Forgot your password?
typodupeerror

OSDL CEO Answers Slashdot Questions 72

Posted by Roblimo
from the bias-is-in-the-eye-of-the-beholder dept.
Here are OSDL CEO Stuart Cohen's answers to your questions, many of which were spurred by the recent release of an OSDL co-sponsored Linux TCO study (that anti-open source commentator Rob Enderle has already/inevitably questioned).
(1) A Movement with the Students
by eldavojohn


This may seem like an inane question but why don't I see more of a push to get Linux into the realm of academia?

I know that Ubuntu [ubuntufund.org] has made strides to incorporate themselves into learning environments but where is the effort to alert students (primarily other than computer science majors) to the benefits of Linux?

When I was a freshman at the University of Minnesota, a friend handed me a CD distribution of Debian that would change my life. I knew of the Linux labs in the University but only now did they interest me. I'm now getting my masters at George Mason University and I don't believe there's a single Linux machine on campus. In fact, the whole Computer Science department has only two Sun servers to offer me an account on! Everything else is Microsoft!

Now you may lay claim that every computer science major these days is running Linux anyway. But how about the other areas of study? I used to take music theory and people would rant and rave about their Macs or one of various composing suites in Windows. I tried explaining that Linux has (certainly more affordable) solutions to offer in this department too but no one would even listen to me. It's not like they were mixing platinum selling records, they were just looking for software to write sheet music with.

I think that both Apple and Microsoft realize that the toys people have in college become the toys they demand in real life. So there are all these [apple.com] efforts [e-academy.com] to garner the student's interest hoping that they will use them in their careers.

They make it free (which Linux already is), they make it easy and they make it available.

So how about it? Why isn't the Linux community minting install discs and distributing literature on campuses? Why isn't Linux tailoring cheap solutions to K-12 schools that don't have the money for Windows anyway? Why do we risk letting someone leave academia without ever experiencing the real fruits of it?

If you are doing this (and I just don't know about it), what steps have you taken?

Stuart:

I have found from conversations with CS professors that while Linux is still finding its way into college curricula, it is being used by students on their PCs and in their university labs. It's also being used by college IT staff for the same reasons it is appealing to corporate IT managers - cost of acquisition, lower cost of management and the ability to prolong the lifetime of legacy hardware. And, in Asia and across a range of developing countries, by contrast, Linux is forming the core of much curricula, especially in countries where governments see Linux as a means to local technology development. In many universities that I visit outside the U.S., I meet with the head of open source for the university. This title doesn't exist in U.S., which implies that the academic tie to local economic development and the advancement of information systems is weaker here than abroad.

I agree that what users become comfortable with in college they take with them into their professional careers. We think Linux has a very bright future in education. We launched a program to bring together leading universities around the world to share their experiences and lessons learned to help accelerate that trend. As a small non-profit, OSDL can't by itself take on the global job of delivering Linux to schools and universities, but we can encourage our members to support Linux in education and we can bring together universities to encourage them to embrace more use of Linux. And we do.

(2) Bias
by MaestroSartori


Since almost all of these studies are funded or organised by a party which appears to be inherently for or against one of the things being studied, will it be possible to find anyone willing to compare them impartially? After all, how many people would believe an Open Source company to be any less biased than MS when it comes to comparing their products?

Stuart:

Questioning the credibility of ALL analyst studies is important. Let me just elaborate on how this particular study originated. I hope the context will help you weigh the impartiality of the study and judge for yourself. Levanta (co-sponsor of the study and a member of OSDL) was the vendor that initiated this study by Enterprise Management Associates. In their own sales scenarios for their Linux product, they were running into organizations that were saying they were apprehensive about Linux because they had read that it was "tough to manage." Levanta believed from its own experiences that the concern did not meet up with the reality -- so they sponsored EMA to get some real-world feedback from end user Linux customers. They also wanted current information. Most of the analyst research published to date was already old, going back as much as four years or more. Linux itself, and the tools available to help customers manage Linux, have come a long ways since then. When we (OSDL) saw an early draft of the report, we were very interested in co-sponsoring the effort because the report echoed what we have been hearing from our Linux User Advisory Councils on three continents for a number of years. It was current information with interviews conducted in November and December of 2005. The customers interviewed were also using current versions of the kernel and current releases of enterprise distributions. It reflected today's market situation. And it was based on real Linux customer feedback.

