Slashdot Mirror
A New Look at Linux vs. Windows TCO
An anonymous reader writes "Laura DiDio, research fellow at the Yankee Group, published a column this morning in which she discusses key findings from a new survey on the total cost of ownership of Windows vs. Linux. DiDio often is written off by the Linux camp as being pro-Microsoft, but she offers excellent, neutral advice for any IT department considering a fundamental systems switch: 'If you do not know what is on your network, if you cannot at least estimate the hourly, monthly or yearly cost of downtime, if you do not know how long it takes to recover from a security outage, if you cannot answer questions about the extent of your company's license compliance, then you cannot truly evaluate whether Linux, Windows or Unix is right for your business. Chances are, if you cannot answer most or all of those questions, it does not matter what operating system you have because ignorance of the core TCO tenets means that your business is not getting the most out of its networks.' "
"but she offers excellent, neutral advice"
The person who wrote this has not been reading her other work. Neutral isnt even on the map.
HTTP/1.1 400
DiDio often is written off by the Linux camp as being pro-Microsoft
Um, well, yes, with good reason. When someone performs public relations work for a large corporation on a long-term basis, one then needs to recognize that further publications by this person should be recognized to certainly be further public relations work for the same corporation. As a side note, one might also consider any TCO studies published by Red Hat Corporation to be somewhat biased.
100% correct.
Generally Windows servers need more attention.
100% back of the envelope, likely wishful thinking, unsubstantiated guess.
There were no real "key findings" in the article. Is it really worthy of Slashdot to mention a survey whose outcome was "well, it depends"? Never mind that, was it really worth writing the original article?
Either the author had nothing better to write about, or they felt like inflating their ego a little by assuming that people in business are pretty thick and need to be told the blindingly obvious.
Like car accidents, most hardware problems are due to driver error.
Criticism of a previous Didio study related to Linux TCO...
... does this sound professional to you?
Laura Didio whines that slashdot does not like her...
Quote from wikipedia by Didio: "The thing about Linux is, you can talk about a free, open operating system all you want, but you can't take that idea of free and open and put it into a capitalist system and maintain it as though it is some kind of hippie commune or ashram, because if you can do it like that, at that point I'm like, 'Pass the hookah please!'". And on another occasion she followed up: "I'm all for open source, and competition serves everyone's interest. But if Linux is really to take its place alongside Windows... then the vendors in this space cannot act like a bunch of hippies in a '60s commune or ashram. There really is no such thing as a free lunch."
Collection of Laura Didio quotes, compiled by Groklaw, on the subject of Didio insisting that she'd seen the linux-sco code comparison and it contained clearly copied code. This was posted, mind you, on the day that the judge ordered SCO to either provide the evidence linux contained copied sco code or drop those accusations from their lawsuit, and SCO resonded by dropping those accusations. In other words, the evidence never existed. There was never any copied code and SCO has as good as admitted so in court by their refusal to specify what the copied code was when the court ordered them to.
Laura Didio has made it clear she is not someone worth giving the slightest bit of attention or media press to. She has in the past shown a complete lack of any idea of rigor in compiling or presenting a study, as well as a willingness to both mislead and outright lie. This is not someone who knows how to do journalism, or how to do an informed study. This is someone who knows how to do one thing and one thing only, and that is shill for Microsoft when Microsoft pays them to. Right now she is shilling for Microsoft. Microsoft press releases released from Microsoft itself may occasionally contain good points or true statements, such as "Microsoft is a company located in Redmond". However, even when this is the case they don't get printed on the front page. Why should Microsoft press releases released through the front of Didio be treated any different?
For example: we have 3 servers (all Windows) in my company. Do we use them optimally? Probably, since we cannot replace them with any other software (to my great sorrow). Do we know how much each server costs us? No, and we will never be able to calculate that. Niether we care, as long as they do their job.
Why should I measure something which is hardly measurable just to be able to say that I use something in right way?
This is typical article where highly payed "analysts" try to spit obvious things in order to sound smart. As usual, they spit crap, but being so "well" informed about the subject, they even don't notice that.
No sig today.
Ms DiDio's report has been here since November 2004.
