When it comes to providing Internet services, Linux again easily beats the more commonly used scholastic systems. Setting up Linux systems to provide email accounts, an NNTP news server, a Web server, file and print services, and firewalling, proxy, and IP masquerading is fairly easy, and certainly rewards the effort. Given that Linux is free, and can run well on older hardware, there is no reason why numerous severs couldn't be set up. For many high schools, creating these systems could be part of the learning experience itself. The benefits could be felt in many areas, including mailing lists for staff and student organizations, for instance.
So far, we've talked mainly about network infrastructure. We all know that Linux excels there. But what about the systems that students, teachers, and staff actually work with directly? These systems have mostly been Macintosh and more recently Microsoft Windows machines. How does Linux fare in that niche?
Unfortunately, not too well at the moment. There are a number of projects working on addressing deficiencies in application software for Linux, but few of them have software currently ready for wide-spread deployment. However, work is progressing rapidly on many of these programs and you can expect to see useful Linux software for schools shortly.
On the administrative side, we've heard teachers say that if there isn't a good Linux gradebook program, Linux will never get onto teachers' desks. Period. Luckily, some of these teachers are also Linux fans and are working on programs that have the features they want rather than those that software companies think will sell. There's K12Admin, an account administration program written and currently in use at the Coast Mountain School District in British Columbia, Grader, a GNOME gradebook program being written by several teachers and students, KTeacher, a KDE gradebook being written by another teacher, and Roster, an account administration program written and used at Laney College. In addition, OpenClassroom, the Authenticated User Community (AUC), LearnLoop, and Zschool all intend to provide Web-based educational "environments" with integrated administrative and collaborative capabilities. Work is also beginning on a general testing and quizzing application that can be run by itself or integrated with some or all of the environments mentioned above.
There are two major features of Linux and open source software in general that will never be matched by commercial software: free availability and openness. We'll examine both of these features as they pertain to education.
Linux and Open Source software have a big advantage over anything else when comparing the cost of implementing comparable systems. Everything else in general use in education is sold with either site licenses or per-seat licensing. Either of these licensing options quickly becomes quite expensive for any school that wants to have more than a few computers in a computer lab that students can only use occasionally. Since Linux has no licensing fees based on the number of users or systems on which it's installed, a school can set up as many Linux systems as it wants from just one set of CD-ROMs. Indeed, if a school is really strapped for funds, it can download Linux from the Internet for only the cost of the connection, or buy CD-ROMs without paper documentation for US$2-7. And since Linux is comfortable and usable on systems on which Windows would be unforgivably slow, it can extend the lifetime of current systems, saving yet more money. A number of teachers have run cost analyses on using Linux versus something else (typically Microsoft Windows), and the differences are compelling. Here's one by David Culp: Getting Linux Into The School.
There are, of course, hardware limits for Linux, too. In schools this is probably seen most often in the size of the hard drives older systems have. It would be difficult to shoehorn a usable Linux distribution and application software onto a drive of 500 MB or less, but many systems in schools fit that criterion. Fortunately, the X Window System gives us another way to extend the usable life of these older systems.
The X Window System separates where a program displays from where the program actually runs. A large program, such as a Web browser or some other graphically intensive software, may be unbearably slow on an older machine, but it might run very nicely on more modern hardware. The X Window System allows the large program to actually run on a single, shared machine so that it can take advantage of the processing power and memory of this machine, and display very responsively on the many older machines, turned into XTerminals, that the students can use. All of this happens transparently, so the students do not know that the software is running elsewhere, they just know that the program runs well on their machines.
Another benefit of the XTerminal-based approach to classroom computing is the simplicity of the individual workstations. Instead of having to manage a large number of individual workstations, a single setup can be created for the simpler hardware. This reduces the amount of technician time spent maintaining the workstations, as well as reducing the number of technical problems that require the technician's intervention during class time. In addition, it's harder for a user to intentionally or unintentionally mess up the XTerminal's configuration and easier for a technician to restore a generic XTerminal setup. This can be a major advantage in the constant struggle to keep all the systems in a school computer lab working and available for student use.