(3) Security Question
by db32


How can we fix the problem of the way TCO studies handle security? In so many of them every OSS application under the sun gets tallied against Linux systems, regardless of how obscure, or unrequired that application may be. Yet all of the 3rd party things that have holes in them rarely seem to even get looked at when talking about Windows security. Firefox for example seems to get tagged frequently when talking about Linux security in these studies, but Firefox isn't integreated into Linux, and it runs on both platforms. IE on the other hand is integrated into the OS, sure you can not use it, but there is a ton of junk in Windows itself that requires the various bits and pieces of IE to operate correctly. What is it going to take for these studies to finally start comparing apples to apples in regards as to what really is part of the OS and what is required for it to run?

Stuart:

It's difficult to say if this will ever be able to be measured "apples to apples."

Just as with Windows, the Linux platform cannot be judged, security-wise, as a "naked" kernel but rather as an OS with a variety of components. In the case of Linux, the component set is determined by the 200-500 packages offered by typical distributions and variously installed by systems administrators and by end users. However, very much UNLIKE Windows, there are few, if any, applications that pervade typical Linux-based applications stacks and carry with them pervasive security issues the way that IE pervades and exposes most parts of a Windows stack. Indeed, Linux is very modular and compartmentalized, such that even if security/TCO studies illustrate holes and exploits in the Linux stack over time, any given breach will have much less impact and patches to repair exploits are far less likely to perturb the entire stack (especially compared to the havoc wreaked by frequent Windows patches and service packs).

(4) Setting up Linux from Win2K3
by digitaldc


Say I wanted to switch from Windows Server 2003 to Linux in a company of about 400 people with the same equipment I already have, generally speaking how long would it take and how much would I need to invest?

Do I need to hire several Linux experts just to get it up and running?

Would you expect this to be relatively easy or would it be very complicated and time consuming?

Stuart:

It's difficult to answer your questions without more information but I can tell you that comparing legacy and migrated systems costs is never an apples-to-apples exercise. Seldom do legacy applications, both COTS and in-house, port on a one-for-one basis and with comparable acquisition and maintenance costs. Certainly starting from the bottom of the stack and moving upwards, it is easy to show improved TCO for comparable "abstract" loads. Linux on Intel, AMD and power hardware will cost less than per-system licensing and the short deployed lifetimes typical of Windows-based stacks. The LAMP stack is demonstrably more stable and more scalable than comparable proprietary Microsoft equivalents - even moving from WAMP variants will yield benefit in terms of reliability. SAMBA, NFS, OpenLDAP and other user-base technologies will scale more cheaply and reliably than Windows-based legacy.

Cost and ease of migration, however, will always vary based on the depth and breadth of specific dependencies within legacy loads and availability of COTS components (or work-alike replacements) on the target Linux system.

(5) If OSDL believes that Linux has a superior TCO ...
by hweimer


... why don't they use it?

Almost every PDF document on the OSDL website has been created on a Windows PC or on a Mac. Even the Desktop Linux Survey Report - http://www.osdl.org/dtl/DTL_Survey_Report_Nov2005.pdf - shows:

$ pdfinfo DTL_Survey_Report_Nov2005.pdf
Title: Microsoft Word DTL_Survey_Report_v4.doc
Creator: Word
Producer: Mac OS X 10.4.3 Quartz PDFContext


Stuart:

OSDL runs on Linux. Many of the reports we publish, however, are formatted by outside vendors or members of OSDL. We don't dictate to others what format they send to us.

But fair enough, we could do better.

I must also say that, while the Linux Desktop is getting better and better with laptops integrated with PDAs and cell phones, it still lacks some features that are standard on other platforms. Take for example the "Linux Desktop Survey" you reference. OSDL's principal analyst Dave Rosenberg wrote the study with Open Office on Linux. Because he had to combine his file with an existing PDF (not available with Open Office today), he found it faster and easier to complete this task on the Mac using Acrobat.

(6) Why Should We Care?
by illuminatedwax


It's a Serious Question. Don't TCO costs end up coming down to how much you will pay employees, how many employees you need, and the price of software? Shouldn't any capable manager be able to estimate the costs themselves? After all, I'm certain TCO varies wildly from workplace to workplace, considering what kind of system is already in place, what software is readily available for an OS, and what skills your current employees have.

My question is: is there really a use for these reports other than for 'defense': positive propaganda versus negative propaganda?

As an aside, do these studies take into account the availability and flexibility of currently extant software? Is there even a way to turn that information into TCO?

