Linux TCO: Less Than Half The Cost of Windows 700
ggruschow writes "Linux Today reports 'The cost of running Linux is roughly 40% that of Microsoft Windows, and only 14% that of Sun Microsystem's Solaris, according to a new study which examined the actual costs of running various operating systems over three years.'"
first? (Score:3, Insightful)
The study (Score:5, Insightful)
Production download their distro from the web? (Score:2, Insightful)
Companies will typically purchase commercial versions of Linux for pilot projects, says Robert Frances Group senior research analyst Chad Robinson, and download free versions off the Web for production deployments.
---
Isn't it the other way around? You want support for your production machine don't you?
Your time (Score:2, Insightful)
Even with free software on donated media running on a donated computer connected to free power, keeping any computer system running is not free because your time is not free. Time you spend tweaking your OS is time you could be spending producing products, selling them, and buying the food, clothing, water, and shelter you need to survive.
Obvious (Score:3, Insightful)
Not Linux Today's Study (Score:1, Insightful)
Windows XP for free...if you are a student (Score:3, Insightful)
For example, at the University of Waterloo, you can get a Windows XP CD for free [uwaterloo.ca] if you are a student. I'm sorry, but I don't know how much Waterloo paid Microsoft for this... so maybe it isn't quite free. I actually got Windows XP from them. I'm mostly a Linux user, but they sucked me in with the free CD thing. Came with a unique activation ID and everything. How soon will it be before Microsoft starts giving away Windows XP to small businesses, home users, then big businesses. They can still make their main money from Office and other things. I think they're going to have to keep cutting costs, in order to match the cost of Linux.
Basically what I'm trying to say is that Microsoft fully realizes that Linux is a somewhat cheaper system to opearate, and this is one way that it is trying to change that. By giving it away free, they've reduced the cost of running Windows by a lot.
Re:first? (Score:4, Insightful)
Re:Production download their distro from the web? (Score:5, Insightful)
We decided that using Linux could help us out w/a couple things that we wanted to do- but we were short on cash to go the MS route.
So I went to Frys Electronics and picked up RedHat. I installed it, learned how to do the stuff we wanted to do, and found out 2 things. Pretty much all the software - and support- you need are available for free.
The community provides so much more than development.
One project we needed was a server running SSH for transfering files over a dedicated T1 between us and a client. You don't need me to tell you that it was cake.
Our other larger project is focused on Apache, PHP and PostgreSQL. There is great, free support out there for all those products.
We bought the box to get started - planned to buy support but dropped those plans when we say that the open source community will provide you with tons of support.
That may not be good enough for some big companies- but for someone in the middle and (always) strapped for cash- it is great.
.
Features and productivity? (Score:3, Insightful)
As for productivity, there will be significantly less time spent on Blue Screens of death, software updates that break most of the functioning, software virii commandeering programs and wiping your hard drive clean and the like.
Your KIA/BMW comparison is highly inaccurate. Linux to Windows is comparing a Honda and a Ford Pinto with a full tank of gas and a bunch of near-sighted rage drivers right behind you rushing home to find their glasses.
Re:first? (Score:5, Insightful)
The question of credibility in this case does not rely on which news service firsts posts an article. Of course Linux Today is going to be looking for more Linux articles than say Wired.
The question of credibility in this case rests on who commissioned the study. We all complain when MS sponsered studies put MS on top. But do we point accusing fingers at one of the Windows News Sites for posting an article on the story? No. If you are going to imply lack of credibility, at least question the right people.
Now, since the Linux Today article doesn't say, who did commission the study from the Robert Frances Group?
Re:Obvious - must NOT be obvious (Score:2, Insightful)
Replace everything with free Linux and you get to send all ten to training courses for Linux desktop and office suite training (10 x 2day training @ $500 at least) = $10,000, and that is before paying for the time it takes to convert incoming documents from MS Office and making sure they look right in MS Office when they are outgoing.
