By Mick O’Neil
When Apple introduced the successful Apple II series, the company also went a long way toward defining what we mean by a workable personal computer. For example, the CPU came in a box along with an expansion bus; it was accessible so that a user could make up for hardware deficiencies by modifying the mother board; cards containing new technology could be added to ‘upgrade’ the system; users could save data to a floppy drive; and peripherals could be attached to expansion cards via punch out tabs in the rear of the computer. In addition, some software was included in ROM (Read Only Memory) to allow the computer to ‘boot’ and read information from a floppy disk. When IBM released the PC a few years later, the hardware was a bit more sophisticated, but the model was pretty much the same.
In some ways I was a ‘power user’ in those days. That is, I expanded my Apple II+ to include: a Keyboard Enhancer, to give lower case characters; a Z80 coprocessor to run CP/M; a Videx Ultraterm Card to display 132 characters across the screen; a Saturn 128K Memory Expansion Card at a cost of several hundred dollars; a PKASO graphics printer interface; an accelerator card; and probably some other gadgets that I simply can’t recall. Then, the personal computer was like a customizable shell presenting users with an incredible (well maybe that’s going too far) array of options. For the most part, expanding a system simply entailed opening the case and adding the new hardware.
With the introduction of the Macintosh, Apple changed some aspects of our view of the personal computer. The Macintosh was supposed to be the computer as an appliance. That is, everything that was needed was already on-board and any notion of modifying the hardware was quickly put to rest by the fortress-like case. In fact, on the original Mac, if you located a long enough Torx screwdriver and managed to open the case, you were still required to cut a resistor to get the system to recognize added RAM. Uniquely, much of the guts of the system software was in ROM, allowing the introduction of the Mac’s renowned Graphics User Interface (GUI).
Regrettably, Apple kept the hardware and software proprietary and at the time there seemed little incentive for the industry to support a new standard. Despite the ‘mini’ Mac’s eventual (and inevitable) success, the ‘toaster’ approach eventually gave way to ‘pizza box’ and ‘tower’ designs supporting the earlier Apple II paradigm with Processor Direct, Nubus and PCI slots.
The reason why the expansion issue is important is that it suggests how we view the personal computer: that is, the computer as a personal power center that users should upgrade on the fly. The two-speed built-in CD ROM drive should be replaced by a twenty-four speed. The 1.2 gigabyte hard drive should be supplemented with additional drives in the tower’s peripheral expansion bays. The processor should be replaced or the cache expanded or more RAM added. Similarly, Windows 95 should sometime be upgraded to Windows 98, PageMaker 6.0 to 6.5, and so on, with little real analysis as to the ‘whys’ of the upgrades.
Unfortunately, the personal computer paradigm has been rendered practically inept by the sloppy implementation of system software. The earlier Windows systems that balanced carefully on a DOS tightrope could not be easily upgraded. This became evident to thousands of users who tried to add a CD-ROM drive to older systems. Windows 95 was supposed to be easier with its ‘Plug and Play’ regimen, and sometimes, if you were real lucky, it was. Frustrated cynics, however, still refer to the Windows 95 system as ‘Plug and Pray.’ (My grammar checker suggested I change the last phrase to ‘Plug and Prey’ which is also not far from the truth.) Similarly, Windows NT, the hardware hog of the ’90s’ still has a difficult time recognizing some internal and external devices. When I suggested installing NT on a portable at a recent Windows NT training session, the instructor suggested that this was probably not a good idea. He noted that Windows NT could be pretty problematic on portables — even powerhouses like my Micron Millenia Transport.
Macintosh, by far the easiest system to expand, has been weighed down to an extent by the flexibility of its operating system. A system that once disappeared into the background to expose the user directly to applications power today often requires user intervention to modify extensions or preferences. Even the incredibly successful introduction of Mac OS 8 was marred by problems with the installation CD and the inability of the new operating system to recognize some IDE drives. As users, we spend far too much time attempting to deal with issues related to hardware and software integration and upgrading. We have to let go of the ‘personal’ responsibility for these systems because we can no longer afford the time and energy it requires. It’s time to ‘think different’ about computing and perhaps network computing offers a direction forward. Though the Network Computer (NC), in conjunction with high speed, remote subscription services, has immediate appeal in the home or at the office, an important testing ground for the concept will be in the field of K-12 Education. Here, a unique set of circumstances makes the NC an attractive alternative to the traditional PC.
The Educational Setting One of the Clinton Administration’s goals is to wire schools across the United States so that all students have equal access to the vast research facilities of the Internet. To an outside observer this means high-speed access from the school to an Internet Service Provider. To the school’s computer coordinator (if there is one), it means routers, school-wide networks, accessible network printers, and all of the administrative support that these entail.
In an ideal world with unlimited budgets and a fully staffed support structure, UNIX might be a viable networking option. In the real world of falling school budgets and few qualified support staff, school systems must presently opt for AppleShare, Novell, Microsoft NT, OS2, or a similar networking package. None of these offer UNIX-like speed and all will require a degree of on-site network support.
