Conclusions

This document summarizes the conclusions of the debates in the Scientific Methodology section.

 

For the betterment of humankind, we are pursuing advances in our scientific understanding of the Universe. We are committed to using the knowledge so gained for the general welfare of everybody. We are convinced that the misuses of science stem not from science itself, but from immature science, which was applied before there was a full understanding of the consequences. So while the objective is to help people, we want to fully understand all of the problems first.

 

Professional science has become very abstract, preferring mathematical models over physical models, and substituting model data for actual observations. It has been argued that scientists favor abstract math that cannot be tested, because then they cannot be proven wrong, which would result in their funding getting cut. So despite their clear advantages in education, experience, teamwork, and funding, professional scientists must conform to political and economic pressures, which can hamper their productivity. Meanwhile, a large number of anomalies in the existing models have been identified, meaning that there are still plenty of opportunities for progress. Historically, citizen scientists have always been freer to explore new opportunities, and despite their impoverished resources compared to professionals, enthusiasts have been responsible for a disproportionate number of breakthrough discoveries because of that freedom. So the opportunity is for the enthusiasts to explore the lines of reasoning not being investigated by the professionals, meaning theories based entirely on reliable observations.

 

With limited resources, enthusiasts must pool the efforts of a multitude of people whenever they can. The Internet makes this possible, and therefore will be the primary instrument of this initiative. Open-access systems have problems with spam and vandalism, so only high-quality software with the appropriate mechanisms for managing accessibility can be used. Even so, the system will also rely on moderators to encourage registered users keep their material cleaned up and consolidated. This includes identifying and documenting fallacious reasoning. Of course, moderators are not universal geniuses, and are ill-prepared to judge good from garbage on all topics. So workgroups should be formed by people with expertise on specific topics, to make sure that the ideas are well-represented. Furthermore, free-form prose sometimes makes it difficult to tell whether statements are verifiable facts, or unsupported theoretical placeholders. To uncover such soft spots, it's useful to break the reasoning into so many stand-alone statements that can be scrutinized individually, and clearly labeled as fact vs. theory. Then users can rank the statements.

 

From the history of modern science, beginning with Galileo, we inherit a scientific method, which we are adapting to online collaboration:

1. Definitions
2. Documents
3. Discussions
4. Observations
5. Hypotheses
6. Predictions
7. Conclusions

The first difference in what we're doing is that the Scientific Method is typically a singular linear process, while we turn it into the framework for the parallel evaluation of multiple lines of reasoning. Thus our conclusions cover the multitude of endeavors within the specific field of focus.

 

The second difference is that as citizen scientists doing theoretical work, this process typically does not drive laboratory research. Rather, we're doing literary research, and evaluating theories for their explanatory and predictive capabilities, and trying to work out the anomalies. So the main body of the work is in the Hypotheses and Conclusions sections.

 

Finally, we include a folder for Documents, so that people can post their theses, and another for Discussions, encouraging the free exchange of unstructured comments. This is a great way to expose new ideas and get quick feedback, but decent ideas can get buried in long discussions. So we employ several methods to extract the value from discussions, making it more accessible. The first thing we do is allow people to insert footnotes in other people's posts. These indicate that there are follow-up comments, and provides links to them, which enables following a line of reasoning through a series of posts, skipping over all of the off-topic and/or redundant posts in-between. Next we strive to maintain thread summaries, to highlight the salient points. The next step is to further refine the ideas into a topical structure. Flowcharts and outlines can be used to organize ideas, such that each unique idea appears in only one place (eliminating flooding and spamming), and in proximity to related ideas (eliminating the distraction of off-topic comments, a.k.a., scattering). These methods are not mutually exclusive. Rather, they complement each other, and both should be used, in conjunction with free-form discussions, to best capture, nurture, and present ideas. And in addition to flowcharts, forums, and outlines, articles can be written to express the ideas in prose. When multiple methods are used, they should be cross-linked so people will know that there is another way of looking at the ideas, possibly including a different type of information. Of these methods, outlines provide the densest concentration of ideas. Collapsible outlines offer the added benefit of hiding sub-topics to get a better overview, and faster navigation in a less distracting environment. This is especially useful on PDAs, where the screen size limits or eliminates the effectiveness of graphics, and prose requires too much scrolling. Guidelines for well-formed outlines are here.

 

In addition to cross-linking within a construct between different representations of the same information, cross-linking between alternate hypotheses when they use similar constructs means that all labor invested in evaluating and improving the one will apply directly to the other.

 

Requiring that hypotheses explicitly explain specific observations, when broken up into individual statements to enable highly-specific criticisms, necessitates that a review article, or a conclusions section, be written to show how all of the pieces fit back together again. Then, the conclusions continue in successively coarser granularities. First there are conclusions concerning a particular hypothesis in the way it explains a particular phenomenon. Then there are higher-level conclusions concerning which hypothesis is performing the best for a variety of phenomena. Finally, there are decisions concerning which general paradigms seem the most promising. So we take hypotheses, break them down into their individual components, and evaluate them. Then we rebuild a world view, first with small assemblies, and then larger ones. And the reasoning is presented at every step, so that it can be reviewed by others, and so that when things change, the implications can be chased throughout the rest of the reasoning. The ultimate objective of this process is the development of a Library of Scientific Knowledge. By allowing ideas to be evaluated at every granularity

, we might include promising new paradigms that might not already handle a great deal of the phenomena to a high degree of accuracy, but might nevertheless open up new opportunities for research, getting past the diminishing returns of existing paradigms.