Bottom line: free _doesn't_ mean cheaper from an IT management perspective. If you are starting an office from scratch and basing its operation on Linux, it is probably going to be cheaper. But converting an org from Wintel to Lintel is very expensive.
Re:Obvious (Score:3, Insightful)
It may seem obvious to you, but it's also wrong. Purchase price / licensing fees typically account for a very small percentage of TCO. So while free software may well have lower up front costs, that doesn't mean the the cost of administering it and keeping it secure is lower, and hence has less bearing on TCO than you might think. Of course, as it turns out, free software typically is cheaper to run anyway, but that's usually because it's running on an OS is designed to support multiple users and remote administration, rather than because of the lack of license fees...
Manipulation of data and definitions.. (Score:2, Insightful)
I also have a difficult time believing that one windows sysadmin can only handle 10 machines, while a linux admin can handle 44 in comparison. Of course this could be a reflection of the quality of the MCSE's out there that are brought into the sysadmin fold. *shrug*
Either way... perhaps someone should do a study of a mixed environment? Find out the optimun TCO mix, not just the black & white versions of all one way or all another.
Comment removed (Score:3, Insightful)
Re:GUI bad, CLI good? (Score:5, Insightful)
Are they built into the OS or do you have to purchase them or download them after the fact? All the Linux, Solaris and Mac OS X Server boxen I deal with have all the CLI tools built-in. I have never seen Microsoft brag about their remote CLI management.
Also, I was calling the "Zombied Windows Server Admins" idiots not all Windows Server Admins. Reading is fundamental...
The most expensive part of windows maintainence (Score:2, Insightful)
Re:first? (Score:3, Insightful)
It could be true, though, if you 'trust' the author's choice of words.
He didn't say 'The Windows technicians, however, managed an average of 10 machines each...'
He decided to stick the word 'only 10 managed an average of 10 machines eacn...'
If you don't believe the author was being truthful or clear, then it's difficult to hold a discussion since base assumptions cannot be verified. Me, I tend to believe, running a Windows and Linux home network, that a competent Linux admin can probably manage 10x as many machines as a competent Windows admin, if nothing else because of XWindows, ssh, and scripts.
The analog in Windows would be... VNC, ???, and batch files?
If you really want to talk misleading...
What versions of Windows were they using? Trying to administer 100 boxes of Windows 95 != Windows NT != Windows 2k
Re:GUI bad, CLI good? (Score:2, Insightful)
But when you're talking about TCO, you have to consider the average admin. The average Windows admin won't know the command-line management stuff (such as it is, I admit I'm not too familiar with it either) while the average Linux/Solaris admin will know the command-line management tools. Does an MCSE cert include these tools? You can bet that a RHCE does. If every Windows admin had that extra training, they'd probably cost as much as a Linux admin does, wouldn't you think?
Besides, the whole point of Windows is that it's easy to use and even a monkey could manage it, right? That's been the prevailing point that Microsoft has been pushing for years now; it's a little disingenuous to turn around and say "but you can run it as easily as Linux if you invest in equivalent command-line training".
Comment removed (Score:5, Insightful)
Re:first? (Score:1, Insightful)
And what if I could flap my arms and fly to the moon?
> because anyone can admin them.
Probably the biggest problem. "Joe from marketting knows how to fix the printer lock-up problem." "Sue, at the front desk, can fix document not found error." "Alice, the accountant can deal with the email problem." "Hey, after 5 minutes connected to the internet, this message saying we're '0w3n3d' appeared. How come?"
> they work in places that only have 10.
Yup, and they have to be right there. Have a remote office? Guess you need a remote admin.
Re:Obvious (Score:3, Insightful)
Applications being used? (Score:2, Insightful)
Okay, you say, don't do those conversions - just use Linux for the stuff that's easy to transfer, like SAP and Oracle. Now I'm stuck with having to support another operating system in my server envirionment.
In large enterprises, diversity of systems is one of the most expensive things we deal with. It's also one of the FUD tactics of our friends at MS - hey, if you can have the same OS on your desktops and your server farms, life is supposed to be easier, right?