Stuart:

Despite the "T" in TCO, meaningful studies must limit the parameters under consideration in order to observe scientific principles. In the recent EMA study, the scope was confined to an area that Microsoft traditionally cites as a strength - systems management. Assuming comparable applications loads, the study showed that IT staffs could manage more Linux-based servers than Windows-based systems. Implicit in this metric is that managing the same number of servers would thereby require fewer or less costly IT resources, with immediate impact on TCO.

Any capable manager should in theory be able to perform such a calculation, but organizations often undercount or ignore real costs of maintaining legacy software, actual headcount needs, and deployment of different types of equipment.

(7) Is it about Linux or better operating systems
by selil


What I would really like to know is why Linux or Windows? Why hasn't there been a really good study that included BSD, Solaris, OSX, or even licensed variants of Unix? Is it all about Linux or is it about better operating systems?

Stuart:

You're right, it should be about the best operating system for you, for your specific needs. Linux keeps winning in the market because it's stable and secure with the ability to handle mission critical workloads. Because our mission at OSDL is to accelerate the adoption of Linux, that's where we focus our resources.

(8) One of the main problems
by petrus4


In looking at Microsoft's TCO claims in particular, I've been unable to avoid noticing that a lot of the company's material on this subject consists of, to put it simply, straight lies. Aside from anything else, nothing is mentioned by them about their licensing fees. How they can state with a straight face that after their licensing fees, Windows can still be cheaper than Linux is beyond me.

Legitimate performance competition is one thing, but I'm curious to know how the ODSL is able to deal with Microsoft's lack of ethics in this regard? Given Microsoft's marketing power, how are Linux advocates able to communicate to people that many of Microsoft's claims in this area are deceptive?

Stuart:

Your comments are very familiar to me. These are the types of thing we kept hearing from our Linux User Advisory Councils - and why when we saw the EMA study, we decided to co-sponsor it. To more specifically answer your question, though - Linux advocates have a powerful weapon at their disposal: real users' experiences. These experiences are very positive and when shared with others, tell the real story. Again, that is why we chose to support this particular study - it's current, it clearly represents user experiences. It's consistent with fact. The other point that is important to make is that it's also supported by market facts -- and that's the growth rate of Linux adoption. It continues to grow twice as fast as Windows and 5x faster than the server market. That means more customers every day are making a business decision to move more of their work to Linux. If Linux has all these problems, then why do people keep making poor financial decisions in a market filled with choices? This logic is catching up to Microsoft. I think Microsoft is a well managed company and they listen closely to their customers. I expect you'll see them moving away from the comparisons they tried to promote in early 'studies' around TCO or security that had no affect on the growth of Linux.

(9) Are the OSS IP Indemnification offerings worthy?
by csoto


We recently had an issue in which Microsoft Office included unlicensed IP (according to a court settlement). Microsoft did not require us to patch existing installations, rather simply protecting our use via the settlement, agreeing to require future installations to include the patch. This seems like a case in which indemnifications worked (although they could have offered some compensation for the extra work - it's cheaper than litigation). For background, see http://www.microsoft.com/windowsserversystem/facts /topics/ipi.mspx.

How do the OSS indemnification plans stack up? Have there been any significant cases involving IP indemnification?

Stuart:

In fact, there has been relatively little litigation around Linux and open source software in absolute terms (a handful of cases) and when compared to the thousands of suits filed every year around proprietary software, practically none. While companies like HP, Novell, and several embedded players offer a range of IP indemnification, uptake and actual use of the policies have not been that significant. Also, while customers look closely at IP issues, it has not been key decision criteria for customers of any size or located in any particular country.

(10) What difference can OSDL make?
by Goeland86


There's certainly been a good few questions asked already, but the one I'd like to get an answer to is, how do companies see OSDL? Do they believe it's a trustworthy group that knows what they're talking about, or does it look like another one of those 'fad-like' groups that's going to fade away? I don't mean to say OSDL is fading out, I'm curious to know what the real-world perception of it is. I've noticed that while many of my friends use linux and are generally well-versed in what's going on, they're usually totally unaware of the existence of OSDL, or its purpose.

How will this change? How will OSDL become a trusted group for IT managers, especially in a world where most of them have only heard of Microsoft's "Get the facts", or have some shares in MS stocks?

I feel that part of the reason that one of the above posters was asking why isn't linux penetrating the educational market is because the trustees funding the schools have a say in what to use, because they're paying for it, and the trustees will usually have a significant amount of MS stocks.

What's the chance of all of this changing? Or rather, what are the means in place for all that to change?

