Developing For The K 12 Audience
Posted on 08 Apr 2002
by David McCreath (mccreath)
Rated 4.27 (Ratings: 8)
- More articles in Site Development
What can a web developer expect when developing for the K-12  market? Is there a dominant operating system? What about older machines and browsers? What can you expect to find installed as student machines? As teacher machines? While I can’t provide national statistics, I am in the position of being able to provide some current statistics about the Anchorage School District, where I work as a web developer.
As with any other web project, when you’re developing a project for a K-12 institution, you have multiple considerations. We’ll look at the two primary aspects of the situation: infrastructure (consisting of desktop hardware, network capacity, and support capabilities) and legalities.
The Anchorage School District is one of the 100 largest districts in the US, in the most wired state in the country . We have over 90 schools and programs, just under 50,000 students, over 3,000 teachers and between 2,000–3,000 support/administrative personnel. We also have administrative and support staff scattered throughout the city in small clusters that add up to about 15 additional sites, all with networks and desktop computers, and connected to our District WAN.
Our platform mix is Mac heavy, and much of our hardware is very old. All told, we have roughly 18,000 computers  throughout the district (including student lab machines, teacher desktop machines, and support personnel machines).
Of those, the support/admin staff are likely to have the most up-to-date, which means anything less than three years old. Those machines are split fairly evenly between Dell and Apple. They account for about 3,000 of our machines. It is worth noting that in the last year, there has been a greater effort to move administrative staff to Windows. This has not occurred on the instructional side, with the exception of three schools (two high schools, grades 9 through 12, and one middle school, grades 7 and 8) that have converted to primarily Windows over the last two or three years. I’ll focus primarily on the machines in the schools for the rest of this article.
We have about 2,500 educational iMacs (266 MHz G3, 64 mb RAM, 6 gig HD, OS 8.6 [although some have been upgraded to OS 9]) that were leased from Apple for three years. Those are the most current machines in the schools, aside from a few admin machines.
The bulk of the remainder is made up of Macs older than iMacs. We have literally thousands of machines that can barely use Netscape 4.08, going all the way back to LCs, and several hundred older than that which can’t run a web browser, period. Here is a detailed list of computers currently in use in our schools, listed by number bought per year.
So what do those numbers actually mean?
We currently have 15,245 computers in the schools. Of the machines that we own outright, 7,048 of them were manufactured before the advent of the oldest G3 (1997), which is realistically the oldest machine we can expect to run OSX reliably. Further, many of these older machines are currently running System 7.5.5 and cannot legally be upgraded to System 7.6.1 because Apple charges for a full upgrade to OSX to make that switch, despite the fact that 7.6.1 is now over 5 years old. System 7.6.1 is important because it is the oldest Mac OS capable of running Navigator 4.08, the oldest browser to support 128-bit encryption. This also does not take into account the cost of adding the necessary RAM to these machines to run 7.6.1.
So that leaves 5,877 capable of running a recent OS and browser. Of those, 2,500 are leased and due to be returned to Apple within the year. If we lose them, we’ll drop back to 3,377.
3,377 machines dating back to 1997 are the cream-of-the-crop computers in our schools. That may not seem unreasonable, but look at it as a percentage of our total. It is less than one-third of all our machines. Remember these computers comprise both student and staff (instructional and administrative) machines.
Complicating the issue is the disbursement of those machines throughout the schools. Typically, when money is found to get new computers, it comes from a one-time infusion of cash like a bond. It’s very, very difficult to get money for leasing computers into the school district’s budget. So when we get that shot in the arm, the critical thing is to replace as many of the real sluggard computers as we can, which usually means doing a whole school at once. What that means is that those 5,877 computers are all in just a few of our schools. We have at least five schools that have no computers newer than six years old, while some schools are populated entirely with the leased iMacs or machines that parents and school business partners scrounged together and donated. The majority have a mix, leaning toward the older machines.
All of these machines run on a network that is barely up to the task. All of our schools, from Eagle River (30 miles north of Anchorage) to Girdwood (30 miles south of Anchorage) are on our district WAN, which is great. But our WAN is old and maxes out at 5 megabits, and our Internet connection consists of 6 T1 lines, for a total of 9 megabits. As I write this, things are pretty fast because it’s spring break, but most of the time I might as well be on a cable modem, and my office is one building away from the outbound connection. Most of the schools are much further away. To compound the problem, many of the schools are wired with 100 megabit backbones, going to 10 megabit hubs in the school labs, not 10/100 switches. Watching a student try to pull up an image-heavy or bells-and-whistles-laden site in the middle of the day is frustrating and boring all at the same time.
We are adding six more T1s by the end of summer, for a total of 18 megabits. This will be great for those of us close enough to the switch to take advantage of it, but without upgrading the entire WAN and all of our schools’ networks, we could actually be facing a worse situation next year, as the schools closest to the outbound connection saturate the network with high-volume traffic.