I welcome your constructive comments on this challenge. I'd love to be able to sell this into m enterprise, but my forehead is already pretty sore from wall beating.
Windows is more expensive because... (Score:4, Insightful)
Re:first? (Score:2, Insightful)
Linux machines get used most often as servers where there are very strict requirements about what is allowed to go on them and who is allowed to have access. Just my two pence.
Re:Follow the money! (Score:3, Insightful)
Their numbers on Solaris are whacko.
To establish my background (and biases),
I've been a UNIX sysadmin for 15 years
primarily on Sun, but including most
other UNIX variants and a little Windows.
I'm currently at a Sun VAR, so I have an
idea about how to price Sun boxes and software.
First of all, it will take about 1 CPU to
meet their Processing Unit definition.
When your basline comparison unit is less
than 1 CPUs worth of effort, comparing using Sun
enterprise class systems is ridiculous.
They're not intended to replace a stack
of 1U servers; they're there for the one
application which needs 99.5% uptime or
better and doesn't split across clusters
of independent systems. The Sun 1U servers
(Intel or SPARC architecture) are $1k each
including OS, not this ridiculous $12.5K
per CPU that they ascribe to software costs.
My opinion: this is a sly slam on Sun made
to look like a boost of Linux.
Here is my followup as promised (Score:3, Insightful)
First of all this thing is 8 pages? The last 15 or so TCOs I've had to read through have been in the area of 70 to 400 pages in length. This looks like something a 10 year old would write in highschool. But I digress. Here we go:
1 - Executive Summary
Who are these numerous executives? (I had coffee with Elvis)
How many is many? Many people have Aids but MOST people do not. (I eat a lot of food compared to a starving Ethiopian)
Survey Participants? Who are they? How do I know you didn't make this shit up? Where is your work sited information? The least you could do is say Client - A, and Client - B if there was a NDA issue.
2 - Methodology
Ok so their WHOLE concept of this TCO factors ONLY web servers? This should be titled as TCO of LINUX WEB SERVERS. Holy Shit I'd fire these guys if they were my consulting shop.
One or two processors in the machines? How do I know the majority of the MS or Solaris machines are not Single versus Dual on the Linux Boxes? Where is my node breakdown summary?
"External" support hardware and software were excluded?! Holy Fuck! Ahem! I consider hardware and software compatibility pretty fucking important. The NIC performance is crucial! If I have to alter my NIC choices based on driver availability that can totally skew a cost per MB in a TCO. The difference, for example, in a web server using Intel and 3Com NICs amounted to $1500 dollars when placed into a Proliant server doing SQL after boiling the TCO numbers. BIG difference if I have to have 100-200 of those.
All prices are based on retail!? What moron pays retail for a corporate purchase on Mid to Large sized companies!? Mother of God are these guys scamming their clients? I get a bulk discount on orders greater than 20 from 3Com. Used to level the playing field? Ahem this is a COST analysis! Enterprise discounts are not irrelevant THEY ARE CRUCIAL! Case in point if Intel gives my 20% off on an order of 2000 NIC cards and 3Com gives me 25% off and Linux won't support the 3Com card (for the sake of argument) THAT IS RELIVANT TO THE TCO!
3 - Cost Breakdown
Software purchase costs per processing unit? Ahem where is the implementation costs, maintenance costs, or as I like to put HOW COULD THESE DOLTS COMPLETELY IGNORE THE SYSTEM LIFE CYCLE!?
Paragraph 3, line 4 "beyond the purchase of the software", ahem DOES LINUX INSTALLS AND CONFIGURES ITSELF!!?? FUCK DUDE I'M SWITCHING NOW! They cover this later; I'll bitch about it when I get there.
CALs are primarily used only in Intranets unless your are running remote services like Outlook's Mail program. But authenticated services are NOT being specified on both systems, only MS. Are we running software here besides Apache and IIS comparisons? Now we have a whole separate TCO on just the application alone! God this TCO is a mess.