Stuart:

There are a lot of questions in your question, and I can't speak for how other organizations or companies perceive OSDL since I represent OSDL. But, with more than 70 members, I can tell you that one of OSDL's key attributes is vendor neutrality. OSDL is in a unique position to bring vendors, users and developers together. In doing this, OSDL bridges gaps among companies that alone struggle to move technology forward but when combined can quickly accelerate technologies via common platforms or testing scenarios. As Linux and open source software move into the mainstream of IT environments, the voice of the user, as well as the vendors, will be more and more important. I think the programs we have today and the ones we will announce shortly will help to bridge the gaps with developers, vendors and users. And remember that our mission is advancing the use of Linux and open source software, and not building awareness of OSDL.

---
This discussion has been archived. No new comments can be posted.

OSDL CEO Answers Slashdot Questions

Comments Filter:
  • by BoomerSooner (308737) on Tuesday February 21, 2006 @12:14PM (#14768844) Homepage Journal
    I have to agree with the article questioning the validity of the study.

    In the end studies are only worth something if they can be verified and are accurate with a sample that has as few biases (sp?) as possible. I use linux, os x, windows, solaris, ... in my businesses. Best tool for the job is my attitude. I happen to like them all!

    Cheers!
    • by khasim (1285) <brandioch.conner@gmail.com> on Tuesday February 21, 2006 @12:25PM (#14768938)
      #1. The "data" is from a self-selecting group.

      #2. The "data" is self-reported.

      That means that you get all kinds of biases. We wouldn't accept that from any Microsoft funded "study". So we should not use the same techniques that Microsoft uses.

      Although he does go far overboard in his condemnation of the the "study". Microsoft has published far more "studies" just like that and he has not complained at all about those.

      He has a bias against Linux and now he has a "study" that he can rant against.
      • A survey is a study. However, any college student that has taken a research methods class would tell you that reliability of a survey is quite low compared to other studies. It is therefore, a much less desireable study to run. Both point #1 and #2 are valid for all surveys, which leads to the reliability issues. What might be more interesting to know would be whether ESA(?) ended up selecting customers that they knew had good Linux experiencees.

        One thing to keep in mind however, is that humans are s
    • Best tool for the job is my attitude.

      Absolutely.. but that's something you can only really know if you use all the tools and then appraise them afterwards. If you're looking for a technology to run a 400 desktop network that would be a bit cost prohibitive. TCO studies are meant to aid people in making decisions prior to spending their money.. you can't find out which tool is best until *someone* has done that.
    • In the end studies are only worth something if they can be verified and are accurate with a sample that has as few biases (sp?) as possible.

      As Stuart mentions:

      Questioning the credibility of ALL analyst studies is important.

      If your claim was true, then History would be a worthless subject as it's almost impossible to get unbiased historical records. Yet it continues to be able to reconstruct the past in surprising detail, in part because the bias lets us know more about the different schools of thought

    • .. only worth something if they can be verified and are accurate with a sample that has as few biases (sp?) as possible.

      Nonsense! I believe biases are inherent and actually beneficial, as all opinions end up leaning to one side or the another anyways!

      Ok, break time over, now back to work on these pesky Diebold machines...

    • Best tool for the job is my attitude

      That's why I still have a win2000 partition for games and Photoshop. For everything else I do the available Linux apps are at least on par with those I can use with windows; but when it comes to games Linux can't (yet) compete with Windows, and I have no desire whatsoever to spend days mucking about with Wine to get a game 'mostly working'. I purchase games for entertainment, not to annoy myself with technical problems in just getting them to function properly.

      As for gr
  • TCP Studies (Score:4, Insightful)

    by EmagGeek (574360) <<gterich> <at> <aol.com>> on Tuesday February 21, 2006 @12:19PM (#14768886) Journal
    So, what we've learned about TCO studies in the last couple years is that they are absolutely meaningless. Whomever pays for the study gets to dictate the results of said study, plain and simple.
    • Re:TCP Studies (Score:3, Insightful)

      by onion2k (203094)
      Not at all. You can tell a lot from a study even if it's been commissioned by a particularly biased party. What features they highlight are likely to be well implimented and impressive, what features they skip over are likely to be very weak, if the commissioner of the study distance themselves from it then you can assume they're resorting to something underhand in order to promote their product .. if it was really good they'd not need to do that. Likewise if the study is especially negative about competing
    • Net effect, but... (Score:3, Insightful)

