Hi All, Great analysis Gustin,
As Szemir mentioned both he and I were there in the morning as well as a couple of CLUGites that I recognized. Overall it seemed that most of the presentation was aimed at either creating uncertainty and doubt OR that when comparing Linux to Windows it was "generally a wash". Both Szemir and I challenged Barnaby on a few points where he seemed to be content to back off or reiterate that "it was a wash". I do wish that some of the more learned and vocal members of LUG had attended. I am certainly not an expert on Linux or Linux advocacy and would have liked to see the comments had others been there. FWIW, Dave Watkins VP CLUG -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Gustin Johnson Sent: Wednesday, April 27, 2005 9:37 AM To: 'CLUG General' Subject: [clug-talk] MS Presentation Sooner or later someone was bound to start this thread, I may as well take the plunge. Below are some of my thoughts regarding the evening. I was at the evening event btw. What is written below is also the basis for an editorial I am writing, though it is far from being complete. Not really a big surprise was that I found the session to be 90% or more FUD. What was surprising was the lack of resistance. There were few people who raised any objections (myself included) during the course of the evening in spite of the strong CLUG presence. I will get back to this later. What this means is that Microsoft is now serious about "fighting" Linux/OSS, which on the bright side somewhat validates Linux and OSS as viable competition (and thus viable alternatives to MS and or other proprietary implementations). Most of the arguments presented were not platform dependant. It does not matter if I run Linux, windows, mac, Solaris, PDP11 etc, a developer needs to be aware of copyright, patent, and licence issues (he did stay away from the patent debate, at least for the evening session). The variety of OSI approved licences simply means that developers have the good fortune to choose a licence that best suits their needs and situation. What was presented was a scary plethora of OSI licences that served to scare the audience. What was not mentioned was that one needs to be even more careful when dealing with proprietary licences and technology, and that open source serves to promote innovation by attempting to remove restrictions on what one can do with the software, not the other way around. The other interesting aspect was that he generally used Red Hat and SuSe for the majority of his comparisons. These are commercial companies that are playing the same game as Microsoft, which is a tough game to play. They are essentially Linux vendors, a point that was hammered home several times throughout the evening. What was not talked about is that the game itself is changing. Do I really need a vendor with the likes of Debian, Gentoo, and Slackware? How about the commercial viability of companies like Ubuntu? The community provides a great deal of expertise to its members. Google is far more help in tracking down issues with Linux and OSS in general. When tracking down windows issues, Google seems to continuously direct me to experts exchange, which wants my money before they show the solution (provided by third parties). Where is the Windows Community that I can turn to when I don't have the answer. These are the ideas that need to be talked about, as this is where Microsoft competes poorly. When there are problems with the software I have (somewhat anecdotally) found that the community provided a faster and more accurate response. In many cases I can contact the developer directly, or at the very least the packager for my distribution. In either case I get to troubleshoot with the person who is most familiar with the package, not some call centre robot reading scripts from a cubicle in Bangalore, and usually for a dollar cost. On the cost front, he showed a pretty graph that outlined where the cost was. Hardware and software were really small, but were also nearly equal. He emphasized that staff and downtime where the greatest cause of TCO. This really has nothing to do with Linux or OSS either, as it applies to both. Incompetent people will cost an organization, regardless of the platform that company uses. Downtime will cost regardless of the platform. What he did NOT do was compare staffing costs, staffing competence between the two "worlds". He did not compare downtime between the platforms in ANY way. He did give some numbers on Server 2003, but did not give any numbers for RHEL and SuSe. What he also did NOT do was explain why software comprises a larger percentage than the hardware in the TCO (based on his own graph as well as experience). To illustrate simply price out the parts for a decent business machine (eg. no $600 video cards). You can get a very capable machine for ~$700. Why is it that when you buy windows XP pro, MS office SBE, and a virus scan product, your purchase price more than doubles. This does not make sense, yet no one mentions this during the presentation. What was surprising was that he told us about the Microsoft Lab that has competent people looking at various OSS technologies as well as various Linux distributions. This lab explains why he chose RHEL, Fedora, and SuSe for most of his comparisons. The presentation was designed to strike where Linux was weak, and in particular, where it tries to directly compete with Microsoft under the same rules. What this means for us is twofold. The OSS developers need to be better. Things are moving along nicely but there are still many issues. Microsoft is now hitting us where it hurts. Recent analysis of the Linux kernel demonstrated that for the numbers of lines of code, Linux was extremely well written, containing far fewer bugs than the average for proprietary software of comparable size and complexity. Even so, we need to be better. Secondly we need to be better with getting our own message out. I am not entirely sure how to do this, but last night convinced me that Microsoft has found an effective method for them to combat Linux/OSS. It was not until I read my notes when I got home that I realized how slick and misleading that presentation really was. Even having taken some behaviour modification courses, I still fell prey to many of the techniques that were used. We need to be better at refuting the subtle FUD when we attend these kinds of presentations. We also cannot be combative nor can we appear to be a zealot. I believe that the great strength of Linux and OSS is the passion with which the people who use and BUILD these packages infuse with their work. We do not need Steve Balmer shouting over and over again that we should be excited about software development. We do however, need to match that passion with effective strategy so that we do not get outmanoeuvred and relegated to a niche market. As the VHS beta wars of years ago, Superior technology does not guarantee success. Cheers, -- No trees were harmed in the transmission of this message, however a large number of electrons were seriously inconvenienced. --------------------------------------------------- This message was sent using Echostar Secure Webmail _______________________________________________ clug-talk mailing list [email protected] http://clug.ca/mailman/listinfo/clug-talk_clug.ca Mailing List Guidelines (http://clug.ca/ml_guidelines.php) **Please remove these lines when replying
