July 17, 2009
"Genius begins great works; labor alone finishes them."
-- Joseph Joubert

The Death of the Linux Debate: A Eulogy

Mar 28th, 2009 | By Mike Dailey | Category: Latest Articles, Linux and Open Source

In a very recent article entitled Think Before You Speak: Migrating the Enterprise to Linux I offered a set of opinions that differed from that of a blogger from the Linux community, who felt that Linux can be justified to IT management as a full replacement for Microsoft Windows.  Needless to say the article has received several thousand hits to date and those numbers continue to climb each day (making my decision to not offer any ad space on my blog one that I might regret).

As you can expect, some from the Linux community are upset with my views.  Linux and open source can be a battle cry for some and an outright religion for others.  Many troubled aspects of our world could take a few pointers from the Linux/open source community, as anyone supporting a worthwhile cause with the fervor that is displayed by the members of these communities is sure to make a difference in our world.  Several who commented or emailed me directly about my article are upset with my views, and some even degraded themselves to the point of launching negative personal attacks against me, which, given the fact that they don’t know me demonstrates the type of fervor I mentioned earlier–albeit in a less-than-becoming fashion.  There were those, however, that supported my views and thanked me for speaking what many of us in the Linux community view as the ugly truth about our own community.

A few people have challenged me to show proof of my claims that Linux is not superior to Microsoft Windows.  Let me note here that at no time did I say Windows was better than Linux; what I have stated and will continue to state is that both operating systems compliment weaknesses in the other, and when peered together in the enterprise provide a very flexible and robust computing environment.  My argument then was that the list of proposed “reasons” to provide to executive management for the complete replacement of Windows with Linux were flawed and biased.  My point was that the Linux community is perceived as spending more time trying to destroy Microsoft than they do advancing Linux and open source initiatives.

Those who refuse to accept this truth post link after link in comments or on their own blog posts, showing “proof” that Linux is a better OS; a more secure OS; a more stable OS.  I too can post links showing that Linux suffers from security issues just as well as Microsoft Windows.  I can post links showing Microsoft servers with better uptime than Linux servers.  But like the links posted by my critics, it is all meaningless.  Either side can dig for dirt, create scenarios putting their claims in a better light, or spin the facts to their favor.  This has been the mainstay of the Linux-vs-Windows debate for years, a debate in which both sides are very passionate in their arguments and cause.

But to understand my views and the purpose of my article you must understand the debate, itself.  For years the debate between Linux and Microsoft has raged; forums, blogs, and articles published in print have weighed in on the battle between “David and Goliath”, known to us as Linux and Windows.  The Linux community is sure of a victory for David, while Microsoft joins in the mocking laughter of Goliath who sees David as more of an annoyance than a threat.  We all know how the story ends.  This OS holy war continues to rage, however, backed by what can only be seen as fervent religious support on both sides.  Microsoft is still playing the role as Goliath, who has now become concerned with the progress Linux–our David–has made in the battle.  No longer is Linux a joke to the great Microsoft.  Linux is now a true threat, drawing the focus and concern of the giant.

The Linux community believes without question that the battle is theirs.  Victory has all but been proclaimed, and to state anything otherwise is seen by some in the community as blasphemy against the religion.  It is to these individuals that I direct my voice and say with the all of the strength and conviction I can gather: the debate is over.  Linux has not won.

“Blasphemy!”  No.  “Liar!”  No.  To see the truth of this you need to look no further than Corporate America, the home turf of Microsoft.  This is the battle ground where the final clash between Linux and Windows will be fought.  The European Union and China may be supporting moves against Microsoft, but the giant will not be defeated until he is felled in his own kingdom.  Here, in Corporate America, Microsoft sits on the throne of enterprise computing.  As much as the Linux community may differ with this view, the fact is that Microsoft is firmly positioned to remain on the throne.

Corporate CIOs and IT decision makers have not been given any viable alternatives to justify the removal of Microsoft from the enterprise.  This is the entire basis for my original article, and one that the Linux community simply fails to see (or chooses not to see).  I have been criticized for not providing any “proof” of my claims, but what better proof can I offer than the unwillingness of Corporate America to adopt Linux and open source as a replacement for Microsoft Windows.  Yes, Linux is enjoying a solid deployment rate as a service platform in enterprise deployments; database servers, server virtualization platforms, web servers, etc.  It is here that Corporate America has recognized the power and potential of Linux.  But in these roles, however, Linux is joining Microsoft Windows in the data center, not replacing it.  Yes, Linux is replacing some instances of Windows in the service platform role, but it is not replacing Windows in the enterprise.  I state on my own resume that I have replaced many Windows servers with Linux to achieve higher uptime and lower TCO.  But again, these were individual service platforms, not entire enterprises. 