      by mengel (13619)
      This is actually due to the fact that if someone commissions a study and it doesn't come out the way they like, they just don't publish it...
    • Obviously another case where UDP shouldn't be used for Slashdot posting.
  • by brlewis (214632) on Tuesday February 21, 2006 @12:21PM (#14768895) Homepage
    Rob Enderle is telling the truth when he attacks the OSDL study on statistical grounds, but who would benefit from a purely statistical study? To tell whether Linux or Windows would have lower TCO for you, you look for anecdotal evidence from a handful of companies similar to yours. Once the sample size is large enough to be statistically significant, it will be too diverse to apply to your situation.
    • by flynt (248848) on Tuesday February 21, 2006 @01:07PM (#14769361)
      Not true. Statistical models (for example, a multiple linear regression) allow you to "plug in" values of the covariates for your particular situation, and get an estimate for your particular situation. I would say that your method of listening to companies in your situation is also probably very important. I disagree that TCO is somehow "non-statistical". This seems like a fine situation to use regression methods to predict TCO given several covariates.
    • Statistical attack (Score:4, Insightful)

      by RealProgrammer (723725) on Tuesday February 21, 2006 @01:16PM (#14769453) Homepage Journal
      I sort of had to chuckle when Mr. Enderle said programmers don't get taught statistics. I don't know about others, but I had to take three different stats courses, though I'd call only one of them truly rigorous -- the others were just business and social science courses, which focused primarily on survey design and analysis.

      But let's take his points one-by-one. Well, wait. He's done the typical trick of repeating several points with different wording, using fancy pseudo-academic fluff words. I'll lump them together:

            1. Population tested is unclear (IT organizations, generic respondents, CIOs and MIS Managers?)
            2. Sampling method is inconsistent, largely self-selecting, and inherently biased (i.e., not blind)
            3. Respondents are not of consistent level or responsibility
            4. vastly different locations and business types

      Using a diverse set of respondents is essential, ensuring "regression toward the mean". That's where your errors cancel if you set things up correctly. Typical MS-sponsored studies question a group of 30 or so people with the same kind of job at the same kind of company, which guarantees a sample bias. It's like asking a bunch of soccer moms whether kids get proper nutrution. They all have the same point of view.

      And I'm not exactly sure how you'd do a "blind" sample for a study like this. Survey ten thousand companies and randomly select the ones you'll report on? Not tell the respondents what the questions are about? Pfah. Either I am missing something, or Mr. Enderle is not being forthright.

            6. Results -- in terms of pricing -- are outside of survey scope
            7. No test ensures that respondents consistently ran both Linux and Windows-
            8. A large percentage of the statistical results were Linux only
            9. Conclusions are not supported by the data. There is very little Windows data, sparse enterprise data, but a great deal of anecdotal commentary.

      Guess what -- Windows pricing vs. Linux pricing is not difficult to analyze, and they aren't sold the same way. Also, Windows behavior is well-known to sysadmins, and the study was about Linux, after all.

      As for number 9, he's stating his conclusion as a supporting point, another neat trick. He doesn't like the results, so if the data aren't invalid (which he claimed was the case anyway) then the something must have been wrong with the way the conclusion was reached.
      • Using a diverse set of respondents is essential, ensuring "regression toward the mean". That's where your errors cancel if you set things up correctly.

        That should have said, "Using a diverse set of respondents is essential, so that your errors cancel."

      • Nicely put.

        Enderle is a joke. For anybody to take his criticism seriously, they have to be Windows shills - like him - themselves.

        I would agree that the OSDL report could have been more comprehensive in its specifics, but that's the nature of these things - they're meant to be read by idiot managers who can't think and don't care.

        Lke Enderle himself.
    • No there are statistical techniques for isolating variables. The idea would be you would end up with a formula like TCO(Linux) = f(# of employees, # of IT workers, total annual hardware expenses, % non hierarchical network communication....)
    • the Enderle effect (Score:3, Insightful)

      by burnin1965 (535071)
      " Rob Enderle is telling the truth when he attacks the OSDL study on statistical grounds,"

      Actually no, he is not telling the truth, he is merely following a predictable pattern of attacking everything open source.

      I find it quite suspect that someone can suddenly doubt the integrity of an organization when it appears they have always had a rather low opinion of the actions, philosophy, and intent of the organization.

      Time and again Enderle has suggested that the open source movement is disingenous and ran by [technewsworld.com]
  • by CyricZ (887944) on Tuesday February 21, 2006 @12:37PM (#14769078)
    There's no point in trying to get people to convert to Linux, or any other open source project.

    The best thing to do is to use it yourself. I've helped various businesses out with OpenBSD installations, and the results have been spectacular. Of course, it particularly shines on web servers and as a firewall, due to its very high level of security. Not only that, but the care put towards security translates directly into higher quality code, which leads to virtually no crashes.