Also there is no lie in the last paragraph, the new terms from MS are terrible.
No lie here, Linux is cheaper cost wise for the software. Too bad like in automobiles labor is the major cost. Even then the data is terrible at best.
3.2 - Hardware Cost Breakdown
Wow Linux only beat out MS by about $1000 bucks? I'd rather pay the $1000 bucks and write that off as application compatibility, hardware compatibility, and more importantly I actually have the MAJORITY of hardware vendors writing drivers NOW for Microsoft. Don't tell me "But that is changing" TCOs are a static snapshot, "What-Ifs" are not allowed, otherwise they become Cost Projection Reports.
Concerning their benchmark concept per processing unit how do we know we have the same data going across? Where are these number coming from? I have yet to see any concrete data. I get the results but how was the data collected? Was the test based on identical web pages? Was it base on client side or server sides scripts? WHERE IS THIS DATA COMING FROM? When I play cards and my friend says he has a full house he's gotta show me the cards. SHOW ME THE DATA!
3.3 - Support
Oh God I loved this part:
"Support Costs Were Those Fees Paid To Consulting Providers or Product Vendors...." "Many administrators were taking advantage of mailing lists, free news groups, support,
Hmmm how many of those groups will mail me monthly CDs like TechNet for a fee? I wonder if that added to the cost? WE DON'T KNOW THEY STILL HAVEN'T GIVEN US ANY DATA. WE ARE SHORT ON BRENT SPENER (Did I get his name right? I'm not a Trekkie) JUST RESULTS. Did your result of 42 come from 40+2 or 21x2 or WHAT!? They say $46,360 for MS. I have been consulting 8 years and have NEVER seen an administrative expenditure like that. Show me the data! Are we looking at 1000 Linux Workstations for every 4000 NT based systems? Right there the admin costs should be x4 as much. No data, no trust. Tell me where that 46k went and then I'll listen.
Oh how about this one,"... for the purpose of this survey administrators...."
Ok so my NT guy that handles my SQL server, Exchange, and 4 other servers is only going to be counted for the web part. Hmmm... Wouldn't that deflate the number of Web specific servers per admin? WE DON'T KNOW! GUESS WHAT? STILL NO DATA! If the 1 Linux admin handles 10 web servers, but all he does is handle 10 web servers that is going to drastically skew results of the MS web guys all share duties on 5 other types of server which is the case.
How about bullet #2 "... System automation tasks... had not been written yet for Linux..." Is this guy drunk?! Holy try going to one of hundreds of scripting pages for Linux you dolt!
4 - Soft Costs
"... Difficult to assign values to..."
Let me think, WRONG. I can roughly estimate over 3 years what those costs are by taking fixed costs and subtracting budget expenditures for the year and I can write THE WHOLE GOD DAMN THING OFF as a soft costs and estimates. That is how you in part determine the next year's budget. The more years you factor in the better the estimate. There is a reason TCOs I read are at least 30 pages long. I accept nothing less.
4.1 - Security
No arguments save one, The reason Linux SEEMS more secure than MS is that is hasn't been as critical of a target. I remember some Linux admins (back when Slashdot was just starting) making the claim that Linux was IMMUNE to viruses. No, people just haven't been writing Linux based viruses. Same with hacks in general. Here is a real solid fact:
Based on the number of attempts and system penetration MS products are 20% more likely to be hacked and infected than Linux. This is a basic arithmetic case of market share. If 60% of the targets are red and the other 40% are blue. Red is 20% more likely to be targeted than blue. It's that simple.
4.2 Availability
Holy this doesn't even get a page?! If I do 4 billion dollar of transaction a week this had better fucking be at least 30 pages long with in-depth up-time analysis including MTF ratings and severity analysis. This is a glaring example of RFG's TOTAL AND COMPLETE LACK OF CREDEBILITY. If I am Amazon or Barnes and Noble if my site isn't up I have no business. How they could blow over this is
4.3 Scalability
One word, Datacenter.
