You make very telling points. Management Accounting also known as Cost Accounting is very poorly practiced in Academic institutions - again the irony we teach but don't do. The gap is particularly noticeable in the service provision area where centralised services are not costed accurately. IT generally falls into this category and the understanding of the cost of provision of IT services is less than desirable. As you state this complicates the comparison between Open source and proprietary software. Patrick Masson gave a very comprehensive inventory of direct costs of providing Moodle and WebCt, however the indirect costs were not articulated. These would include management overhead, contract management, training, space, etc. In his case I don't think they would have made much difference but they can be significant and can reverse business cases.
The main difference with an Open source architecture is the growth of internal responsibility. This requires increased management and communication to undertake many of the functions that a vendor does in providing product research, development, documentation, training and consultation. For some this is being done in any event, large shops in particular may be quite sophisticated in this regard. However, for smaller shops that have vendor based solutions, Microsoft houses for example, the degree of expertise to undertake the broader managerial competencies in addition to the technical skill sets may prove less efficient and effective and ultimately more costly.
----- Original Message -----
From: "Stephen G Landry" <[log in to unmask]>
To: [log in to unmask]
Sent: Saturday, February 7, 2009 9:22:12 PM GMT -07:00 US/Canada Mountain
Subject: Re: [CIO] Savings with Open Source
I’ve been very interested in this discussion. I think there are good reasons to like open source software. On the other hand, I took Economics 101 some thirty years ago when I was toying with a business minor, and I’m nagged by a couple of questions about the cost of open source versus commercial software.
One difficulty I have in comparing the cost of open source versus commercial has to do with how costs are budgeted on our campuses. For example, when I license a commercial software product, the indirect costs associated with the product – the software firm’s building rental, their electricity and networking costs, the benefits for their employees, their legal and accounting costs, etc. - are funded from the IT budget in the form of the licensing fees I pay for the product. When I participate in an open source project, those indirect costs are absorbed by many different campus budgets outside of IT. The cost of the license versus the cost of a developer are incommensurable quantities – there’s different things bundled in each - and so that may not be the full story in determining whether I’ve saved my institution money by moving to open source.
I also worry that in comparing the costs of open source versus commercial software I may run up against a mild form of the freeloading problem in economics. I suspect that a relatively small number of institutions do a disproportionately large amount of the heavy lifting for these open source projects. In contrast, software vendors tend to discount “per seat” costs for larger installations, and so they spread the costs very differently across licensing institutions. This can distort our cost comparisons, especially if we’re not one of those large institutions. I also suspect that some of those institutions that do the heavy lifting are the ones that will have the greatest challenge in an economic downturn, since they rely more heavily on endowments or state appropriations to sustain their operations.
This is not to say that open source systems are not worthy projects. Seton Hall University uses a large number of open source systems to provide mission critical IT services to our community. But we have to be realistic in evaluating their costs. I have no reason to believe that the employees of software vendors are less efficient or effective than those employed by colleges and universities. In the end, the cost of software is the cost of software, but different models may account for the costs differently.
By the way, I have the same issue of cost accounting in other areas of my University. Seton Hall University has started to move toward a responsibility centered management approach to budgeting that will account better for the costs of our various services, but right now we don’t do a really good job of understanding the costs of our programs and services and aligning our objectives and incentives accordingly. I have a colleague who is a program director and who is proud of the fact that his program brings in $1 million for only $800K in direct costs. He believes he’s netting the University $200K. But for the University overall, there’s $0.50 in indirect costs for every $1.00 in direct costs, so by another reconning, he’s losing the University $200K. In the end, we tend to achieve the objectives we’re given; his objective is to maximize revenue, so that’s what he does. In the absence of good cost accounting, if my goal is to minimize IT expenditures, rather than institutional expenditures, it might sound good to me to drop some software licenses and replace them with open source, but I worry that I’d be shifting costs as much as reducing them.
Stephen G. Landry, Ph.D.
Chief Information Officer
Seton Hall University
South Orange, NJ 07079
Email: [log in to unmask]
For the latest news from SHU IT, see the IT News and Announcements blog at: http://tltc.shu.edu/blogs/projects/DoIT/
For the latest news on teaching and learning, see the ePirate blog at: http://tltc.shu.edu/blogs/projects/epirate/
********** Participation and subscription information for this EDUCAUSE Constituent Group discussion list can be found at http://www.educause.edu/groups/.
This communication is intended for the use of the recipient to whom it
is addressed, and may contain confidential, personal, and or privileged
information. Please contact us immediately if you are not the intended
recipient of this communication, and do not copy, distribute, or take
action relying on it. Any communications received in error, or
subsequent reply, should be deleted or destroyed.
Participation and subscription information for this EDUCAUSE Constituent Group discussion list can be found at http://www.educause.edu/groups/.