Solar Eclipse Mini Grant (due 12 Jan 2017)
Best viewed online at https://freecollegeproject.miraheze.org/w/index.php?oldid=1191. Update by clicking "Page" above.
This experiment creates a scale-model that measures the size of the penumbra of a solar eclipse in a way that permits comparison with a simple formula. It can be can be performed on any sunny day, and costs virtually nothing, provided a quality cell-phone camera is available. The calculation assumes that the umbra on Earth's surface is negligible compared with the size of the Earth, and that the Sun is very far away compared to the distance to the Moon. Such a calculation combines the rigor of Euclidean geometry with the essential skill of approximating in a manner not unlike the approximation of parallel rays used in deriving single-slit Fraunhofer diffraction.
Aside from low cost, this experiment is significant in that we are unaware of it ever having been performed before. We got the wrong answer (with an error of nearly 20%), but decided not to repeat or attempt to improve the procedure. This lack of attention to detail by the original investigators will permit a community of student researchers to share in the thrill of scientific discovery as they attempt to reproduce and improve our result, reporting their findings under a Creative Commons license that will permit publication in forums such as WikiJournal of Science or Showcase. The former is striving to be a peer-reviewed journal, and the latter is intended to highlight the best student efforts at Wright State University Lake Campus.
Our goal is to give young scholars the opportunity to build scholarly reputations by publishing under a Creative Commons license before they graduate from college or high school.
Many active efforts promote open source ("opencoursewre") educational materials. One of the first, established in 2001 (albeit in very crude form) was MIT OpenCourseWare. It should be no surprise that one of the nation's most elite technical teaching institutions recognized that bright and dedicated young people could self-educate by reading material that has been posted on the internet. The Wikipedia:Education program is another successful effort, which has engaged 1200 students in 640 courses since 2010 (see, for example, the "dashboards" at Brigham Young Biophysics at and the B K Shah Medical Institute ).
At least two efforts to produce college textbooks are underway under a Creative Commons license that permits free distribution: Chemwiki, and OpenStax College. The latter has developed textbooks for three courses taught by Vandegrift (astronomy, calculus and algebra based physics).     It was the recent release of these textbooks that permitted Vandegrift to abandon parallel efforts on Wikiversity, and focus on two other project that are described next.
Quizbank is a bank that currently contains 808 exam questions. It is based on 1640 lines of MATLAB code, awkwardly written over 3 summer vacations by a non-programmer who teaches Physics and Astronomy. If all goes well, the code will work well enough to be used for the openstax College textbooks by 2018.
This sample "dashboard" is a simplified inter-wiki version of WikiEd's highly successful version, which creates a course management system that permits instructors to monitor student efforts on Wikipedia. The inter-wiki version has been recently modified to include links out to a collection of private wikis that can be found at:
The reader should note that the successful open source efforts at UC Davic, Rice, and MIT are all designed to grant recognition to the authors. While Wikipedia's CC-by-SA license does include attribution, this attribution is quickly buried under many layers edits and rewrites. The Miraheze private wiki-farm will not only facilitate attribution to student authors, it provides "secrecy" that is essential to course management in large introductory courses, where instructors tend to assign a common writing assignment to all the students, letting them compete for the top grades. The ability of students to effortlessly "peek" into each other's efforts would greatly frustrate this incentive.
In 2010 Guy vandegrift consulted with Tuway on the ergonomics of a mop design, which generated a gift to Wright State University that was originally intended to fund an amateur boatbuilding project. The goal was to design and build a low-cost sailboat that was as easy to build. The boats vaguely resembled upside-down birdhouses, and had no curved parts (we wanted strong boats and cheap plywood does not bend well). Two boats of satisfactory performance were built, but the project was terminated because only a handful of students were actively engaged. These funds laid fallow for five years until a worthy cause was found. That cause is the support of Miraheze, because without it, this effort to reduce the cost of higher education cannot succeed. An appreciation of the shoe-string budget under which Miraheze operates can be understood by visiting this page. Note how Wright State's modest $695 donation doubled the funds available at the end of 2016.
If computers can drive cars in city traffic, why can't we automate education? Some obstacles to this cost-saving transformation will never be overcome. Teaching will always be a person-to-person interaction, especially at the pre-college level. Computers are not likely to be soon writing textbooks, assessment and curriculum. Trained and qualified humans will always be necessary to intervene when the program is not working properly. But if an effort is made to involve students in their own education, the costs can be significantly removed. Students can propose multiple choice and short answer exam questions, and then explain why the questions are important or interesting. The private wikis on Miraheze will allow educators to share "secret" exam questions that are not yet on the study guides.
It is even possible to automate the supervision of student efforts using something called Calibrated Peer Review (CPR). In this system, students first write and submit an essay. Then, they are given guidelines for judging similar essays, and must prove their competence in this peer review by correctly judging three essays that have already been evaluated. Only then, are students tasked with assigning scores to three essays written by their peers (all articles and student peer reviews are anonymous to the students). Students are graded both on their own essay, as well as their ability to agree with the parallel assessments of their peers. In a private conversation, Professor Cavanaugh of Wright State Lake Campus claimed that this is a cost-effective way to teach, provided one subtracts out the time it takes for the instructor to learn the system. According to one published study of CPR:
- "Students who received feedback by the instructor in a traditional way did not improve their writing and critical reasoning from assignment to assignment, but students who responded to an identical writing prompt and worked though the CPR process did."
Perhaps Miraheze could someday develop a wiki-based version of CPR
- Both efforts were initiated by Vandegrift in an effort in an effort to promote wikitext in academic communications.
- Though situated in California, the only Wikipedia mention Chemwiki seems to be on the German wiki
- wikiversity:Physics equations and wikiversity:Astronomy college course
- . --Phil Bolger http://home.netcom.com/~horse/shoebox.html
- Though one must wonder if exam banks can ever be made "secret". See Isn't it insane to let everybody see the testbank?
- See http://cpr.molsci.ucla.edu/cpr/cpr_info/index.asp . A free version is available at http://cpr.molsci.ucla.edu/cpr/cpr/login.asp
- Balfour, Stephen P. "Assessing writing in MOOCs: Automated essay scoring and calibrated peer review (tm)." Research & Practice in Assessment 8 (2013) http://files.eric.ed.gov/fulltext/EJ1062843.pdf
- This author is not qualified to know whether this claim is a SMOP error.