My argument is, and always has been, that Linux and open source–in its current state–is not ready to replace Microsoft in the enterprise.  There are too many third-party enterprise-class applications leveraging Active Directory.  There are too many third-party enterprise-class applications designed to operate on or leverage Microsoft Windows services.  The term “enterprise-class” is key here; you can not replace Microsoft Exchange with Postfix and claim that you have provided equal services or functionality.  You can not replace Active Directory with LDAP and claim that you have provided equal services or functionality.

Do not offer Linux as a replacement for Windows on the desktop when it requires the re-education and re-tooling of the entire enterprise just to adopt it.  Offer a version of Linux that leverages the current experience of existing non-technical users; a version of Linux with a similar look and feel to using Windows.   Do not offer Postfix as a replacement for Exchange.  Offer a fully capable collaboration platform with the same features and functionality as Exchange, Lotus Notes, GroupWise, or any other enterprise-class platform.  Do not offer Linux as a replacement for Windows Server when it can not provide the same set of supported applications by commercial vendors.  Offer a Linux that has wide-spread third-party vendor support, including enterprise-class applications that are supportable by someone other than a group of part-time volunteer developers that may or may not feel like working on a technical issue when it occurs.  These are not my opinions; these are the demands of business.

What the Linux community believes is that the business world should adopt what they have decided to offer as a solution.  In reality, software developers must provide what is required by the business.  If your product doesn’t offer what is needed it is not adopted.  This is the simple truth of supply and demand.  This is a business decision, not a technical decision.   If Linux and open source are to succeed in the enterprise their respective communities must begin to deliver solutions that leverage the needs of businesses, not what they as developers see fit to deliver.

I am a supporter and user of both Linux and open source software.  My web site runs on Linux by choice.  I use Suse and Fedora Linux.  I use Firefox, Thunderbird, GIMP, and WinSCP to name only a few.  I too am passionate about Linux, although admittedly not to the admirable level of many–or the fanatical level of some– in the Linux community.  I do not view Linux as the savior of server platforms, brought upon the Earth to triumph over that great kingdom of evil, known to us as Microsoft.  The fact that my article has the appearance of support for Microsoft has brought upon me the wrath of some in the Linux community, and as my article continues to make the rounds on the Internet I am sure there will be more flaming arrows, hurled stones, and even the possible raining of fire and brimstone to contend with.  The truth is that I am not supporting Microsoft, nor am I supporting Linux.  I am supporting both, as I believe that both have their value in the enterprise.  Past articles I’ve written will attest to this.

Just because some in the Linux community disagree with my position does not mean that my argument is any less valid or true.  The battle continues, and the winner will be judged and declared by the united voice of IT executives and decision makers who, thus far, continue to look in the direction of Microsoft.  The undeniable fact remains that Linux and open source solutions are not seeing the rapid adoption or deployment rates that would indicate either is being accepted by the industry as a “better” platform for the enterprise.  This is the only “proof” I need to offer.  If Linux is truly the best solution for the enterprise, where are the raving reviews and mass migrations?

Any further discussion of which platform is better is meaningless, as “better” will be decided by budgetary dollars and not blog posts.  The debate between Linux and Windows is dead.  Consider this a eulogy.

 
Tags: , , ,

Mike Dailey is an Information Technology Architect and Senior Network Engineer specializing in the design, integration, and management of complex computer network and data security solutions for medium and large enterprises.

Email this author | All posts by Mike Dailey

22 Responses to “The Death of the Linux Debate: A Eulogy”

  1. anonymous says:

    This is from your resume:

    “Migrated a multitude of Windows NT/2000 systems to Red Hat Linux to lower TCO and enhance system stability and performance. “

    • Mike Dailey says:

      You are correct. That is from my resume. It says I migrated a multitude of Windows systems to Linux, not entire enterprises. There is a big difference. It is very easy to justify moving an Oracle RAC implementation from Windows Server 2003 to RedHat Enterprise Linux; the performance gains and stability are good, and it is a platform recommended by the vendor, Oracle. The TCO is there in this case. This is a world of difference from migrating an entire Windows enterprise to Linux.