A further benefit of XTerminals is that they can be created from very different computers and still provide the same user experience to the students. A student lab of XTerminals could be made up of older 486 PCs and 680x0 Macintoshes; as long as they were all running as XTerminals with their programs running on the same application server, the students would see exactly the same screens.
The other advantage Linux and open source software in general has over the commercial alternatives is that of openness. All of the commands that make up the programs are completely visible, so these programs are uniquely available for students to study to understand how such software works. In addition, the complete openness of the software makes it possible for individual students or schools to modify it to suit their needs, or to suggest changes to the creators and maintainers of the software. This capability is the major reason for one of the best features of open source software -- the extremely quick discovery and repair of bugs. In many ways, this openness is analogous to the academic concept of peer review. When everyone can see and critique the work people do, flaws tend to be discovered quickly and remedied equally quickly.
So far we've talked about Linux's utility for running servers and administrative systems. We haven't addressed its usefulness on the desktop, the place where Linux perhaps lags behind other OSs. However, there are many efforts to bring Linux up to and past the others in the educational field. In addition, many of the perceived shortcomings disappear when examined more closely.
As in many other areas, in education you may hear that there's no standard office software for Linux. This is usually code for "Microsoft Office isn't available on Linux." However, there are many alternatives that will work just as well. We all know them, so there's little point in listing them. The hard part is persuading teachers and administrators that they shouldn't be teaching students how to use Microsoft Office (or any specific product), but rather should be teaching the concepts behind word processing and spreadsheet calculations, etc., in general.
One of the biggest shortcomings for Linux in the educational field is pedagogical programs, the sort of educational programs often called "courseware." Courseware is almost always produced by small companies that have few resources to use in porting their programs to OSs for which they don't see any market. There are a few ways we can address this. First, we can develop Open Source courseware that runs on Linux. There are some efforts being made in that area already. Second, we can identify Linux software that would be usable as courseware with proper lesson plans, and develop such lesson plans. We've started to identify such programs, but we haven't yet started work on lesson plans. There are organizations that develop lesson plans and post them on the Internet for any interested educator to use, and we hope to collaborate with them on this. Third, we can advocate Linux ports of existing courseware. In many cases, tools exist which can greatly facilitate the porting of existing software; we describe this in a commercial port advocacy mini-HOWTO that should help anyone interested in making such an effort. It's available from the LDP or from SEUL.
Yet another way to address this lack is to provide authoring tools that teachers can use to create their own courseware. Examples of this sort of software are Apple's HyperCard, HyperStudio, and [shudder] Microsoft's Visual Basic. These all provide occasional users with the ability to create courseware with personalized content.
One feature of all of those programs is the visual nature of the program development. This kind of interface is much less imposing than the traditional IDE we're used to. Because its primary users are comfortable with the more traditional interface, Linux is a bit short on the visual sort of tools. We know of two that are arguably ready right now, and one for which we have great hope.
MetaCard is a commercial program that was designed for corporations to use in creating computer-based training. It runs on Linux, Windows, and MacOS, and can read HyperCard stacks and convert them into its own format. MetaCard makes a limited version of the program available for free download, and has very favorable educational licensing terms for K-12 (non-collegiate) educational institutions. In addition, the MetaCard interface is written in MetaCard, so it could be modified to be even more appealing to the educational community.
Visual TCL is a similar program for creating Tcl/Tk programs. It's available for free download and could also be modified to fit the educational community's needs. It of course requires that Tcl/Tk be installed on the system.
Squeak is an effort to create a new SmallTalk programming environment. The Squeak development team is working on their user interface, Morphic, with the intent of making it usable for everyone from students in the lower grades to professional software developers.
