Perceptions of the Linux OS Among Undergraduate System Administrators
At the start of this academic year (September 2001), I was asked to teach a new module in my Institute's B.Sc. degree program in Information Technology. This final-year undergraduate module, entitled “Network and Systems Management”, covers a wide range of system administration technologies, practices and principles. In effect, students of the module are the system administrators of the future.
As is probably the theme at the majority of third-level educational establishments, student's exposure to OS technology at the Institute of Technology, Carlow is Microsoft-focused and desktop-based. This is easy to understand, for the desktop is very much a Microsoft stronghold, and if an institution can use the same PCs to teach business undergrads Excel and science undergrads programming, then they will. However, what many of my students often fail to recognize is that, as system administrators, they will find themselves managing servers running OS technology other than Microsoft's.
So in an attempt to expose my students to a more realistic view of the technologies in use in the real world, I try to deemphasize Microsoft's technologies in favor of the alternatives. As you can imagine, Linux features quite heavily.
At the start of this academic year, I informally surveyed the 31 students enrolled in the module about their exposure to Linux. Most (if not all) had some exposure to the OS. I probed further and asked how many students had used Linux as the basis of their third-year project (the previous year). One or two hands were raised. Then the first shock came: someone blurted out, “nearly everyone who used Linux last year went on to fail their project”. It came out that a number of individuals were missing from the final year due to failing the project element in year three. When I probed for the root cause of the project-failing problem, I got my second shock: “Linux is too hard to install”. I was shocked not because these two statements were necessarily false but because these 31 students had pretty much convinced themselves that success was tied to Microsoft and failure to Linux.
While I covered Windows 2000 and Linux as case studies, pointing out the advantages and disadvantages of each OS, I gave the class an assignment that would require them to do some simple research and, as a consequence, allow them to learn a little more about Linux. The task was simple enough. I stated: “Despite considerable success as a server platform, Linux will never threaten Microsoft Windows as a desktop operating system.” I asked the students to research the subject area, form an opinion as to whether they agreed or disagreed with the statement, and then present their case in no more than three A4 pages of typed text. As I marked their assignments, a number of themes recurred. Additionally, numerous myths became evident, and—perhaps not unexpectedly—a number of blatant untruths presented themselves. These I classified as clangers. In the remainder of this article, I present the themes, myths and clangers uncovered, in addition to my own personal commentary.
Note: for this purpose, I define a theme as something that is generally agreed to be true. If a comment occurred repeatedly throughout the students submissions, and it was true, it became a theme. A myth is defined as something generally held to be true but is, in fact, not true. Even if a myth occurred repeatedly throughout the submissions (and many did), it can't be a theme as it isn't true. A clanger is a statement that is just blatantly wrong.
The majority of my students felt that “more desktop applications are required for Linux”. No argument here, the more the merrier. And, Microsoft obviously has a distinct advantage in this regard. This theme appeared in many different forms in the student submissions. The most depressing (but still true) form was: “The average user does not care what operating system they are using, just so long as it runs Microsoft Office.” And Microsoft knows this. The real crown jewel in the Microsoft arsenal is the Office Suite. The fact that Redmond and Cupertino engineers have already ported (most of) the Office technology to Mac OS X indicates that a port to the X Window System would not be too difficult. But let's face it, porting to Mac OS X on the PowerPC-based Macintosh will never directly threaten the Windows monopoly. Porting to Linux on x86 is an entirely different matter. Were this to occur, the implications would be huge. This theme was further generalized by one student as follows: “The desktop operating system with the most third-party software wins.”
The students felt that “The KDE/GNOME choice confuses most newcomers to Linux.” This frustration was also expressed as follows: “A commonly-agreed upon GUI environment is needed.” Most felt Microsoft has a definite edge here, as Windows 9x/ME/NT/2000 and now XP look essentially the same. There's an argument that the choice of GUI (or desktop environment) is a good thing in the Linux world. However, I'd have to side with my students on this one, as I'd really like to see one single, coherent GUI environment combining the best features of both KDE and GNOME. There is nothing inherently wrong with all Linux desktop GUIs looking the same, is there? And I suspect such an environment would be welcomed by the vast majority of Linux GUI trainers.