      • Frank says:

        –This is a world of difference from migrating an entire Windows enterprise to Linux.

        I have found that there are only a hand full of “Windows enterprises” outside of Microsoft. As one SA to another, I feel that you give Linux too little credit. We’ve found it to be fully capable of running not only our enterprise, but the OTLP back-ends of EVERY tier 1 network operator in the US, in addition to a few in Europe and the Mid East. If you own a cell phone, you are running on a linux enterprise (with some HPUX and AIX mixed in).

        –Corporate CIOs and IT decision makers have not been given any viable alternatives to justify the removal of Microsoft from the enterprise.

        I have found that most corporate CIOs and IT decision makers don’t know what an enterprise is, let alone what the alternatives are. They believe that a web-site + groupware + workstations constitute an enterprise. If you’re filling out a time card in excel or Openair, you probably don’t have an enterprise (or a real manager for that matter).

        Frank

  2. Xackery says:

    Discrediting the messenger does not make his statement that much less true. Sadly, both yours and the article you were debating were based on opinions, and I doubt you will find any ground shaking conclusions from pure facts either, as the only real facts you can gather are by countless surveys and experiments which would most likely result in things going back and forth on which OS is best, or at least most likely won’t have conclusive evidence to really support one is truly superior.. Especially since there are some factors that are really tough to properly measure, e.g. “security”, which becomes a question of who set it up (as you noted), and even then how secure a fully set up server is, which could be measured by vulnerabilities reported while a fully secure system is in place (exploits), and even if you DO find that Microsoft is superior, or Linux is superior, and have a conclusive list why everyone should move to the other O/S, you still have to convince people, which is tough since most would simply state it’s your opinion anyways, or blindly say that they already use X product why should they move even if it’s more secure migrating will be expensive, anyways? It’s a hassle, etc.. I mean look at when Microsoft was the underdog, they didn’t go around stating truths and people simply adopted because it was the superior product. No, they undercut the competition at a key point and when they had a substantial chunk of the market, a business or person would notice all their friends were using Microsoft products, and this convinced them that it was the smart thing to do! Even if other products were superior at the time, people didn’t care, they just wanted compatibility and the comfort that came from using the same product as their neighbor. This falls into the word of business too most of the time, unless the choice is blatantly obvious, which when it comes to an operating system is rarely the case.

    You know how I get people to use Linux products? I show it to them. I say, look, this is linux! Isn’t it nifty? You can do everything you would ever likely want to do with your computer, and it looks cool! That sells more people than saying “It’s more secure! (Most people don’t even apply patches anyways) It costs less (Most people don’t even realize they’re paying for a microsoft product when they are buying a computer at 300 each), etc..

  3. Mike Dailey says:

    Well put.  Good comment, and good points.

    You are correct, my articles/posts are nothing more than my opinions, and in most cases that’s all we have out here.  I try to base my technical opinions on my technical experience as in the real world that is all you have to go on.  But to your point (and as I stated in my article) nothing will be settled here as the “facts” are too often biased and opinions just shift from one side to the other.

    My overall point is that enterprise IT decision makers are not being convinced to look at Linux beyond a few point deployments.  My opinion is that Linux/open source is lacking crucial features that are turning away these decision makers.  I could be wrong; maybe there is some other reason I am not seeing, but the fact remains that Linux is not being adopted at the rate I or anyone else in the community would like to see.  Instead of fighting with Microsoft I hope the community will focus more on delivering a product that is superior in the eyes of the decision makers.  If that happens I will be the first in line to cheer on Linux/open source.

    -Mike

  4. Hans Bezemer says:

    Mike,

    You’ve posted an excellent comment to my article. Why not put that into a post? You may be right, we might be closer to each other than you think. Why leave it at a few flaming posts? Ok, let me make a few steps to your side. I’ll change a few things here and there, make a few links. You decide whether it is good enough to come to a mutual understanding. This is not getting us anywhere, I guess. I’d rather make peace than war.

