Wednesday, December 23, 2009

Add your voice to the strategic planning process

Welcome to the President’s blog, created with our strategic planning process in mind.

Strategic planning needs to be open and transparent and it needs to involve a broad cross-section of College employees.

I’m excited about adding a blog to our process, in addition to the information on the strategic planning website. The blog is a venue for timely updates and deeper discussion. It is a way for you to add your voice, and to encourage your colleagues to do the same.

Initially, most of the content will relate to our seven key discussion themes. The blog will provide information about the themes as well as a timely summary after each themed discussion group meeting, so that you can keep up to date even if you can’t attend a particular discussion group. And for those who do attend discussion groups, the blog allows you to contribute further ideas or commentary, in a format that’s easy and convenient. This way, we can have a richer, deeper dialogue on each theme.

If you’re unfamiliar with blogs, check out How to follow blogs.

And also check out How to post a comment.

I look forward to your input!

1 comment:

  1. This is anonymous for two reasons. First, I need to feel free to say what I think needs saying. Second, anonymous in the sense that I don’t see my concern reflected yet in topic heading.
    I should begin by saying – I’m not luddite. But I’ve grown a little alarmed at some of the ways in which computers are being pushed at the college, as an arm of management.
    Up until a couple of years ago, we had a human administered writing assessment test. It was slow, but internal studies showed it was offered students good placement advice. There were various ways in which it could be made to run faster, but instead, it was replaced by a computer driven test that has proved to be erratic at best, and which generates a large number of anomalous results. The Accuplacer test has so far absorbed considerable amounts of staff and faculty time, attempting to fine-tune it, but its results are still, the last time I looked, arguably worse than the results generated by the old test. What is the use of offering students poorer advice faster?
    More troubling is the use of Infosilem to create faculty schedules. The program so far has a miserable record of creating workable schedules, yet management’s latest response is not to say that faculty may continue to make their schedules the old-fashioned way, but rather to say, you must appoint someone who is specially tasked to implement the program in each department.
    That not only violates faculty’s rights, it also shows a poor grasp of the underlying problems.
    Scheduling is an optimization problem. I’m not a mathematician, but optimization problems of this sort (like the “travelling salesman problem”) are classified as NP-Hard; some of the latest solutions to the TS problem have absorbed as much as 17 years equivalent CPU time. To expect a commercially available program to generate acceptable results, given the multiple overlapping constraints, seems to me naive at best, and dangerous at worst, an attempt to take away faculty’s ability to organize their tasks in a way that best suits both they and their students.
    Why have these two problems in particular cropped up? I believe they result from a certain naivety on the part of some upper management personnel. The results have been disappointing because these two applications are at the limits of what computers can do. But why use computers at all?
    It would seem to me a better choice to make the human-based testing more efficient, rather than simply replace it with a machine test. Again, scheduling can be done reasonably well by humans, but only haltingly and quite poorly by machines. The program tries to optimize, a near-impossible task, given the scale of the computer resources being applied to the problem; a human tries for the least worst, or a best-fit solution to the problem. Which means that a human is always going to have to “adjust” the machine results to make them workable. So again, why use a machine at all?
    Some management personnel pushing these “solutions” seem to have in mind a different college than the one we actually have, one with a very differently composed workforce and and more academically homogenous student population. To try to impose “solutions” here which appear to work in those other types of institutions strikes me as foolish; it’s as if management were trying to make Douglas over into a different image, so that they – rather than students or faculty – will feel comfortable, will know better what to do.
    But that’s not who we are. We are overwhelmingly a university transfer institution. We ignore that side of our operations and its demands at our peril. We should not be trying to remake the institution into something it was never intended to be. And we should not be so naive as to think that collegial co-management can be replaced and streamlined by the use of a few computer programs to push faculty aside from areas in which their expertise can be valuable, and which, indeed, belongs to them – in part, at least – by long-established practise.