Openness: September 2003 Archives

MIT for free, virtually (serendipitous link discovery via ResourseShelf)

The Massachusetts Institute of Technology is making its course materials available to the world for free download

"One year after the launch of its pilot program, MIT on Monday night quietly published everything from class syllabuses to lecture videos for 500 courses through its OpenCourseWare initiative, an ambitious project it hopes will spark a Web-based revolution in the way universities share information."

Let's see how far (in time and space) this ‘revolution’ will reach! Maybe, if each school does not have to (re)create the course materials from scratch, the tuition will go down! :) Or maybe someone will be making more money.

Nevertheless, in terms of information and/or knowledge sharing there ought not to be any doubt that this is a step in the right direction. Hopefully, the potentials can be utilized to benefit the society in general.

In An Open-Source Search Engine Takes Shape there is an assumed relationship between open source, open ranking, and fairness of returned results.

Currently, all existing search engines have proprietary ranking formulas, and some search engines determine which sites to index on the basis of paid rankings. Cutting said that, in contrast, Nutch has nothing to hide and has no motive to provide biased search results.
...
"Open source is essential for transparency," he said. "Experts need to be able to validate that it operates correctly and fairly. Only open source permits this." If only a few Web search engines exist, he said, "I don't think you can trust them not to be biased."

I think this relationship is sounds. How does one test and evaluate that indeed the opens source search engine will result in 'open ranking' algorithms and thus lead to fairness?

The next issue to be dealt with is the scope and the understanding of fairness in the context of search engines. Should fairness be understood as proportional (returned results vs. the total number of searched documents), or equal coverage of the queries even though some topics of interests might be less represented on the internet. In addition, considering that no one single search engine can cover/index the entire webspace, what would be the criteria for domain/URL inclusion for indexing?

I believe that the open source search engine might be better in fairness, but there still remain a lots of issues to be dealt with as important factors in tilting the 'fairness' one way or another.

In The Age of Corporate Open Source Enlightenment Paul Ferris enlightens us with a fair and balanced analysis of the open source software appropriation in the corporate environment.

The ideological connotations and the analogy to the church vs. state might be overemphasis, however, the article is very insightful and brings forth the philosophical background which appear necessary in the process to contextually understand the open source software and its environment.

From The Mercury News in Open source helps education effort in Third World:

"In Africa, in Asia, in much of the world -- especially in the developing nations -- open source is looking like the best way to usher in the information age. Money, flexibility and plain old independence from a monopolist's clutches are a powerful combination."