Has anyone succeeded in resolving their problem of which OS to use based on this insightful article?
I wasn't sure why Dido was reviewing linux...
I mean, she sings so well, why the sudden career change?
Sigs are for the weak.
Zotob Worm Hits CNN and Goes Global
Windows 2000 machines are infected. Linux and Unix aren't.
I clearly see what that woman means.
Seriously, anyone who doesn't even have a clue how much their systems are costing them are only ever going to make it worse - and more expensive - by randomly moving to another platform because someone's friend told them it was free. Anyone making such kneejerk reactions without the figures also has a high probability of being a moron that makes anything they touch turn to sh*t anyway, making it doubly worse.
In essence - if you don't even know what you're doing with what you have, don't make it worse by changing it to something else that's so different.
I don't know if MS should be pointing this out in their marketing though - one one hand it's inherently true and a great way for them to fight the leaks to OSS - but it's also pointing out to the majority of customers that they're stupid.
Here's what she said a while ago... "But if Linux is really to take its place alongside Windows... then the vendors in this space cannot act like a bunch of hippies in a '60s commune or ashram"
And now... "Neither server system will consume the other. Both will coexist..."
Is it her point that Microsoft has in recent days started acting like a bunch of hippies in an ashram?
Or she has acknowledged that Linux is not about free lunch or beer, but true freedom for the customer, and hence compatible with capitalism?
Looks like after her FUD in the SCO affair fell flat, she's presnting more scientific FUD in doing a TCO comparison... why should she choose to study the methods of hippies, outcasts ans communists?
-
If you keep throwing chairs, one day you'll break windows....
That doesn't sound neutral to me, it sounds more like:
"We know that you, and everyone else on the planet, can see that Linux is cheaper so I'll try to convince you that you don't know enough to judge the TCO in the hope that you'll then take the easy option of sticking with M$"
Was there a specific point to that article? It seems she tried to avoid any nitty-gritty details after her attempt at building a point that Windows isn't threatened by Linux, and vice versa. After that opening I expected something with more depth.
Instead, she threw in mindless details of the very obvious and took on a condescending tone toward businesses in general; of course, she did toss in random facts and figures that stuck with the general theme. To be honest, it looks more like a rant with ramblings than a neutral article with a point in mind.
Fun Zoid RPG
The biggest threat to Linux is not Microsoft, but rather integration and interoperability issues among various Linux distributions and their applications.
Yes, this is an issue that needs to be resolved; but, to say that this is the 'biggest' threat is completely over-the-top.
I would say that the biggest threat to Linux is integration and interoperability between Microsoft and Linux/Open Source solutions.
Linux distributions don't use proprietary file formats, APIs and protocols. Microsoft can easily integrate with Open Source software. But if you're developing Open Source software that needs to integrate with Microsoft software, be prepared to pay up.
Competition creates a win-win situation for everyone.
So, why doesn't Microsoft make its file formats, APIs and protocols open and free (as in beer)?
Why is Microsoft constantly in trouble for anti-competitive practices? The only conclusion I can sensibly come to is that Microsoft doesn't like competition.
Linux/Open Source/Anti Microsoft News
Regardless of what the original poster thought, she does not supply neutral advice. If anything, what I see here is an attempt to appear neutral, but throwing in veiled threats.
Windows commands 65 to 70 percent of the server operating system market, while the Linux share stands at 15 to 20 percent
What market? What segments? What percentage of computing power? When you say MS runs 65% of the market you imply (and this is where Laura also gets here reputation) that everything else is a marginalized entity.
The high-level findings show that there is no universal clear-cut TCO basis to compel the corporate masses to do a wholesale switch from Windows to Linux
Gee, one machine, operating system, language, et. el. does not meet all needs. And this was insight? To whom? If anything this should be more concerning to MS, but it is also a plug for MS (as the owners of the market). But she follows it up with:
The majority of wholesale defections to Linux continues to come at the expense of midrange Unix installations...
The we see her drive home the MS competes with its own products (because they are just so good you need not upgrade), while Linux competes with its poorly defined systems, lack of support and array of distributions.