From a cpu standpoint MS leads, note the fact they kind of gloss over this section. Damn near a page for software costs with some statistics but virtually nothing here. This fucking piece of trash looks more like a poorly disguised outsourcing bid from some half-assed Linux shop. With the advent of cheap blade servers and AMD's entry into SMP this should have been a 60 to 100 page section! What about support RAID systems, Fibre Channel links, high speed switching systems, clustering, FUCK THE LIST JUST KEEPS GOING ON AND ON AND ON....
These people have NO FUCKING clue how to do a TCO. NONE. ZERO. ZIP. My 14 year old non-techie niece could do a better cost analysis.
I am not going to even bother on the rest of this crap. I have only one thing to say in my summary:
IF RFG IS WORKING FOR YOU, FIND BETTER. THE HOMELESS GUY DOWN THE STREET MIGHT EVEN BE BETTER.
Suggestion to web sites that quote this: Give Linux a real chance to succeed Bullshit like this doesn't help.
Re:first? (Score:5, Insightful)
Basically: scripting is everything. If you understand shell scripting and perl, you can make Unix machines dance. A real Unix wizard can nearly bring about world peace from the command line.
Scripting in Windows is much harder. It can be done, but it's relatively alien to the system, and some complex things are unscriptable.
Windows 2000 has improved this capability a lot. If they have been studying for three years, chances are that a lot of the machines are still 4.0. A true from-the-bottom-up 2K network is A LOT easier to administer than 4.0. One poster in the linked thread claims to be running about 200 clients and 37 2K servers all by himself. With 4.0, I don't think that would be possible. Things would break faster than you could fix them. With 2K I can just barely imagine doing it, though I bet that guy is incredibly busy.
Linux is easier still to administer. Perl, ASCII text configuration files, and separation of services beat Kixstart and the registry hands-down.
Re:Perhaps... (Score:3, Insightful)
I'm not saying Windows is easier/harder to amin, just that you need someone with equal experience admin'ing both to make a fair comparison.
Offsets the free as in beer copies issue... (Score:3, Insightful)
Grab the PDF from... (Score:3, Insightful)
The link at the bottom goes to ibm.com(/FUD)
I wonder if they would per chance be biased against solaris in anyway?
Maybe that explains why they think people will use 6800's to run a web farm. Their hardware comparison is dell pcs and 6800's and 4800's.
Really...
Re:first? (Score:3, Insightful)
Actually, statistics are not usually crap. The real story around statistics is in the interpretation. Media almost always try to read more into statistics than is really there. So, some statistical result which really isn't as cut-and-dried as it seems to the untrained eye gets turned into a headline like "Linux TCO: Less Than Half The Cost of Windows".
Some folks like to think they are informed by being able to spit out information like "The cost of running Linux is roughly 40% that of Microsoft Windows." A number like that is meaningless without understanding of the study, the data, and all of the factors that went into getting that data. Otherwise, it's just meaningless hype or FUD, depending on the point of view.
Re:Windows XP for free...if you are a student (Score:3, Insightful)
The idea behind giving away a such a controlling OS isn't to lower the cost of anything, it's simply an easy move to infect the cash-strapped with a way to easily perform a "mind and movement" control endgame manuver around free operating systems. It's a noose. A digital chasity belt to keep us in line--keep us from ripping, mixing, and burning.
Maybe they think if they can fit the young with this prophylatic DRM measure, they won't have such a hard time later when they really start tightening the screws and throwing away the keys.
Re:first? (Score:2, Insightful)
If you're able to pass the tests without ever looking up a SINGLE braindump, my hat's off to you - they're quite difficult.
Using braindumps, I'm sure anyone with an IQ of 80, with no computer experience, could probably pass the test just from straight memorization of the questions.
That is the problem with paper MCSEs, and thus the dilution in the value of an MCSE certification.