  5. Vonskippy says:

    Most Linux zealots are morons. If they knew ANYTHING about business, they would understand that business is all about money (as in making it). If (and that’s the key word) Linux was a better deal – businesses would be lining up to change over to Linux. Since they aren’t, it’s plain and simple that Windows with all it’s warts and costs is a BETTER deal for enterprise then the mostly free and feature poor open source substitutes. I have no clue why this argument is hashed over time and time again. Simple put – when Linux (and it’s apps) is better then Windows (and it’s apps) then businesses will change over – and not a minute before no matter how many fanboys are in their IT department.

    • Me says:

      Actually, Vonskippy, it’s all about Microsoft marketing. You are correct, it’s about the money, which they put into marketing rather than making a better product. There is nothing feature poor about Linux. It’s about the pointy-haired-bosses, who believe anything a marketer tells them. Funny how the PHBs like Microsoft, yet the techies – the ones who actually have to work with the software – are the ones who like Linux.

  6. j00p34 says:

    Mike, don’t you think your experience with “the linux community” has something to do with your false assumptions and uncareful reading?

  7. Mike Dailey says:

    No, J00p34, I think my experiences with the Linux community are like those that others have experienced when anything is deemed “against” the community.  People are not upset over my difference in opinion with you; they are upset because they perceived that I am “against” Linux and “for” Microsoft.  If anything represents “false assumptions”, it is those that respond to my article under the assumption that I am pro-Microsoft.  Read more carefully and you’ll see that is not the case.

    Understand that I am not attacking you, or your ideas.  My ideas differ with those you posted in how we go about migrating to Linux, but believe it or not we are on the same side.  I simply take a different approach–one I see as less “dangerous” in IT political terms–when recommending Linux to IT management.  It doesn’t mean I am right and you are wrong, it just means that I believe in my approach and will defend it.  You believe in your approach and you’ve defended it.  If we, as a community, are not allowed to disagree with one another how can we label this a “community” at all?  Isn’t it then just a bunch of coders with overzealous egos?  If that is the case we might as well change the open source licensing models and hand over the code to Steve Ballmer, as Microsoft has already won.

    Thanks,

    Mike

  8. bigbearomaha says:

    Actually, I think there is too much of this taking sides business. As usual, the truth lies somewhere in between.

    There are a number of reasons MS products are used in the enterprise that have more to do with money than the quality ( or lack thereof ) of the OS.

    Large companies, look toward volume costs as well as training. There is an entire cottage industry built around education and certification.

    Once a person and/or a company invests in paying enormous amounts of money to have techs trained and certified on a certain OS they are highly motivated to depend on said OS the money was spent on, to not do so would be to lose value on the money spent. All too often, the cost of education and training, especially for proprietary systems can soar into the tens of thousands or more in a large company/ enterprise system.

    that’s a lot of incentive to stick with what you have.

    Also, it is important to notice that the re-training needed for those trained on console based systems originally is considerably lower than for those who were trained soley/primarily on a GUI centered platform.

    This is another reason why Unix shops find the TCO and training issue less of an obstacle than do MS product based organizations. The transitions are more of a hop than a jump away.

    Something else to consider is that hardware is hardware. No matter what OS platform you use, the actual hardware is going to cost the same.Enterprises are loathe to re-purpose existing machines because it is often cheaper, per the volume purchasing agreements they achieve, it is cheaper to keep a machine with what is running than to put the time and effort into changing OS and/or software. Just wait till it’s time to replace the machine and make the change at that point.

    The last thing I want to touch upon in this post is the cost of support. In terms of vendor support, IF an enterprise is even going to consider a Linux solution, it will be from one of the big companies pushing Linux like Red Hat, Sun, Novell Suse, etc…

    The cost of support from these companies is pretty much consistent with that of any proprietary OS vendors support and is on a per host scale. So the total cost of adding Linux, if one chooses to add in vendor support, is negligible anyway.

    So really, the death of the Linux debate is mis-reported as there are a number of areas that could be rectified and improved on through discussion in so many areas that impact the broader topic than just “this OS is more broke than that OS”

    Big Bear

  9. Leandro says:

    “is business, not technical decision”

    The “marketing” part of that decision must not be ignored. Most companies and users, not being able
    to judge alternatives technically, chose the ones that convince them better, and this convincement
    is based on marketing.

