Machtelt Garrels wrote: >>In my experience developing curricula for both bound hard copies and online >>distance learning, educational institutions (both colleges and universities) >>prefer to create their own non-redistributable and proprietary courseware. >>(Anyone remember the big fuss when MIT launched opensourcecourseware?) >> >> Sure there was a fuss. Any sufficiently useful attempt to do things differently will upset anyone who has money or reputations to be made by keeping material proprietary. However there is also an understanding, more common in academia than in commercial IT training, that building on the works of others can be a Good Thing.
On my personal website I maintain a listing of freely-distributable courseware for Linux instruction; I look forward to being able to add BSD-based materials to this list: http://telly.org/free-training >If a certification body also launches courses, whether these are Open Source >or not, there is a conflict of interest. > This is true in the sense that a certification is a standards-setting exercise, and for a standard to be truly open anyone should freely be allowed to create implementations of that standard (ie, training toward that certification). While I agree that the standards body itself creating definitive or reference implementations ("official" -- but not exclusive -- books and courseware) is technically a conflict, there are multiple reasons why doing so can be beneficial or even required: 1) In the case where the conventional publishing industry may not believe enough sales will exist to justify creating materials, having "official" documents may be the only thing available until demand is demonstrated. The lack of *any* materials can be a significant obstacle to the certification's acceptance. This may indicate production of official materials as a short-term tactic to start momentum. 2) In the absence of an elaborate and trusted system to verify the quality of third-party materials, having a reference version guarantees at least one implementation that is likely to meet all the objectives of the cert. 3) Certification itself can be very expensive to produce and distribute, even when avoiding the VUE/Prometric path. Psychometricians aren't cheap, and depending on a totally voluntary staff can make deadlines and rotations difficult to enforce. I recall that the rule of thumb for commercial IT certs was that creation of any new psychometrically-validated certification level would run about $1M. LPI, with substantial volunteer help, was able to bring costs under $200K per level, but that's still significant. The BSDCert donate page, which indicates a (far too low IMO) target of $30K, is currently only at 3% of its goal. In such cases, publishing official courseware may be a 'necessary evil' that can mean the difference between stability and instability. - Evan _______________________________________________ BSDCert mailing list [email protected] http://lists.nycbug.org/mailman/listinfo/bsdcert
