings of their respondants.

The whole point of this of course isn't to compare the platforms or make a suggestion on which is better, it just conveys the feelings of their respondants.

Should this be used as a basis for a decision for what to use? Of course not!!! Is this an interesting insight into the current thinking of corporate IT departments? Yes.

It isn't FUD and isn't pointless, but if you take any of this as FACT, thats your mistake. This is simply an interesting look at current thinking. If this thinking is correct or not isn't the point. Its like saying a poll finding 80% of people are against the war in Iraq is FUD. That poll wouldn't wouldn't mean we should or shouldn't be there (as the respondants may not really be qualified to know), it would just give an interesting view of what people are thinking.

Read this article as such.

Re:Not FUD!

[nmos]

I don't see how this survey can be considered FUD. They aren't saying anything either is better or worse than the other. They simply relay feelings of their respondants.

Maybe and maybe not. From the article we really don't know who was surveyed. Given Didio's history I wouldn't put it past her to have selected the people/companies being questioned to give whatever results she's being paid to find this week.

Enough of this TCO crap

[js3]

The only we learned from TCO studies is they don't know what the TCO is.

Re:No comment...

[Brent_Edwards]

Very true. Also most organizations are using A LOT of Windows especially on the desktop, so for them to admit that Linux is better (more secure, cheaper, whatever) would be admitting they are doing it wrong... Thus not a common trait of managers.

We're still reporting DiDio?

[lakeland]

Laura Didio is a paid shill, who writes whatever will give her the most hits, regardless of factual content. I've read tabloids with a better grasp of the truth than her.

According to the article, she's now claiming to have done yet another study which no doubt will get debunked within a couple hours but still cited by microsoft (sans the debunking) months later.

So, a simple question for everyone? Why bother debunking it? Anybody with more than half a brain already knows Didio is a paid liar, so she's not going to care if you drive a truck through her arguments. The other people on ./ already know she's a shill, they won't care. And the PHBs reading the article via MS's links won't get to read your debunking. Why waste your time on her?

Humor the Thought

[Anonymous Coward]

Suppose they are equal for now. Then what would happen if, say, one of them destroys the other in the next 30 years?

Case 1: Linux ends up suiting everyone's needs "for free", so (very) many people leave Microsoft. Cost of ownership (inflation-adjusted): equal or less than today.

Case 2: Palladium (I guess it's called NGSCB now) becomes reality; Linux, Apple, and all other competition is destroyed. Cost of ownership: 10, 100, or 1000 times more, or maybe just whatever your business can afford.

Investment: even if the two actually are equal today, and even if TCO is the only factor I consider (i.e. setting aside my enjoyment of Linux), I still have my preference.

(Did I miss something? It seems like people don't talk about Palladium on /. anymore...)

Deployment, etc

[paugq]

Some important features Linux has been lacking for a long long time has been ease to deploy software & patches from a central location and a highly-integrated and easy-to-use directory service.

Sure, you could use scripts and LDAP, but they suck. Those are time-consuming and limited solutions.

With Novell entering the Linux market (I'd rather say Novell is betting everything to Linux), these is changing: eDirectory, ZenWorks (6.6 works great, I can't wait for 7.0), etc are superior tools and services.

A lot of companies (mine included) are starting to use SuSe because of the awesome integration of Novell tools with SuSe. We are even deploying it to our clients. Is this the return of Netware, in shape of a Netwared Linux?

Re:I wonder if they surveyed a company

[korielgraculus]

Assuming that the server is doing something crucial to the company, you do realise that you have successfully managed to give a 27,000 person company a single point of failure (you) then? Congratulations!

Question for Accountants

[SunFan]

Are companies in denial about the costs of lost productivity due to Windows-based outages? Labor accounting sort of sinks that information away, IMO, and is harder for businesses to quantify as an IT-related expense.