So far we've talked about what things are currently available that can help Linux make inroads in the educational community. What should we be working on in the future to make Linux the OS of choice for education, and what can we do now to promote its use?
One thing that we'd like to see is window managers and program interfaces that can be modified to be appropriate for people from kindergarteners to college students. Ideally, these would be very simple and easy to use (if limited) for the younger students, but could be gradually made more complex and powerful as the students' abilities increased. We'd like to see first and second graders working with simple versions of programs like AbiWord or the GIMP, and slowly moving into using the full versions.
There are a number of Web sites and mailing lists that discuss one or another aspect of using Linux in education. We can't list them all here (and would surely miss some if we tried); take a look at the worldwide links page at SEUL/edu for connections to a number of them. Some of these groups work on advocating Linux to schools or on developing software and documentation for scholastic use, while others focus on implementing Linux or various applications in specific schools or areas. Whatever your interest, there's probably a discussion group that addresses it.
A well-recognized problem for using Linux in education is the paucity of documentation on installing, configuring, and using Linux and most applications. We're working on defining a template for such documentation that will make it easier to create such documents and provide them in a variety of formats, along the lines of the Linux Documentation Project. You can take a look at a draft proposal. Once we get everything worked out, we hope to see many people contributing quality documentation that will help teachers and students use Linux.
One major need for Linux to be a viable option in most schools is quality local support. In the Linux community we're used to on-line support via mailing lists, newsgroups, and Web forums. While some schools may feel comfortable with only this, many will want to see a warm body they can feel is concerned with their local interests. The lucky schools will have a Linux fan and user on their staff already, but for the others that can be convinced to try Linux, outside support will be needed. There are commercial organizations that support Linux for a fee, but their prices will most likely be beyond what perpetually cash-starved schools can afford. What does that leave? LUGs (Linux User Groups). One of our big dreams is that LUGs around the world will create educational outreach programs where they adopt local schools or school districts and provide the maintenance and support the schools will need. This would be beneficial to the LUGs too, of course, since they'd be creating enthusiastic new users by making Linux work smoothly for the schools. A local LUG/school relationship could make for both stronger LUGs and stronger schools, besides making Linux advocacy easier.
Linux has some obvious advantages for education, but also some obvious shortcomings. Our job is to fix the shortcomings while making sure the advantages are shouted to the world!
Doug Loss (email@example.com) is the Data Network Coordinator (with a side helping of information security administrator) at Bloomsburg University in Pennsylvania. In his copious free time, he leads the education section of SEUL (Simple End User Linux), writes weekly Linux in Education reports, and stirs various pots around the Internet. In the time that's left, he reads hard SF, plays board wargames, and putters around the house at his wife's insistence. He is survived by a wife and an 8-year-old son.
Pete St. Onge (firstname.lastname@example.org) has been involved with the Simple End User Linux education project, SEUL/edu, since its inception almost two years ago. Prior to returning to studies to complete an M.Sc. in Biology at McGill University, Pete St. Onge taught Operating Systems and Applications as a staff instructor at the college level.
The Simple End User Linux (SEUL) project is a volunteer project to help Linux become an operating system that the average user can install and operate easily. SEUL is currently working on the non-technical aspects of bringing Linux to end-users -- Linux in education, advocacy documents, managing and coordinating communications between projects, coordinating feedback from end-users, and hosting related development projects. The end goal of SEUL is to have a comprehensive suite of high-quality applications (productivity applications as well as leisure/programming applications) available under the GPL for the Linux platform, as well as a broader base of educated users around the world who understand why free software is better.
We're eager to find people interested in writing editorials on software-related topics. We're flexible on length, style, and topic, so long as you know what you're talking about and back up your opinions with facts. Anyone who writes an editorial gets a freshmeat t-shirt from ThinkGeek in addition to 15 minutes of fame. If you think you'd like to try your hand at it, let email@example.com know what you'd like to write about.