Cost of Open Source

I walked into the office of a colleague the other day (the director of our IDT graduate program) and on his office table was a new Xserve blade and a Xserve raid (the Macintosh server with software and storage). Very nice equipment that will allow the IDT students to each have their own accounts and offer many terabytes of storage for streaming media.

“When are we going to have this system online,” I wanted to know. “I am waiting for someone to come to get everything up and running” was the reply.

So, I have been trying to estimate what this combination of hardware, software, and services was worth. Hard to know for certain. There were terabytes of storage in both the blade and the external RAID. The hardware had a very fast CPU. OS X server is not free and then someone was going to set up the entire thing.

I am in my “lab” on a Saturday morning working on my own server (the setup running this blog). It is a G-4 cast-off from a campus lab that I picked up as surplus. I use all open source software – the server, database, blog, and middleware software responsible for this blog and the other applications I run are free. Total cost = $0.

Clearly, I would not want to put my equipment up against the resources now available to the IDT students – no way the combination of software and hardware are close in terms of the users that can be serviced simultaneously, volume of content that can be streamed, storage capacity, etc. Still, $0 is $0 and I actually have and can continue to add more servers of the type I have described. Anything multiplied by $0 is still $0.

I happen to like operating my own hardware and software, but here is the other perspective on the cost issue. The other perspective would recognize that I am up here on Saturday morning. I am spending time doing things that would not be required of the users of more “ready-made” systems AND I have spent a considerable amount of time to get to the point that I am able to keep open source software operating. If I were not interesting in doing this “for fun” and say I was operating these same resources for a K-12 school, what would my time be worth?

My task this morning was to upgrade the blog software that serves the content you are viewing. If you are not familiar with open source, here is a description of why this singular task may be more complicated than you realize. To provide you access to my blog with free software, one of the challenges is to keep multiple free components synched (I don’t mean this as a technical term). The server software (Apache), the database storing blog “pieces” (MySQL), the middleware conntecting the database to the server and providing the processing necessary to assemble blog pages sent via the server (PHP), the blog software (WordPress) and blog software plugins (the add-ons that deal with spam, insert the tags, etc) are all continually upgraded. These upgrades make the software work more effectively and also reduce security threats. I could get everything working and then not mess with the new advances, but the university would be upset if I did not do what I could to reduce security problems. The challenge is that you have to deal with many separate parts and you have to be careful that the parts remain compatible. If you pay attention to only some of the pieces, you often find that the piece you have upgraded assumes a more current version of another piece and your upgrade has in some mysterious way “broken” the entire system. I don’t mean to make this sound impossible – everything went smoothly this morning. My purpose was to identify some challenges that are hidden in the use of open source software.

I learn a great deal doing what I do. I think acquiring the experiences I have allows me a different perspective on the way things work that those who rely on ready made apps simply do not have. Some may feel this knowledge is unnecessary for practitioners. It really is hard to say what is the best way to spend your time.

Blogged with Flock

Loading

Ubuntu

Every once in a while, that geek part of my personality forces its way to the surface. The time has come to describe Ubuntu.

I use open source software on several servers I operate (this one for example) because I have access to bandwidth, but not a lot of money. My hardware consists of older Macs (the cast-offs from computer labs) and the Mac OS is built on a linux foundation (BSD). From time to time operating this type of server requires a basic understanding of linux. Such requirements in combination with reading several accounts of the work of Linus Torvalds created a personal interest. Working with linux commands was kind of a throw back to my first experiences working with computers. You enter text command rather than interact with the computer through objects and menus.

I first heard of Ubuntu when listening to podcasts (probably Leo Laport but I can’t remember for sure). It was described as a version of Linux that could be run from a CD on a variety of hardware. The cool thing about this is that it was possible to experiment without having to worry about messing up the operating system of the computer (which was always somewhat of a concern when I was working on one of my servers).

When I finally got around to looking for Ubuntu (the holiday break), I encountered Edubuntu. The edubuntu community is attempting to develop a host of educational applications around ubuntu. Again, experimenting is low risk and no cost. You pretty much download a collection of resources associated with a CPU (e.g., Intel x86) and then burn a CD. You boot your computer from this CD. A user interface has been developed so you will have no trouble launching a variety of open source productivity (e.g., OpenOffice), Internet (Firefox) and educational apps.

The vision here is to create low cost alternatives for education and the resources are intended to be installed (not run from a CD) or run in thin client mode (programs are loaded from a file server when run).

If you have heard about open source alternatives to popular productivity tools or linux and were curious, this is a reasonable way to explore a bit. I would suggest you create a CD even if you have a specific interest (say OpenOffice). Getting OpenOffice to run on the Mac is a little more complicated than just downloading the program (e.g., you need X11 installed first), but running from the edubuntu CD just works (I must admit I did not deal with the issue of how I would store files I might create).

Blogged with Flock

Loading

NUVVO – Teaching/Learning Portal

During the flight to my recent research conference, I was listening to a randomly selected educational podcast and encountered a description of NUVVO. I made a note to investigate this product when I returned home and had some time. Here is my comment.

I would describe Nuvvo as a CMS (course management system). At the low end, Nuvvo is a free service allowing anyone (a stated goal is to “bring out the teacher in all of us”) to build an online course and offer it to “learners.” The free version limits course authors to 25 MB of content. The free version also contains advertisements. If you wish to explore the capabilities of this service, there is a five minute screencast that explains core features.

I think this is a significant and interesting development. It is probably too early to speculate about broad topics such as what the opportunity to offer a free or inexpensive educational content management service might mean. Will ventures such as this challenge the place of traditional educational institutions in serving nondegree seeking students? Will such services challenges companies such as lynda.com or atomiclearning.com that provide specialized training? Will educators in K-12 settings use such products to extend classroom experiences?

Exposure to this service reminded me of the efforts of some of my University of North Dakota colleagues to develop a CMS product (HTMLeZ). University ventures have contributed much to the field of technology (e.g., the original Mosaic browser), but I wonder if such projects should be approached with a commercial focus. Sometimes a commercial opportunity emerges, but I think this cannot be the reason research and development projects are undertaken. Alternatives to the high end commercial products (e.g., Blackboard) are needed, but it is unclear how such alternatives will emerge. Will new ideas come from small commercial startups, from open source initiatives (e.g., Moodle), or from companies such as Nuvvo.

Loading