But, contrary to what the headlines would have us believe, the biggest threat to Microsoft's continued dominance, at present, is not Linux. It is older versions of Windows. The biggest threat to Linux is not Microsoft, but rather integration and interoperability issues among the various Linux and open-source distributions and applications. The lack of enterprise-level application support and documentation for the aforementioned software packages also is an issue.
Sorry but this is typical DiDio, with some enhanced editing to ensure it give the appearance of being unbiased. But she hit all the MS talking points.
When I switched to Linux on my desktop, I was forced to learn quite a bit more then I knew about my OS. Someone who switches to Linux because their frind told them it was better is in for a similar experience, and could end up saving his company untold fortunes by getting educated about his/her OS.
People say my sig is the best thing about me.
What a moron. Windows is not the biggest threat to Linux. That I can agree with. Windows is the best reason to switch to Linux. Microsoft isn't a threat, obvious patents filed by Microsoft are the threat. Sure, the patents are bogus, but it takes lots of money to fight off the hydra.
I took the quote from someone else's posting. I refuse to give a hit to a site publishing her trash.
Necessity is the plea for every infringement of human freedom. It is the argument of tyrants; it is the creed of slaves.
Didio is still the same old 'analyst.' Right there giving the Microsoft sales pitch with the same Microsoft paid for 'research.' This just represents a shift at Microsoft not with Didio.
It is nice to see Microsoft is realizing the Get The Facts campaign failed and they are regrouping as Linux continues to advance.
Linux is making big moves in the large companies and governments. The folks that do do their own TCO. Microsoft is just trying to infuse FUD here. "Did you do your own TCO?" "Are you sure Linux will save you money?" This is aimed at fighting the coattails that the big Linux wins will be bringing out of Microsoft's market.
Microsoft lost the debate (TCO) because it was an impossible ground to defend. Now they are trying to appeal to charging cattle.
Then you have bad admins. A (good) network or system administrator has to know at all times what is on the network and investigate where the bottlenecks and other problems are. If you do not know that, then you are indeed in big trouble (or you have Windows administrators which do indeed not know what is running on their Wintendo). Then you should not consider what systems you are going to switch to, you have to consider what people you are going to switch to.
Custom electronics and digital signage for your business: www.evcircuits.com
"Rival vendors improve the inherent performance, reliability, security and scalability of their core offerings."
She says it like she`s citing the weather information in North Alaska.
I haven`t seen proof of security, I see only headlines of new virusses being written because of security issues.
I haven`t see any of that performance increase either. My XP system`s performance is completely gone after only 4 months of operation, where my old w2k system at least pulled through for the last 3 years without much problems. Granted, it boots faster.
Reliability. Well, it`s good to know that there is more reliability, but it`s a bit difficult to test it. How much more reliable was my previous version of visual studio on win2k, compared to
Scalability. Yep. On a desktop I guess XP scales a lot better than w2k.
With great power comes great electricity bills.
Look, TCO is sales device that has gone amuck. Back in the 60s the mainframe guys came up with TCO to justify the purchase of more expensive iron against lest expensive systems by bundling the kitchen sink and some intangible, "soft" numbers with actual prices. TCO had all but dissappeared by the early 90s when Gartner suddenly came out with the now famous TCO report that applied the old TCO concept to newer computers. Sales people everywhere rejoiced because you could easily:
Bundle software, hardware, networking and professional services and compare that against existing infrastructure and the IT departments salaries.
Include whole cloth fabricated numbers such as "downtime costs" "lost productivity costs" and so on that existing systems have that superwhamadyne new systems don't have.
IT executives liked TCO because the CFO like numbers. Salespeople liked TCO because they had underutilized MS Excel chops and could create pretty convincing slides with cost comparisons. CEOs liked TCO because they like bar graphs.
Finally, the IT media which really could be rebranded as "PR Newswire for Dummies, Technology Edition" liked TCO because their articles took on an air of gravitas that they never had before.
-- $G
If you do not know where you are, just consider the cost of finding out. If your solutions have grown all over the place, it can be ridiculously expensive to do the analysis in order to find out how much they are costing. The analysis you need to do, is what solutions do you need? Once you know that, you can do some sums.