    So while other businesses are fidgeting around with Windows Server 2003 or XP (and the massive hardware requirements needed to run such systems), and running into weekly network infections and other issues, others who know of better systems won't have such problems. They'll be able to run their servers on more economic hardware, using far superior software.

    It's most often small businesses who are most responsive to using technology like OpenBSD and PostgreSQL, rather than Windows 2003 and SQL Server. And they're the ones who can gain the most from it, as well. Little makes the owner of a sporting goods shoppe happier than telling him you can set him up an OpenBSD web server for his online store at a mere fraction of the cost of a Windows-based solution. Even saving £500 for such a person will make them very pleased. And when their system doesn't crash every other day, they'll be even more impressed.

    • The best thing to do is to use it yourself.

      Often, especially in business settings, you have to convert others in order to be allowed to use Linux. For all its improvements on the desktop, the place that Linux really shines is still the server room. Since servers are shared infrastructure, any changes to the servers usually involves getting the okay from upper management. In order to get this permission, you have to be able to say that Linux is going to save the business money over the long run, not ju

      • In my experience, users who end up using Linux or BSD in a work environment are often quite pleased. As long as their system works, they're more than willing to adapt to it, and likely will find themselves working far more effectively.

        In the past, I've set up desktop networks running OpenBSD. KDE can be configured to look and feel very much like Windows, if that's necessary. But after a day or two of using even a default KDE installation, many people were more than comfortable. The massive boosts in product
        • All of your points are valid. However, the point remains that its a lot easier to see these benefits after the implementation than before. If one is trying to convince a reluctant manager who has been using Windows products throughout their career, having some 3rd party sources to back up your word can only be beneficial.

    • I use Linux.
      The problem with Linux and OpenBSD on servers is that the bar for a Windows admin is much lower than a Linux one. For Windows you can go to any community college and take a course and then "admin" a windows server. Will it be a secure? Will it be stable? I really doubt it. With Linux you still have to have a clue and be willing to study on your own. Linux admins are just not as common as Windows admins even if the Linux admins tend to be better.
      The problem with Linux on the desktop is that EVERY
    • True enough, don't try to change anyone else's os.

      but refuse to touch any machine not running linux.

      I do. [ but then I'm lucky, my own company and I refuse to buy any ms product, or any software that is designed to run on ms product ]
  • I'm president of the Computer Club here at Grand Rapids Community College. We have several Linux users in the club, including myself and the Vice President. However, we don't advocate Linux to the student population.

    You see, we don't have time to support it. To borrow from Star Wars: There are always two, a master, and an apprentice. And the master better have time for all the apprentice's questions.

    Getting someone into Linux is a big time investment. You become there "go-to" guy, they guy they turn to when they don't RTFM, when they RTFM, but don't understand it (Try "man 7 regex"), and when they totally borked their documentation resources.

    I know. Years ago, I was the apprentice. And my "master" seems to have been on a sabbatical since then.

    I'm taking Calc 2 this semester. If I wasn't, I *might* have time for one or two Linux apprentices. We'll see what happens this summer.
  • Huge error (Score:4, Informative)

    by davidstrauss (544062) <david&davidstrauss,net> on Tuesday February 21, 2006 @12:42PM (#14769130)
    Microsoft ISA Server isn't a web server. It's a firewall and routing server. Microsoft IIS, the web server, comes free with every copy of Windows Server. You can also run Apache on Windows for free. This alone is a $24,000 mistake. MySQL and Eclipse also run on Windows. It seems like they chose the most expensive Microsoft-only configuration and even took some liberties there. How am I supposed to trust this study?
    • Re:Huge error (Score:4, Insightful)

      by Shimmer (3036) <brianberns@gmail.com> on Tuesday February 21, 2006 @01:36PM (#14769656) Homepage Journal
      You're right that ISA Server isn't a web server -- that is a huge mistake and does indeed call into question the credibility of the study.