The students had plenty to say about the (lack of) reliability in Windows. An eyebrow raising comment said, “for the sake of convenience and familiarity, most users will put up with Windows crashing on a regular basis. In fact, everyone knows it's quite normal for PCs to crash.” A more general observation, along the same lines, was, “People like predictability, and they don't like change, so they will put up with Windows' shortcomings.” This is a shame but it is true: it has become okay for a PC (running Windows) to crash once a day (or more often). One student referred to this as “normal” behavior. Like it or not, the average user expects their PC to crash and are trained to switch it off then back on again.
Compared to the infamous reliability of Windows, Linux did well: “Linux is technically superior to Windows: it runs longer and consumes fewer resources. Linux also has better security, stability and scalability.” No argument on this front from me, either. Unfortunately, the PC world is littered with dead technologies that were technically superior to the alternatives available at the time of their launch. Or perhaps I should have said “dead companies”. Of course, it is not a company (like Netscape) that Microsoft is trying to kill with its attacks on Linux, it's a community (which is a little harder to kill). So, Microsoft's past tactics may not (hopefully, will not) work.
A number of students highlighted the market perception of Linux as a problem to be overcome: “How can Linux really threaten Windows on the desktop when the vast majority of PC users haven't even heard of it?”. Another slant on this was, “The Windows brandname is too strong to threaten”, and “The Linux community are no match for the marketing machine that is Microsoft.” This visibility problem isn't helped by the fact that the mainstream computing press have all but stopped covering Linux since the dot-com bubble burst. The Windows brandname is as strong as Coca-Cola, but the Coca-Cola brandname didn't stop Pepsi from having a go (and doing quite well, too). Again, I think the strength of the Linux community has bearing here, despite the fact that a lot of my students thought that “which desktop OS dominates has more to do with marketing than technical expertise”. Nearly every student agreed that “Linux needs to shake its image as the techie/programmer's OS”, and that “Linux is seen as a geek's OS. Programmers love it and that puts everyone else off.” Yes, image (market perception) is everything, and Microsoft knows this. This helps explains the anti-Linux FUD campaign coming out of Redmond these days.
The fact that Linux tends to run well on any old PC came in for praise, typically as follows: “Newer versions of Windows tend to obsolete todays hardware. Linux, on the other hand, runs quite well on older PCs.” Yes, the new version of your chosen operating system shouldn't require a major upgrade (or replacement) to the hardware it runs on. If only more people would realize this, and act on it.
More than one student had this warning for Microsoft: “The new XP licensing arrangements may result in many IT shops reassessing their allegiance to Microsoft. Coming on the heels of the recent economic downturn, this may hurt Microsoft to the benefit of Linux.” Yes, we should all be screaming this from the tallest buildings we can find: users (i.e., IT managers) need to resist Microsoft's attempts to “lock 'em in” as much as possible!
|Non-Linux FOSS: libnotify, OS X Style||Jun 18, 2013|
|Containers—Not Virtual Machines—Are the Future Cloud||Jun 17, 2013|
|Lock-Free Multi-Producer Multi-Consumer Queue on Ring Buffer||Jun 12, 2013|
|Weechat, Irssi's Little Brother||Jun 11, 2013|
|One Tail Just Isn't Enough||Jun 07, 2013|
|Introduction to MapReduce with Hadoop on Linux||Jun 05, 2013|
- Containers—Not Virtual Machines—Are the Future Cloud
- Non-Linux FOSS: libnotify, OS X Style
- Linux Systems Administrator
- Validate an E-Mail Address with PHP, the Right Way
- Lock-Free Multi-Producer Multi-Consumer Queue on Ring Buffer
- Senior Perl Developer
- Technical Support Rep
- UX Designer
- Introduction to MapReduce with Hadoop on Linux
- RSS Feeds
Free Webinar: Hadoop
How to Build an Optimal Hadoop Cluster to Store and Maintain Unlimited Amounts of Data Using Microservers
Realizing the promise of Apache® Hadoop® requires the effective deployment of compute, memory, storage and networking to achieve optimal results. With its flexibility and multitude of options, it is easy to over or under provision the server infrastructure, resulting in poor performance and high TCO. Join us for an in depth, technical discussion with industry experts from leading Hadoop and server companies who will provide insights into the key considerations for designing and deploying an optimal Hadoop cluster.
Some of key questions to be discussed are:
- What is the “typical” Hadoop cluster and what should be installed on the different machine types?
- Why should you consider the typical workload patterns when making your hardware decisions?
- Are all microservers created equal for Hadoop deployments?
- How do I plan for expansion if I require more compute, memory, storage or networking?