Often the analysis needed is a high level overview. If you have lots of users in similar environments all using home crafted spreadsheets - chances are you need to replace it all with some properly organised reporting from a database. Especially if you have reason to suspect that you have only 1 original copy of Office for a hundred users. If you have lots of users endlessly copying documents - chances are you need to document management system, a central repository, perhaps an internal print shop. If you have offices full of inkjet printers, you will save big money in consumables and reports by a proper deployment of laser printers. If you have loads of deskbound light email users all using Exchange server and Outlook - you are wasting an expensive resource because you could put them all on a low overhead server and Thunderbird.
Once you know your needs you can do some planning - which may be to stick with the Windows you know and love and clean up the shit. But it might be that when you expose what people actually need to do, and how they need to cooperate and share data, you would be better off building on a Linux platform.
Summary of that ramble: You do the TCO on where you need to get to, not on where you are today. Because it is practically guaranteed that you are wasting money today; you just need to find out where, in fairly broad brush terms.
I guess that analysis is why I could never have worked for McKinsey and other obsessional bean counters. But ask yourself; if you suspect you are knee deep in shit, is it better to analyse the composition of the shit or to look for a ladder?
Panurge has posted for the last time. Thanks for the positive moderations.
Unix {including Linux} only ever stops working for a Good Reason; so Unix sysadmins actually have to do hard work reading and deciphering error messages and fixing stuff {and Linux admins in particular must add googling for other people's related misfortunes and extrapolating}.
You can always spot the Linux n00bs because they are the ones who reboot a misbehaving box, see exactly the same error messages as before, and think there is something unusual about that.
Je fume. Tu fumes. Nous fûmes!
A friend of mine hired a linux guy to handle machines for several sites. About 100% of his time spent doing any kind of technical work (maintenance or repair) is spent on repairing the MS-Windows machines, which another group is actually contracted to maintain and repair. However, if he doesn't do it, then it doesn't get done. The linux machines, fortunately outnumber the MS-Windows machines significantly at those sites and he only has to touch them when he wants to try a new distro. He's always got one machine for 'experiments' at each site, to get feedback from customers on new programs, distros, or configurations.
Fortunately for my friend this fellow doesn't mind too much and has nerves of steel, a necessity since he rose up from a MS-Windows background.
Beta is broken and the link to classic doesn't work. Stop wasting our time or there won't be anybody left here.
Ok, here's the back of my envelope:
1) Everywhere I've worked, the Windows admins outnumber the Unix admins, at least 2:1, per managed machine.
2) The Windows admins seem to work in a half-frenzied state, much of the time, while the Unix admins try to look busy, much of the time.
3) Windows admins are hard-working, loyal, dutiful, do-as-you-say-Sir types, whereas Unix admins are chronically lazy -- and lazy like a fox.
4) Windows admins are excellent at solving "hit the box with open hand, right here, and it will go" problems, or "magical" solutions, but are narrowly focused on the Windows world (there are so many incantations to learn, I guess), whereas Unix admins can typically solve most problems on Windows PCs and can also manage Cisco, Macintosh, Sun, IBM, DEC, BSD PC or Linux PC, etc. (i.e. platforms that usually "just work" and are designed well, not just marketed well)
Seeing that salaries, in most organizations, grossly outweigh hardware and software costs, per year, I think I can safely toss out nearly all TCO studies and just "hire smart". I'll let my people decide what works, not a TCO study, thank you.
I'm afraid this is incorrect - OS have acquisition costs and maintenance costs. The person who applies an OS patch most likely doesn't work for free.
One thing they teach you in Management 101 is that if you can't measure it you can't manage it. The argument that Linux acquisition costs less than Windows is essentially a moot point, since the biggest single cost in IT is personnel, not hardware or software.
Salarywise I think a good Windows admin should command about as much as a good Linux or Unix admin, unfortunately the majority of Windows admins I've seen can't even spell enterprise, much less act as part of one.