      However, I don't think Apache, MySQL, or Eclipse (rather than IIS, SQL Server, and Visual Studio) are very common in the Microsoft world. I think it makes good sense to compare OSS against a Micrsoft-only configuration.
    • okay. iptables. and there's some routing software in fedora too, if you need it. so what? Oh horrors. You might have to learn some arcane language to implement rules. Guess what, if you don't know anything at all, the question "Do I let port 137 leave my LAN?" is equally hard to answer no matter what firewall you are using.
  • by morgan_greywolf (835522) on Tuesday February 21, 2006 @12:54PM (#14769254) Homepage Journal
    TCO will always vary based on what your application stack looks like. One thing we're studying where I work is the TCO of moving a mix of COTS and custom developed code from Solaris to Linux. Most of the rest of the company is on Linux, but this one particular application stack has for years ran on Solaris and is only now becoming feasible to run on Linux, since our major COTS vendor for this stack has ported their code to Linux. But because of the massive amounce of customized code in the application stack, moving the entire application stack to Linux doesn't look like the best way to go ... it's looking more and more like moving only some pieces of the stack to Linux and keeping some of the pieces of the stack on Solaris may look like the best approach. My point is that TCO will vary widely from application stack to application stack and ultimately whether Linux or Solaris or Windows is cheaper depends on a lot of factors that have nothing to do with the OS platform. (Oh, and BTW-- you CAN embed an existing PDF into an OpenOffice document on Linux, it's just not very straightforward. Doing so involves converting the PDF to a PostScript or EPS document, or into multiple PostScript or EPS documents, depending on the number of pages of the original PDF.)
  • Bad statistical reports are commonly used to manipulate people, most notably, in politics, but it can happen in a wide variety of consumer venues as well. Don't take any report from anyone at face value if you're going to depend on the results. Look underneath and behind the results in every case, and make sure you aren't being played.

    There's a case of a pot calling the kettle black for yah...

    • "There's a case of a pot calling the kettle black for yah..."

      True, but OSDL, as the open-source kettle, should not have released a study like this and blackened itself. Many of the criticisms levelled at the original MS-funded study can be levelled right back again at this one, and the OSDL should have seen them coming.

      Never wrestle with a pig. You get covered in dirt, and the pig enjoys it.
  • >First, be aware that programmers probably don't have to study statistics and marketing studies in school like business majors do

    Thats right. Thats sooo right. I mean, I needed higher level math to get into university, and then 3 years of advanced math to graduate in engineering.

    • by Anonymous Coward
      BS.

      Well, maybe the business students take marketing studies classes, but I call bullshit on the statement about statistics classes.

      Where I went to school, the statistics classes offered to satisfy the business majors were more oriented towards psychological and marketing purposes than the other classes, such as a Mathematics, engineering or CS major would have taken. In one upper level undergraduate class I took there were two business majors at the start of the class, and they both dropped it in the first
    • I agreed with the article that the report had weak statistics- but for him to suggest that business majors know more about statistics that programmers is insulting. If anything I would expect that the mathematical background of programmers would be higher than that of business majors (but then, I majored in math and computer science, so I'm probably biased).
  • From the study:

    "This study did not set out to compare the TCO of Linux directly against the TCO of Windows, but concentrated primarily on management effort in the Linux environment, particularly where sophisticated management tools were in use. Where the data is available, however, the analysis highlights general comparisons between the two environments."

    The study did a good job of what it set out to do. It did not set out to be a TCO comparison, and is not a TCO comparison other than in an anecdotal/gener
  • First Question (Score:1, Interesting)

    by Anonymous Coward
    I'm now getting my masters at George Mason University and I don't believe there's a single Linux machine on campus. In fact, the whole Computer Science department has only two Sun servers to offer me an account on! Everything else is Microsoft!
    I'm a freshman at GMU and computer labs dual boot Windows / Redhat. The clusters are Tru64 Unix. In the main building, the computer are running Sun operating systems.
  • Beating a dead horse (Score:3, Informative)

    by TheSkepticalOptimist (898384) on Tuesday February 21, 2006 @01:49PM (#14769797)
    TCO estimates and debates are a complete waste of time. Either your trying to promote your product over some other product, or your biased against the product your testing. TCO's are inherently flawed "faq's".

    The bottom line is, with Open Source software, while the source code can be inexpensive or even free, whether or not you have the write skill set to make use of it will determine your TCO.

    Have a staff of people that are not skilled with OOS and your TCO will skyrocket with long development cycles (unless your getting your OOS work done on a voluntary basis). Switch to OOS solutions and if your IT department isn't skilled, your TCO will skyrocket because of flawed implementation or not recognizing proper "free" solutions to implement, thus costing your company money to resolve the issues.

    Same goes on the retail software side. You can't expect retail paid for software to work miracles either, unskilled staff will drive up prices of both OOS and retail software/technology solutions.

    I just grow tired of the whole, "My product is better then yours because it has less TCO", mentality. Anybody TRULY objective about TCO estimates will realize that you can't look at nice, trite, situations where the solutions work flawlessly with skilled staff. Any OOS proponent that claims OOS will always cost less then retail software is just deluding themselves and others. Same goes for retail software proponents.

