Monday, August 17, 2009

Open Source... from the Source

An Open Source IT guy responds with a bit of perspective:
For me, this post highlights a paradigm that many an IT employee in education understands, and that is that faculty drastically underestimate the amount of time and effort it takes to maintain an enterprise quality system. For the record, its a lot harder to hire a quality sys admin for open source than for Windows and also a lot more expensive. And please don't just say, "but the savings of software licenses make up for the salary difference". If you did a total cost of ownership analysis you would know that it isn't that simplistic.

I suppose that by now you wouldn't guess that I am an open source guy. Well, I am. I've replaced large enterprise (Nortel) phone systems with open source (Asterisk, sipx). If there is one system that is scary to go unsupported on, its the phones. You've got 911 and other security and safety problems. When the Internet goes dead people get annoyed. When the phones go dead, people get fired.

So let me turn this argument on its ear a little. The reason that open source doesn't take off more is that end users are generally unwilling to put up with the slightly more downtime that is possible with unsupported open source. Sometimes an upgrade will introduce a new problem with a feature, sometimes you have to wait for a patch from the community and sometimes you are communicating via a listserv with other folks trying to get things straightened out. Even a really good Linux admin will get stuck on problems and need to go to the community for support. Support from the community takes more time than a call to Microsoft (thought it is much cheaper).

When a system is down and faculty are upset, they are usually ok with "Microsoft is working on it" or "we've got a call into the vendor and they are logged into our system right now trying to get things straightened out." They normally are not ok with, "I'm just really struggling to get this freaking thing to compile, but I've emailed my friend in Bolivia and I'm waiting for his reply".

If people would chill out just a little and bring their expectations down from five nines (99.999% uptime, all but five minutes a year) it would be a workable solution. Technology has created such high demands for perfection and instant gratification that IT admins are in a catch-22. You can't spend the money for 24x7x365 support, but you're expected to provide just as good of service. These unreasonable demands are what is driving my friends and I to bail on the IT industry, it is just too stressful of an existence.

I'm all for teacher innovation, I'm all for using collaboratively developed software that holds to the open source model. My job is to get people to use technology in their instruction. We just need people to understand the trade off and to understand that an hour or two of downtime each academic year is worth the money we didn't send to Microsoft.


  1. Ok, so we keep Windows, but lose the applications that cost money - use Google Docs, Gmail, Blogger, OpenOffice, and the rest. How often do we really have problems with the applications? It's usually the network or the OS, not the apps.

    Think about the savings just from not licensing Microsoft Office.

  2. Imho, tech leaders in schools rarely do enough to set appropriate expectations for their user base. I taught for seventeen years before working in IT, so I understand intimately the demands of classroom work. After 11 years of school IT, I also understand intimately the demands of running, in our case, a $35 million business.

    Unless you educate your faculty about what it takes to live in this bipolar world--the warm, fuzzy kid-centered world and the business side of running a school (which few teachers have any clue about), they simply can't understand the context in which IT leaders make decisions.

    Without that context, humans invariably attribute responsibility for decisions based on their personal perceptions of the decision makers. This nearly always results in a flawed understanding of why decisions are made.

    IT leaders often complain that people don't understand what we do. We have a responsibility to tell them, in clear, accessible language, and in whatever variety of ways it takes for them to hear us.

    When we do that, people at least understand the bigger picture than just the slice of school life they inhabit. They realize that we face many of the same dilemmas they do in their classrooms -- short term vs. long term, individual vs. community, safety vs. convenience, cost vs. benefit. And then if disagreements do arise, the arguments are at least about reality rather than someone's uninformed and imagined reality.


Note: Only a member of this blog may post a comment.