TCO is high because companies choose to trade skillz for salaries - and rather than hire a good Windows admin for $80-$80k a year or so, they go for the $45k inexperienced MCSE who's only demonstrated that he passes tests well. God help them when the $45k MCSE costs them a couple hundred thousand bucks in downtime because he couldn't figure out how to fix a problem.
Friend of mine works for a major processed food producer and told me yesterday that all their Windows 2000 machines and some of the older Windows XP machines got hammered by Zotob yesterday.
I told him that the patch to fix that vulnerability was released a week ago and there was an MS security bulletin on the thing - and that any admin worth paying knows that in this day and age, once the vulnerability is made public the worm won't be more than a couple days behind. Anyway, the company he works for apparently didn't think it was real important to patch the machines and Zotob took their entire production line down - costing them a couple million bucks, I guess.
First thing I'd do is ask the CIO why the machines weren't patched and then ask for his resignation - but I would give him the opportunity to fire a couple of middle managers berfore I kicked him out the door.
TCO? The main cost is and always will be good people.
we see things not as as they are, but as we are.
-- anais nin
Perhaps, but it's in line with my experience. For me it's a bit simpler, just go through and count the trouble tickets. Systems hosted on Windows machines get twice as many complaints about outtages or down time than the sites running on a LAMP platform. If I'm going to get woken up in the middle of the night, it's almost always a Windows box.
Sure you can argue it's the application or the coding skill and it doesn't take into account the popularity and traffic of the application, but that would seem to even out across 30 or 40 apps.
At least across the companies I work with, the Windows servers need more attention. And they need to be rebooted more often, rebuilt from scratch more often and they will slow down over time. I'm not sure how you can label my observation wishful thinking. It is what it is. I run Linux at home and my personal web sites on LAMP servers because I don't have to dork with them as much.
I wonder when the last time Didiot actually had any hands on experience with server? Armchair quarterbacks aren't limited to Sunday afternoons.
That's our life, the big wheel of shit. - The Fat Man, Blue Tango Salvage
Seriously, for the past few years theres been these TCO studies and they always flop back and forth (ussually depending who the writer prefers). There's just so many, trying to include so many variables, instances, and situations that the whole TCO thing has become unhelpful.
My experience has been that companies who invested in good people, who can think objectively, for their IS departments have smoother running operations than those companies that let the TCO studies make their decisions for them.
Where I work, we run a majority of Windows boxen that we simply have to buy because there is a lot of proprietary software that we require that only runs on Windows. But for the REALLY important stuff (billing, client records, and vital for survival stuff) we have an AS/400, a couple big HP Unix servers, a Linux based Oracle database, and a little Linux email server.
The Windows problems are typically caused by the wierd quirks you find with proprietary software, but typically leads to us being told by software's support group, "Reboot the machine and lets see what happens". Honestly, most of time the reboot does fix the problem.
The AS/400 and UNIX/Linux servers never need to be arbitrarily rebooted because a program is misbehaving, they just sit there and run. If a program is misbehaving, we can kill the program, find the problem, fix it, and start the program again.
What it comes down to is having good people running your company's IS department. They are the ones who will know what works and what doesn't, and they will buy the hardware appropriate for the task.
All that said, as far as I'm concerned, 99% of all TCO studies fall under the category of FUD, irregardless of which OS they come out in favor of.
I'm not sure why the Slashdot crowd is so vituperative about this article. Laura Didio's past rantings aside, this article basically says that most corporate TCO comparisons are inaccurate because the corporations don't have enough information to make a proper comparison.
As a long time consultant, I'd say that the article is pretty accurate. It's rare that IT projects are forced to do a proper business case to justify the expenditure.
Why doesn't Slashdot ever get slashdotted?
Many organizations only use their networks for mail, file, and print services. If they are working to everyone's satisfaction and are within the accepted budget then performing countless metrics can just be a waste of time.
Why should you make major changes if all is working well and within budget?
Ninjas don't carry tic tacs
"Maybe the person who wrote this has been writing her previous work."
No doubt. Seems to me "an anonymous reader" is either DiDio or her agent trying to jack up the hits related to her "study" in order to show future editors why publishing her tripe is profitable and worthwhile for them.
I say avoid the linked article like the plague.