All of the desktop machines in the district are supported by a two-person help desk, three field techs (software), and three hardware techs. The network is supported by a staff of four.
There is another department called “Instructional Technology” that is staffed by 10 certified teaching personnel whose stated purpose is to help teachers integrate technology into the curriculum. More often than not, however, they end up troubleshooting file servers and helping set up networks because our three field techs can’t get to everyone who needs help.
Our entire Information Technology staff, including the twenty-two people listed above and some other related departments, consists of about seventy people, including the CIO and his administrative assistant. This includes server administration, programmers, and the operators for the old VAX that runs our student information system. Most of the programmers and operators are dedicated to administrative systems, so the real support for those 15,245 machines is in the hands of the twenty-two people listed above, and ten of those people only do it because there aren’t enough hands to go around.
The ADA makes it quite clear that educational opportunities cannot be denied to anyone because of a disability. If you create a web site or application that does not take into account assistive technologies like screen readers and that site or application goes into wide use, you suddenly open yourself up to potential lawsuits at worst, and at best you needlessly deny some students access to your work.
Section 508 requires “Federal agencies to make their electronic and information technology accessible to people with disabilities”. While this doesn’t currently apply directly to public school districts, it does set a clear precedent for the federal government to regulate accessibility.
It might be instructive to look at the example of the Children’sInternet Protection Act of 2000 (copy hosted by the Internet Free Expression Alliance) and Universal Access (E-Rate) funding. Under CIPA, all schools and libraries that receive E-Rate funding were required to install Internet content filters on their networks or lose that funding until a filter was installed. For the ASD, that funding rounds out to about US$1 million per year. Did we want to install a district-wide content filter? Not necessarily. Did we concede? You bet your boots.
Given the unambiguous nature of the ADA, the federal government’s attitude toward the accessibility of its own web sites, and the history of Congress tying critical funding to compliance with laws, it only makes sense to stay one jump ahead and code any K-12 project with an eye toward both laws. There has not been a test case yet, but I would not be surprised to see one within the next two or three years.
A couple of years ago, the e-commerce developers that I know used a rule-of-thumb that assumed the user had a 4.0 browser and a 56k modem at best. That still may be a pretty realistic assessment of many K-12s.
The hardware figures that I give are only for the Anchorage School District. Obviously, you would do an assessment of any client that you took on, but I think ours are not atypical. The browser bugs and long downloads that we find mildly annoying when shopping online are the sorts of things that can bring a classroom full of K-12 students to a complete halt and disrupt instruction as a non-technical teacher tries to figure out why QuickTimeInstruments won’t load.
The legal issues are very real, and will become much more important as web-based instruction gains acceptance in K-12 schools. There are a few districts in rural Alaska that use web-based instruction already. The web holds tremendous promise for education, and as the technologies that drive it mature, the possibilities for inclusion only increase, but care must be taken and attention must be paid.
Of course, any job requires a realistic study of the client’s capabilities, and if you are doing specific contract work for a particular school or school district, your numbers may be very different than those presented in this article.
But it’s critical to keep in mind the wide array of machines and personnel (both users and support) that any given K-12 organization is likely to have. We regularly do work for departments within our district (not schools) that are all Windows, and running newer machines. It doesn’t significantly change how we code, though. We have yet to create a web application that won’t run on NN 4.08 for Macintosh, but it makes more sense to keep all our code reusable.
Also remember that communication between departments in K-12’s may not be optimal, and you might be half way through a Windows-optimized development job before somebody at IT finds out about it and tells you that you have to support Macs. So before you get too far into a project, be sure that the people you’re dealing with are being realistic about their requirements. Tell them you need to talk to somebody at IT.
Finally, look forward. Don’t stick a school or district with a product that will be obsolete or unsupportable in two or three years. Make sure that if they can’t contract you to add to the product, or fix it, or just support it, that they can hire somebody else to.
For our non-US readers, K-12 means “Kindergarten through Twelfth Grade”, or primary and secondary education. It roughly covers the ages of 5 through 18, give or take a year on each end. This market is dominated by public school systems, but also includes many private institutions.
return to article
This article started as a response to a question on thelist, where I was pulling numbers from memory. Any numbers that have changed from my original email are more accurate numbers and give a more meaningful count of the machines that would actually be accessing a site.
return to article
That lease is up next year, and there is currently a discussion centered around buying the leased machines for $1 each or negotiating a new lease. It’s an interesting conundrum, because we have machines that are capable of running the newest Mac OS, but they’ll all need RAM upgrades at the minimum, so that $1 turns into more money by the time we buy the RAM and pay the people to crack the cases. Add to that the fact that in three years, when these machines will be the equivalent of an LCIII in today's terms and we have to replace them, we're back at square one. Obviously, leasing would be ideal, but it’s hard to explain to the taxpayers why we need to upgrade our computers every three years. In all likelihood, we will take advantage of the $1.00 buyout option, whether it’s in our best interests or not, because we cannot afford to lose that critical percentage of “modern” machines.
return to article