IIRC, Bill Gates himself said a while ago that companies lose two weeks per year per employee due to Windows' downtime. Having worked on a mixed UNIX and Windows network, this seems about right to me. The Sun's were rebooted a couple times a year for maintenance, and the Windows PCs got rebooted every day. Even on the Windows side of the place, they conceded to Solaris for all the infrastructure where they could, such as e-mail routing, scanning, DNS, etc. Still, that didn't help the Windows server "cluster" that served many of the user accounts.

Laura Didio is a con artist, liar and..

[I_redwolf]

general jokster. In all honesty, who really cares what she has to say? Or for that matter, anything else coming from the Yankee Group in general. It's all joke, smoke and mirrors stuff.

You know what would be a good idea. A bunch of geeks getting together with a bunch of researchers in their respective fields. Creating honest, non-biased "this is the way it is" anaylsis and reports on TCO/Software/Hardware/etc. Sort of like Consumer Reports(tm) but with more detail and analysis of specific topics.

Laura Didio? Whatever.. If I want to run my business or anyone elses into the ground; I'll take her advice. Until then, I won't even pretend to RTFA.

So all OSes are built exaclty the same?

[Infonaut]

The only reason Linux is any better is that UNIX machines have been Internet connected by default for about 15 years while with windows its only about 8.

This is the same argument as the old saw about how simply because Windows is the dominant consumer operating system it is the target of more malware. It ignores the fact that operating systems are not all built in the same fashion. For example, what about pre-OS X versions of the Macintosh? What about OpenBSD or Bastille Linux?

These discussions about OS security tend to ignore the fact that the *NIX distro or Windows version you're using can significantly impact security. Just as all OSes are not the same in terms of usability, I think it's a gross simplification to say that they're pretty much equal in security.

Re:Good as each other

[bersl2]

I did read one stat one where which said somrhthing like 66% of linux workstations use a cracked copy of windows.

Non-sequitur.

I find windows easy to use and install, call me a point a click nub, but if i can do alot without having to remember commands, it works in my favor.

Appeal to tradition.

What happens when say, a linux company starts comming as big as Microsoft, are they being as bad

Appeal to fear.

and lets be honest, alot of windows problems are C++ buffer over runs which as it fault of the languge, not the programming,

False premise.

also being the most widly used OS its bound to have more holes discoverd as more people (like 12 year olds now can get on net and learn how to crack programs and find security holes) are doing it on windows systems.

Red herring.

Also another side and my second closing (lol) is that Microsoft does alot of good. Gates give the most to charity (i think he gives the most in the world),

Appeal to wealth and honor by association.

there traninig scheme is good with alot of people doing, I for 1 am starting my MCSE in september, and have a big reasearch department,

Untenable appeal to authority.

look a mobile computing, wireless, the whole wireless home idear with windows media centre.

Meaningless statement.

I wonder how much of the advanced we have today, would be here without MS.

Historian's fallacy and/or hindsight.

Got any more fallacious thinking?

Re:MOD PARENT UP!! :)

[ergo98]

Security will always be the winning point of Linux - that's the matter of system design.

Wow, where do you get this stuff from?

What is the amazing system design element of Linux that yields such remarkable security? The reality, you know here in the real world, is that there is nothing special about Linux. In fact the architecture of the NT line of Windows operating system has more embedded and pervasive security functionality. Security is far more of a system design element of NT and greater than it ever has been with Linux, Microsoft just has a habit of grabbing defeat from the hands of victory.

Of course then there is Windows 2003 - I know you're probably arguing based upon the same old tired rhetoric from the '95 days, however 2003 is a rock solid operating system. Apart from being tremendously stable, it is extremely secure by default. SP1, released a few days ago, enhances and improves on the security that was even there.

If you really think security is the big winning point of Linux, then you lost the game two years ago.

"will be"

[Anonymous Coward]

Windows will be cleaned up

This pretty much invalidates what you said about Linux needing a lesson from history somehow having to do something about Windows.

Re:DiDio. Why am I not surprised?