Given that school networks are an essential element in providing Internet access, it makes sense to revisit factors like expense, installation and maintenance, security, and examine how the Network Computer could make a difference.
Expense Schools are constantly in transition, buying faster machines with more RAM and larger hard drives to run more powerful software. They have no choice because, as the software publishers move to 32 bit software, the schools are forced to keep current. While caught up in this kind of high tech rat race, it’s easy to lose sight of the fact that the single most important application used in schools is the word processor and students will not write faster on a 300 MHz PowerPc than a Mac IIci. Though sufficient speed is important to retrieve sound, graphics, and movies from the Web and to process them on a work station, we should be cautious about defining both sufficiency and obsolescence.
Schools are sometimes forced to buy computers with features they will rarely use. It’s not as easy today to purchase customizable systems that meet a school’s specifications, and many school systems are forced to buy computers designed for the business market. In many instances, PCMCIA slots, built-in modems, internal CD-ROM drives, fancy sound cards, accelerators, and the like, simply go unused in the school writing labs.
Application software installed on work station hard drives is expensive to purchase and expensive to maintain. Though some software publishers offer educational discounts, schools often face the choice of paying far too much for network licenses or looking the other way while teachers install needed applications.
Installation and Maintenance Installation and maintenance of systems software are real problems in an educational setting. A teacher’s time is too valuable to spend trouble shooting a computer operating system as complex as Windows95 or NT. Installing system upgrades with all of the resulting confusion concerning software compatibility is also a waste of teacher time.
PCs and Macs are vulnerable to inadvertent errors that can cause disabling system crashes. Teachers will sometimes try to improve their systems by installing new hardware and software without fully understanding the impact. Something that works well at home may not work in a network environment. Windows is particularly vulnerable to these user ‘attacks’ and even the Mac can be sometimes problematic.
Security Security of work stations and networks in a school setting is an enormous issue. Whereas in a business or government office, the network administrator worries about protecting sensitive databases, school administrators also have to be concerned with the integrity of the system as a whole. Generations of student hackers try to demonstrate their prowess by disabling school work stations or networks, installing games on network servers, broadcasting obnoxious messages, hacking the supervisor’s password, etc., etc., ad nauseum. Security of work stations requires constant vigilance as long as workstations have modifiable systems software and/or floppy and CD-ROM drives.
Network security under Windows 95 and Windows NT is a nightmare. At a recent NT training session, I graphically demonstrated a breach in NT security and asked the instructor for a strategy to preclude this breach. He indicated that there was no foolproof system to secure NT workstations or servers. Windows 95 is even less secure. Networks are vulnerable to virus infiltration from floppy disks, CDs, and cartridges. As long as students can access systems with removable media of any sort, the network will be at risk as will individual work stations.
Think Different, Think NC Clearly, in the field of K-12 education, we have to rethink how we deliver computer services to kids. The closest we can come to the perfect educational computer system is a descendant of that ‘appliance’ that Apple introduced in 1984. The Mac NC could offer possible relief to schools in terms of expense, installation and maintenance, and security.
The basic work station will probably come without a floppy, CD-ROM, or hard drive and with much of the system software in ROM. The only expansion capabilities built into these systems might be to allow replacing the processor, adding RAM, and upgrading the system ROM along with a Firewire or SCSI interface for those instances when a peripheral might be needed. These stations could be priced much lower than the cheapest PC on the market and because of their low entry level price, school systems should be able to afford a whole lot of them.
There could also be other flavors of NC for different functions around the school. Graphics NCs could include internal hard drives to support Computer Aided Design or heavy-duty photo processing, while Multimedia NCs for the media centers might sport an internal CD-ROM. The real value in these work stations will depend to a large extent on the network server. A local Rhapsody-based server, for example, could offer UNIX-like speeds, network versions of ClarisWorks Office or Microsoft Office, and licensed third party applications. Because the NC networks could quickly grow to standardize school computing, economies of scale would allow software publishers to significantly reduce prices.
A network computer sporting a Mac operating system with server based suites of grade level software would be an educational computer coordinator’s dream. Most installation and maintenance would be done on the server, and without the security gaps provided by work station floppies and CDs, the network would become much less vulnerable. The NC could free the users to concentrate on information retrieval, processing, and publishing with the operating system once again slipping invisibly to the background — where it belongs.
My Perspective I have been working in educational technology for some thirteen years now for one of the finest school systems in the world. The Department of Defense Educational Activity is, in fact, spread all over the world with DODEA schools located in virtually every U.S theater of operations (including the United States). We are committed to providing whatever technology is required (and within our budget) to help all of our students succeed. Towards that end we have developed a flexible, vital set of technology standards that determine to a large extent how our technology program evolves. For further information about DODEA, please visit the DODEA home page at http://www.odedodea.edu. (Note: the opinions expressed in this article are strictly my own and in no way represent DODEA or the U.S. Department of Defense.)
Mick O’Neil (firstname.lastname@example.org)