MeepMeep
Development Costs (Score:2, Insightful)
It seems to me that a lot of the push for Windows comes from the development side, especially in the web environment. They like being able to open up Visual Studio and have it write a solution for them. Granted, it will be full of buffer overflows, but it will be finished by the deadline, if not before.
Correct (Score:2, Insightful)
They should have used X1's or at least compared the Sun's against mid range intel Xeon HW, with all of the same redundant power, etc. The comparison they make is stupid. The Sun HW offers the flexibility to replace processors without shutting down the system. That kind of function costs money, but in this comparison it has no value.
If I were they I would have looked at the results and said "Holy sh-t, this can't be right" and investigated where I went wrong in my calculations.
Its the unquantifiable stuff that makes Solaris, and Linux so good. The security, the reliability, the sane licensing. All the things that they admit do not factor into their TCO measure.
Re:Number security advisories this year (Score:2, Insightful)
of applications that go along with it. Microsoft's listing here is just for the
OS. This was discussed earlier this year.
One other note. Many of the advisories against Linux this year, in general were
put out before 0-day events, and the hacks were derived from the advisories. M
ost the Microsoft advisories were post 0-day events.
Also, Why aren't you listing debian, mandrake and the other Linux distributions
here? Just curious?
Going out and listing the advisories against any
OS isn't necessarily a good way to define weather or not it's secure. There are
a lot of other issues involved.
1) How easy is it to secure the OS
2) What kinds of security holes are cropping up.
I'd rather have a dozen local exploits to 1
remote exploit.
3) What kind of response do you get for fixes to
the exploits. Microsoft has gotten much
better in this respect this year.
Also note that each system/OS has it's niche. I wouldn't run a terabyte databas
e on Microsoft or Linux. I wouldn't run a web frontend on a Sunfire 15000.
This article seems to be providing alot of fluff,
and not putting much hard evidence out. I know Linux inside out, and could conf
igure it to be cheaper faster and more stable than Microsoft in many application
s. That's because I know it well enough to do this. There are $M people that w
ill say the same about $M products. I enjoy going back and forth with them abou
t this. We keep it on a razzing level, and not to serious. Everyone else out t
here should too!!
W.Kid
Re:first? (Score:3, Insightful)
It's a total of 8 tests, each of which costs about $125 a piece per try to take.
But that doesn't say anything about the competency of people coming out the other end.
Could a Windows sys-admin with a few years of experience pass with no specific preperation? Maybe, but probably not.
Doesn't that mean that the test is not an accurate assessment of skills? I would think that it *should* pass an experienced Windows admin. Otherwise, if they're passing these inexperienced-but-looking-through-expensive-MS-st
I rather suspect that it would make more sense to have "subject tests". If you're going to be administering IIS, then test IIS. If you're going to be certified to administer NT 4.0 for stand alone workstation use, test that.
That being said, I think that a fair number of the most capable IT sorts learned all this on their own well before the whole idea of "testing" came up, and so are a bit scornful of tests.
Re:first? (Score:2, Insightful)
I have an NT4 MCSE. I only failed one test once: the SQL Server 6.5 Administration test, and I had never used it outside of class and I waited over a year after the class was over to take the test. I passed it on the second try.
I don't cram study or use the brain drain/cram/dump/whatever sites. I learn by using it, in this case in a school lab or home test setup.
I also was way more experienced with Win95 clients on a NetWare 4.1 network and hadn't used a domain environment outside of MCSE class or home.
I am very sharp with fundamentals, though, and have a knack for remembering technical gotchas.
The main trick for the MCSE's from my point of view was seeing the world through Microsoft's eyes and answering the questions the way Bill Gates would answer them. In my case it was learning the domains. At least 3 of the tests asked questions about different domain setups for different applications. Think like Bill, not a Linux geek. I swear several times I had to go against my better judgement and decided Microsoft would see it this way and got the question right. (At least in the practice tests; I don't think they tell you which exact questions were right & wrong in the real test.)