  10. In my organization 2003 with InActive Directory is used for authentication mostly, who can do this to that. Strangely, I have run even larger systems that used only LDAP for authentication, and they worked very well. The fancier features of InActiveDirectory have the effect of locking-in users to M$, not giving wonderful benefits to business. Feature-bloat also has the negative effect for business that it takes longer to learn how to use the features and that the system is unnecessarily complex.

    TFA is also contradicted by surveys such as KACE that find many enterprises are seriously examining ways and means to gain independence from M$ by migrating to MacOS or GNU/Linux. see
    http://www.kace.com/about/releases/systems-management-appliance/computer-management-software-alternative/11_19_07.php
    see
    http://www.kace.com/pdf/?doc=Windows-Vista-Adoption-Trends-Technology-Professionals
    ” * 42% have considered deploying non-Windows operating systems to avoid a Vista migration compared to 44% in the 2007 survey
    * 11% are already in the process of switching operating systems, up from 9% in the 2007 survey, and a further 30% expect to switch within the next year, up from 25% in 2007
    * Mac OS is the most likely operating system to be deployed in place of Vista (29%)

    It is interesting that not everyone holds the opinion that GNU/Linux will not fly. M$ has proven repeatedly that it is not a dependable partner for many enterprises. A dependable partner would not force you to chuck working equipment, tolerate malware, and pay far more than the cost of the service. Why should anyone accept a single-source supplier who charges a lot more than competitors?

  11. Mike Dailey says:

    Thanks Robert for the interesting links and reading.  While the surveys are almost 18 months old I think we’re still subject to the same reasons why the debate is dead.  As many have pointed out, no matter how many links, surveys, studies, etc. we can cite the fact is that the information can be compiled to support whatever argument a side wants to win.  Look at the Microsoft studies that cite the superior performance and stability of Windows.  Do any of us believe it?  Not really, given the fact that MS paid for the studies.

    We can easily flip the KACE studies on their head by pointing out their own study.  “44 percent of companies are considering deploying non-Windows operating systems as an alternative to upgrading to Vista, with Macintosh the most likely OS to be deployed as an alternative.”  This tells me that the majority of the 44% considering the switch would choose MAC over Linux.  This could be interpreted as Linux being the last choice given the options.  Doesn’t sound good for desktop Linux in this context, does it?  Of those 44%, how many actually decided to switch.  That would be an interesting follow-up study.

    You are right; Microsoft has repeatedly proven that the wallet outweighs the customer, and they know they have the global customer base over the barrel.  But until we get those 44% to pick Linux over MAC and Windows we still haven’t won the war, and the debate is still meaningless in my opinion.

    Thanks for the good feedback Robert.  I wasn’t aware of the KBox product capabilities.  Do you have experience with these?  Worth a look?

    -Mike

  12. Sid Boyce says:

    In the news over the last few days — Canadian Cabinet Minister had all his emails wiped and UK parliament having to have their PC’s cleaned up of a serious virus infection.
    I haven’t had any such problems in using Linux since I made a complete switch around the days of Windows 95.
    My daughter’s Vista laptop keeps losing it’s WiFi setup, I could go on with tales of woe.
    Am I missing something? If so, what?
    Some articles are written to light a touch paper so that any strident responses will allow the “Linux bigot” label to be attached and of course pro-Windows articles slating Linux can’t have such a label as Windows and bigots never can be placed in the same box. Most Linux users came from Windows and have never regretted the move.

  13. Hello,
    You made some valid points.
    Open Source software is still working to get there in terms of feature for feature in some corporate missions.
    But it is working really FAST.

    There are visceral resistance for the change, some management miopya, some corporate people personal “interests” (I will not dig now but seasoned pros already saw this happening), and lack of long term vision, as bottom line is hurting now.
    Most non-techie people hates to learn things not directly (I mean instant DIRECTLY coupled) related to their primary job task.
    Even when these things help them in short term.
    Ask some clerks about using paragraph styles at their text editors, for example. And I do not cite editing structured texts with topics.
    Windows desktop is not intuitive nor consistent. People are simply used to it. And given that most invested many years to be “trained” on it, they are unwilling to learn new things.
    My son use Gnome and KDE since 3 years old (now it is and has a login at home Debian machine since then. They are not intuitive for daily tasks?
    Given that at corporations, desktop is preconfigured by sysadmins, the clerks simply use it, without fiddling with conf files, drivers, etc.