[Anonymous Coward]

Now let's see the Linux cost. I'm going to get a maximum of $50 off those Dell machines for chosing Home instead of Pro, I can't 'not have' Windows on it. So that makes it $299/machine. Let's say the cost of installing Linux on each of these is $0

It was $79 when I checked. But with the right login scripts, it seems you can actually make Home play nice on a Windows network. As for not getting $50 off for not choosing XP home, you can blame MS for that. In order to get the $50 price, OEMs pay for a license for every system they sell, whether or not they actually have Windows installed.

Now let's look at my army of typists. None of them know how to use Linux/GNOME, OpenOffice or Evolution. So I train them. I hire a training guy to come in for a day to give them a crash course on how to use Linux, and he charges me $200. However, I've got to pay my typists anyway, $100/ea for the day. So that's $100x10 + $200 = $1200

If not OpenOffice, would they be using MS Office? Normally $450 for the pro edition, but on Froogle you can get it for $350, or small business licenses for $250, or _academic_ licenses for $150. You can forget about a Windows server with Exchange for all the Outlook users. That'll be another $3000 at least. If you do go OSS all the way for apps, then there's really little or training cost to get them using Linux, except with Windows they're more likely to screw up the system. You'll probably be running Linux on the servers anyway, for Samba file sharing.

A lot of ERP software only runs on Windows. In our case, that's the only thing that seems to be keeping us with Windows. But at the price, operating system cost isn't really a factor. I just like the functionality of Linux.

Laura Dildo is a Microsoft lapdog

[Anonymous Coward]

Based on Laura Dildo's reputation, I'd take any article written by her as garbage until proven otherwise.

Re:Brought to you by...

[abandonment]

exactly, why does slashdot even give didiot the time of day any more - she's 'proven' her 'independence' so many times in the past year or so with the sco/linux debacle, let alone the fud that she spouts consistently.

it's obvious she's a microsoft employee, and the yankee group have destroyed any credibility they 'might' have had by continuing to employ her.

non-story, complete fud

Re:here's the survey

[killjoe]

DiDio is a well known shill. The fact that she is claiming that the TCOs are equal means MS has given up trying to convince people that linux is more expensive.

How long is a piece of string?

[cardpuncher]

What's a "corporate computing environment"?

The number of Linux *desktops* in "corporate" environments is vanishingly small, so I don't know how any responses could be statistically significant.

Linux *servers* will be more in evidence, but the role they play will vary significantly. In smaller companies, they may well be used for file and print services. In larger companies, they probably won't because AD makes much more sense in that environment. Mostly, Linux servers will be web servers and the TCO will depend almost entirely on the type of application you're trying to build and the development and support time using the chosen tools (J2EE/PHP/CGI/Perl/bash...).

So for *most* Linux deployments, it's not a question of Windows vs Linux, but the TCO of Visual Studio/SQLServer/IIS vs Websphere or some FOSS solution.

Which might explain why the survey "reveals" so little...

Re:DiDio. Why am I not surprised?

[clare-ents]

Haven't you forgotten to buy Word and Outlook ? 10xWord + 10xOutlook might swing your calculation the other way.

Now lets add on the fact you might need a server to store the files with 10 CALS. Then you might need a copy of Exchange so that email can be managed. Suddenly your costs are rocketing in both technician time & software costs.

Do any TCO studies make sense?

[walterbyrd]

Frankly, I don't see how TCO studies make any sense. To me, it seems that it's completely situational.

Things that actually drive decisions in business

[agtwilight]

1. Value, the ability to make or protect earnings
2. Enterprise quality - any idiot can make a webserver, but making 500 of them or rolling out the app to 10,000 people is what you need
3. Vendor Stability - he who has the most money in the bank wins kthxbai
4. Audit/lawyering - SOX needs to die, but right now its important

Notice in here there is nothing about Linux, Windows, Opernsource, blah blah blah. If the business needs an app that they are going to make a buck on that needs SCO - YOU BUY SCO. If its runs on MSSQL YOU BUY MS. If it runs on Oracle - YOU BUY ORACLE...you have to have controls and process around all systems - just because you are using apache doesnt forgive the need to have security, patching, monitoring, and compliance.

I know I am talking to a brick wall..but I thought I would try - my cynicisim hasn't kicked in yet this morning.

twi