An unexpected challenge with teaching on-line

Three (or more) operating systems times three (or more) versions of software with bugs unique to one or systems (that I don't have) means too many systems for me to manage teaching. Thank the FSM they're not using Linux, too. (Let me see that would be Ubuntu Linux, RedHat Linux, Debian Linux, Yellow Dog Linux, Vine, Turbo, Slackware, etc.. It quickly gets to be too exponential.) Nope, sorry, three versions of Microsoft Office on three different operating systems are bad enough. This semester, I'm teaching an on-line for the first time ever. The subject isn't new to me. I've taught bioinformatics off and on in different venues for almost ten years. It is strange for me though to communicate entirely through e-mail, an occasional video, and through writing. I miss seeing the students face to face and without the ability to watch what they're doing when they use computers, I'm slower to diagnose problems and figure out how to help. I've also found a new unanticipated challenge and I wonder how other people deal with this. This challenge is dealing with all the varieties of computing platforms, versions of different software, and bugs that seem to pop in the versions that I don't have. Now, there are many challenges to teaching bioinformatics in any situation, largely, I think, because it involves learning how to use several new kinds of software programs, at once, and usually a new language. The students who are experienced with programming are often disdainful of the academic databases and freeware programs that we use, (and sometimes ticked off because biologists use words that they don't understand). Even many of the biology students do not know enough biology to understand the words that are written in a database record. Many of them are not good with using software at all. They can't switch between windows, they don't know any keyboard commands, and they loose files that get downloaded. I can attest that everything that Joel said about computer users, in general, is correct about students, too. To paraphrase Joel:

1. They cannot use the mouse. 2. They do not read instructions. 3. They do not remember.

Perhaps there are challenges too, because of the way that I teach the subject. I teach bioinformatics the same way that I would teach cooking, media preparation, or gene cloning. The students get lots of assignments and no tests. My view is that a big part of bioinformatics revolves around analyzing molecular data and I would like students to finish my class having acquired the ability to do this, at least to a certain extent. This semester, we're working on a project that I introduced in another class last fall. We analyze chromatograms that were derived over the past four years by sequencing bacteria isolated by students at JHU. The students identify the bacteria via blastn, and count the different kinds of bacteria from different biomes. Then, we identify polymorphisms in the genes that were sequenced, align sequences, make phylogenetic trees to look at the diversity of these sequences, and see where the polymorphisms map in the structure of the ribosome, and whether they map in regions that are bound by antibiotics. It's pretty cool since we get to apply a lot of different techniques and since we're working on a semester-long project, we have some continuity. So what's the challenge? The challenge is that my students (and I) all have different kinds of computers and different versions of Excel. As I said above, I like to have the students count and graph the numbers of each bacterial genus in each biome. If we could do this with Google Docs, we would, but Google Docs doesn't have a feature for doing pivot tables and I've found that pivot tables are by far the best method for this kind of work. I thought that things would be fine even though pivot tables are little complicated. But, I realized from student's questions that this was harder than I thought. So, I made movies to show how to do this. Still, I didn't realize the extent of the problem until I saw the assignments and found that only one student had managed to do the graphing correctly. This was downright depressing. The problem turned out to be a bug in Microsoft Excel that only appears on computers with Windows. I never saw it because this problem doesn't happen on my Mac. I thought I had the problem solved though, when I borrowed a computer with Windows XP and Office 2003, reproduced the behavior my students described, and figured out a way to work around the problem. Then one of my students sent screen shots to show me that his computer doesn't have a wizard and he's using Excel 2007, on VISTA. Crap. I don't have a copy of VISTA. I don't have a copy of Excel 2007. Yes, I can require that students purchase certain software, but asking them to go backwards and get Windows XP for one, maybe two assignments, is not a realistic request. So, I wonder. Do other instructors have these problems? Do other instructors just avoid assignments that involve commercial desktop software? I would quit using Excel and Word, but having taught in community college where 30-50% of our students were graduates from the University of Washington, I strongly feel that college instructors have a responsibility to leave students with at least a few marketable skills. I also know for a fact that my former biotech students had a big advantage in the job market because they were really good at using Office programs really well. But, back to my question, are these problems that other instructors have? How do you deal with the multiplicity of platforms and software versions?