    There is the problem of brand awareness:
    http://www.zdnet.com.au/insight/software/soa/Is-it-Windows-7-or-KDE-4-/0,139023769,339294810,00.htm

    but corporate america is blind.
    Read the provocative text (as for long time it has been unseen):
    http://www.shirky.com/weblog/2009/03/newspapers-and-thinking-the-unthinkable/
    It will bring some ideas to the debate.

  14. Rich Richardson says:

    Nothing gets the page hits like proclaiming linux dead and then castigating the nutters, huh?

  15. Apple Uber Alles says:

    Actually, Rich, Mike said nothing about Linux “dying”, but since you brought it up, I’ll say it. Check out google trends, for example: “http://www.google.com/trends?q=linux%2C+windows%2C+apple&ctab=0&geo=all&date=all&sort=0″. At the rate it is losing mind share, I predict Linux will be OS/2′d by about the year 2015. I predict companies like Canonical and Novell will go bankrupt around 2011, and Red Hat will probably be bought out and scavenged for parts by the likes of HP or Oracle in 2013 or 2014. Enjoy!

  16. jecker says:

    Mike,
    I believe you are 100% right in saying, “both operating systems compliment weaknesses in the other, and when peered together in the enterprise provide a very flexible and robust computing environment.” As you have pointed out, some of the community’s philosophy is flawed. Don’t get be wrong, I am a Linux user also, but just because you have a superior product doesn’t mean everyone is going to rush out and change. Look at Apple: Apple has a far superior product to Microsoft, but because of price and lack of enterprise support, Apple still has a very small market share.

    Calling for the demise of Microsoft is no better then the FUD Microsoft puts out. We as a community should concentrate on improving our product and let time tell. As far as I can see, there will always be a place for both OSes.

  17. Earhardt Thoosle says:

    maybe I missed someone else pointing this out but isn’t the real future about technologies that do away with the concept of a dominant OS? Isn’t that where MS should be worried, if they’re worried? I switched over to Linux several years ago and still run a few Windows apps under Wine. I’m not trying to say Wine is the future but I have the feeling that future technologies will be driven that blur the difference between OS’s. This can only be bad news for companies who depend on the fact that they own the OS. How would sales of MS Windows ver xxx be when it became a fact that you could just as easily run MS software on the FOSS XYZ operating system? This is what I think is coming.

  18. hmmm says:

    While there may be some valid points here, I find it highly ignorant to declare the debate over. That’s like standing at a certain point midway along a river and declaring that the river stops there, then being frustrated as it flows right on by. In such a fluid and dynamic environment, the debate will never be over and positions can always change.

  19. Zsolt says:

    To Apple Uber Alles: I am sure that Google would not agree with your conclusions. I have Ubuntu installed on my machines, others I know have either Fedora, Suse or something else. The “performance” of the different distributions varies widely depending on the distribution. So the word Linux says very little as long as you haven’t named the distribution. The fact that the interpretation can be wrong is most evident if you apply the following logic: suppose Linux takes completely over in a few years. By that time the word Linux will be dead. There will be no Linux. Just “the OS” and the different distributions. So the trend might very well be the sign of Linux taking over. QED

    I agree that Linux has not beaten Windows. I agree that it will not do that in the coming five years. BUT I do not agree that there is place under the sun for both (desktop) operating systems. I’ve been using Linux for almost 15 years. I used it when it was a complete rubbish so I can appreciate the difference between what it was then and what it is now. Windows has evolved as well. Even if I don’t consider the Vista fiasco the rate of evolution for the two OSs is uncomparable. MS is loosing its breath in the competition with the OpenSource community. Not that obviously in the market shares as MS has the strongest marketing department. But the OSS “company” has a much-much larger “production department”. The rule is simple. Any software that is needed by a sufficiently large market will be best delivered by the OSS “company”. Special technologies including those for the enterprise segment can be the quarry for the software companies to fight over. If Windows will not be able to benefit from the enormous codebase of the OSS “company” it will necessarily loose the game. In the field of general purpose software nobody can survive if deprived of the work of the millions of other developers and dozens of large companies contributing to FOSS. Or at least I hope that GPLv3 will take care that this doesn’t happen. So on the long run Windows can only survive if released to the public.


Leave a Reply