    In the end, the bottom line is, if you find a solution that works, stick with it. I find too often that companies paying for retail software/technology licenses suddenly decide to switch to Open source solutions, thinking it will save them money and they will get the same quality of service. If their retail solutions ARE working, and they are not affected by the cost of licensing retail software, DON'T SWITCH! The moment some CEO decides that it is time to switch to Open source solutions in order to make a few extra million on top of their already profitable business, then that is when companies fail.

    If your in the process of looking for a solution because your existing solution is costly to implement AND isn't meeting your needs, then by all means look at ALL solutions, both open source and retail.

    The last thing that TCO reports do is given management a false sense of security when moving to some solution just because some biased 3rd party wrote up a convincing argument that suggests it is cheaper to use their solution. TCO reports are inherently flawed, don't make the mistake of implementing a solution on cost alone.
    • I'd have to agree with you about what is best varies depending on your situation. My work uses a combination of Open Source software, Windows products, third-party vendors and software we created. We try to go with whatever will be best for us as a company- which means if there is a good open source (or freeware) solution that is the way to go. If there's not, we can either make it ourselves or try to buy it off someone else. There isn't yet a good open source solution for every problem.
  • Hardware drivers (Score:3, Insightful)

    by smitty_one_each (243267) * on Tuesday February 21, 2006 @02:03PM (#14769926) Homepage Journal
    It seems that video and wireless network interface drivers are two of the stickiest wickets for the desktop, irrespective of OS.
    Under my WindowsXP boot, my Netgear card died, after configuring WPA-PSK against the Linksys router, with a wpa_supplicant error.
    Can't run CivIV at an interesting resolution with any amount of eye-candy, or the nVidia driver craps out.
    Clearly, these are multi-vendor problems.
    What I hope to see from OSDL, and I think the Linux kernel community is driving towards, is more advocacy for open drivers. I don't know whether it's all intellectual property crap or what, but the peripheral vendors, as a whole (and without stooping to finger-pointing) come really close to cartel-like behavior that achieves little besides disadvantaging the end-user.
    Is there a way to promote a hardware compatibility database on http://www.osdl.org/ [osdl.org] that focuses strictly on chipsets, standards, specs, and test suites so vendors are positively encouraged to sell gear that doesn't suck?
  • im studying physics at mcgill university. our expermintal methods class requires us to use linux. we code in pico and compile with a linux prompt. i was a little surprised with the first question. anyone else required to use linux for school?
    • My college, the University of Missouri-Rolla, requires some (not much) linux familiarity for all its computer science majors. I think all CS degrees should require some experience in both Windows and Linux based enviroments (who knows what you'll need to use once you get a job- we use both every day at my work).
  • People who use Linux, like it. Or probably more accurately, people who like Linux, use it.

    Yeah, that's real informative.
  • by brokencomputer (695672) * on Tuesday February 21, 2006 @03:19PM (#14770585) Homepage Journal
    I'm now getting my masters at George Mason University and I don't believe there's a single Linux machine on campus. In fact, the whole Computer Science department has only two Sun servers to offer me an account on! Everything else is Microsoft!
    All of the Computer Labs in the Science and Technology buildings at George Mason dual boot WinXP and Red Hat Linux. However, Windows is usually what you see when you go into the computer labs since most people don't know how to reboot the computer and select a different operating system from GRUB.
    • Sounds like you need to make the default OS Linux in the bootloader. So then they won't have to figure out the complicated operation of rebooting and selecting "Linux".

      Just make sure the install is slick and has all the tools (OO, Firefox, GAIM, etc) that the user expects and in a way they can find it, and you might have some impressive results.

      And oh, "Redhat"??? Is that actually commercial stuff from the last few years, or are the Linux installs really that old? Sounds like it is time for an upgrade to
  • Because he had to combine his file with an existing PDF (not available with Open Office today)
    Combining PDFs is easy with pdfTeX. Converters from OO -> TeX exist.
  • I used to be really big in advocating that everyone I know switch to Linux, and looked forward to the day when Linux was the dominant desktop OS, and when Microsoft filed for bankruptcy.

    Over time, though, I started to question why it mattered. I prefer to work on a Linux platform. But what did it matter if someone else preferred to work on a Windows platform? What did I gain by getting people to use Linux?

    I still don't really understand the motivation. Personally, I decided that not only did I have nothing

If money can't buy happiness, I guess you'll just have to rent it.

Working...