![[Home]](http://www.bayleshanks.com/cartoonbayle.png)
Sometimes you hear people talking about how it would be desirable if academia moved away from the tenure system. They advocate having more nontenure track professors.
This would be fine if it were done right but the problem is that right now the tenure track professors don't want to give status and power and money to the nontenure tracks; so currently the nontenure track profs are secondclass citizens. To make it work, the nontenure track ones should:
Also, when you're up for review for tenure it's a very stressful process. You don't want to make this process happen more often for the non-tenure track people. A suggestion: for non-tenure track folks, let the professor decide when to be reviewed, within a 4-year margin or so.
currently/at some point in time academia effectively rewards "idea acceptance futures"
i'm not sure, but perhaps Western science has been arrogant by assuming that their method exempts them from having "schools" of thought about a given topic. i'm not sure if Western science thinks this, and i'm not sure if it's wrong if they do. so the following are all tentative hypotheses, but right now it seems to me that:
Until the invention of the scientific method, there was a long history in various cultures, and various topics (natural philosophy, philosophy, martial arts, medicine, political philosophy) of having different "schools" of thought about knowledge that was passed down. These schools had different topical and disciplinary foci, but also different opinions that caused disagreement with other schools on topics that both schools included.
Then came the scientific method, and for awhile there were less "schools" -- yes, there were locally isolated communities of thinkers who knew more about certain topics and skillsets than others, but as for differences of opinion about scientific conclusions, there were some but they were temporary.
This was thought to be a consequence of the scientific method; the scientific method would allow the community to reach consensus on scientific theories, allowing the eventual elimination of any particular "school" (although topical and disciplinary boundaries would still exist).
As knowledge grew, specialization would increase, but it was not thought that this would cause a reemergence of "schools".
However, now it seems like it was just a consequence of the paucity of knowledge that had been generated throught the scientific method early on. As knowledge has grown, it seems that there are multiple ways to partition the available knowledge into specialties. Schools have re-emerged as different groups of people study different overlapping subsets of knowledge. These schools describe the same thing with different terminology. Some of the vocabulary ontologies are not isomorphic, in that atomic terms in one ontology can map to complex terms in others.
This leads to situations in which some topic is included in multiple schools, and the people in one school are not well informed of the progress made on that topic in the other schools, due to the large amount of time that would be required to communicate across schools. Therefore, although in theory the scientific method could lead to a convergence to consensus among the entire community on any given topic, given sufficient time for all schools to communicate their results to each other and understand the others, in practice the rate of progress within schools exceeds the rate of convergence to consensus between schools, leading to an ever-growing list of shared topics upon which the schools disagree -- divergence.
Anecdotal evidence for this can be seen by considering how the graduate thesis topics of many students in science, in order to be "done right", would seem to involve familiarizing oneself with more than a decade's worth of results in related fieds, in addition to relevant results in the student's chosen field. The student is not permitted to spend more than a decade on their thesis (and often does not desire to do so), hence corners are cut and the student picks and chooses among the related results that they actually learn. In other words, specialization is insufficient; for many thesis topics, there is no set of fields that completely covers all previous results relevant to the topic, and which can all be mastered within the time limit.
This problem could (and should) be ameliorated, but not solved, by decreasing the scientific career rewards associated with the production of new knowledge within one school, and increasing the career rewards associated with the translation of results from one school to another.
It is amusing that this problem has been faced by other communities of scholars that predated science, and that science scoffed at this, taking it as evidence that the other scholars were not interested in verifiable evidence and objectiviy, but merely in spouting opinions. In fact, the relative lack of schools in early science seems to be a result of the paucity of knowledge that science had generated at that time, compared to the large amount of knowledge in other communities.
i think cs and in particular A.I. could use a lot more of this:
"From 2002 until now I have been involved in revising the OOPSLA conferences. In 2002 I introduced Onward!, which is a track or subconference for new ideas and new thinking. Sort of an antidote to the salami-slice approach to computer-science research (normal science in Kuhn's terminology). In 2005 I added Essays which is the other end—deep reflection from a personal vantage point of technical topics or philosophical issues in computing / software." -- http://dreamsongs.com/OOPSLA.html
"Computer Science is now so obsessed with rigorous evaluation that programming language research has become stultified." -- http://alarmingdevelopment.org/?p=451#more-451
http://dreamsongs.com/MFASoftware.html
---