If I ever get around to updating to MSCE 2000--ahem MCSE
But I shopped my MCSE around a couple of years ago and found it worthless in my case. By that time everyone knew about paper MCSEs and I guess I was one since my experience dealt more with Win95 & NetWare & un*x. Besides I was very disheartened after spending all that money on a cert that expired a year or two later. (Well, MS backed out on expiring it but it's basically worthless, anyway.) Yet my Associate's degree is still as important as when I got it 8 years ago. I'd be better off finishing a BA or BS and selling my experience next time than chasing these vendor-specific certs.
But all this ranting leads back to TCO for Windows & Linux, but that misses the point I think. What you're looking for is return on investment. Someone's calculation of TCO doesn't apply to your company. If you're a fortune 500 company with 250 MCSE's scattered about and an IT support staff that's dealt with Windows for 10 years your Linux ROI won't be lower unless you want to lay off all the Windows guys who don't already know Linux because you'll have more downtime and more paper MCSE's scratching their heads and butts. If you do lay off your whole IT staff and find Linux guys then they won't know your business and probably spend much of their time trying to migrate business practices to fit Linux rather than applying linux to the company's business.
I've had ideas about making a custom Deiban distro for my company. I'm quite sure I can use free software to handle all critical apps. But the managers and analysts use Word and Excel and I'd never get them to like OpenOffice.org. There are plenty of other problems, but I'm already off topic again.
Re:Perhaps... (Score:3, Insightful)
What you might also not know is the update packages MS sends out are horrendeously (sp?) documented, use different (undocumented) switches and have a small chance of blowing up whatever you're patching. This is the biggest patching problem, not the logistics of making the patches run. Are there similar problems with patches in Linux?
Funny sig btw
Re:Does it matter *what* explains the difference? (Score:2, Insightful)
Then, later, they decide that the ftp server could be linux and no-one in the company would know.
Then, later again, they decide that they're going to invest their precious time in learning linux and python/perl and they slowly migrate from a windows admin to a unix/linux admin.
Also, the fact that if you get out of the loop for two or three years as a windows sysadmin you have to relearn everything from scratch because windows servers will look completely different doesn't help. Some people decide that they've learned enough and don't want to mess with re-learning all the tricks every time MS wants to change its paradigm.
"viruses" is greatly outweighed by "and stuff" (Score:2, Insightful)
Saying "But it's a Mac ..." is simply the logical fallacy of special pleading. If you don't understand why network admins wouldn't want a personal machine on the network, you don't understand security.
Re:first? (Score:2, Insightful)
Great. You can set up a boot option labeled N that logs on to netware and fires up a script there to install an image (or maybe just an OS install with a custom
That's not very complex on the whole, and not complex at all on the windows side. Netware's doing all the work.
So what else can you do?
Re:Typical MS admin tripe.... (Score:3, Insightful)
Plus, the risk of a Mac OS virus infecting an x86-based machine is exactly 0. Different opcodes.
Irony, or just gross bias? (Score:4, Insightful)
Basically, the LINUX TODAY article is saying the TCP (Total cost to purchase) is equivalent to TCO (Total cost of ownership)... in effect saying that any positives and negatives Windows may have in the software itself has NO value. As other above have said, you can calculate TCO in many different ways. If you want to assign EVERYTHING other than software licensing a VALUE of $0... maybe this article strikes a chord with you.
All I am saying is consider the source!
Re:Experience at Excite@Home (Score:2, Insightful)
This has nothing to do with open source, but more to do with the which communities evolved the techonlogy and the underlying motivations of companies hawking their wares.
That's everything to do with open source. You can't have a truly good technological community if one party has total control and can "take the ball and leave". Unless you only meant you don't have too look at the source to benifit from open source.
Anonymous only because it's off topic. Maybe I should make a seporate off topic account.
Re:Irony, or just gross bias? (Score:3, Insightful)
and) Linux Today quoted the study, they didn't author it.
and) The study included maintanence costs, not just capital outlay.
Therefore) Quit self-moderating.