Research Enhancement Proposals


Retrieved from: http://undergrad.usc.edu/RESEARCH-Word-collage.jpg
Retrieved from: http://undergrad.usc.edu/RESEARCH-Word-collage.jpg

Research Enhancement Proposals (REP) are recommendations for improving the quality of research findings that can be distributed to the public for free, similar to the way “free software” and “open source software” are publicly available. Its main purpose is to recommend techniques for sustaining and distributing meritorious research work (i.e. compositions) by breaking up corrupt or inefficacious practices within current industries. Practices that disallow research work from becoming freely accessible, being publicly critiqued and being shared among peers without untoward barriers are to be curtailed. The principles of free and open source software can indeed be applied to any composition for the betterment of human ecology.

In REP1, the concept of a repository alludes to any type of housing for compositions. However, REP1 has been written more so with the view that a repository would be a digital one connected to the public internet i.e. world wide web.

REP1

  • A composition within text, image, video or audio media, pertaining to any form of knowledge, may be stored and shared via a public repository without asserting any type of monetary burden upon the users for accessing the repository and its contents.
  • Author attributions and copyright licenses associated with a composition must ensure that the composition is perpetually free to retrieve, view, modify, copy and distribute for any purposes by any member of society.
  • An author, who pushes a composition to the repository may remain anonymous via an alias.
  • The privilege of critiquing a composition must not be denied to any member of society under any circumstances.
  • A “corresponding author” is expected to moderate the comments pertaining to their composition only after the comment has been posted in a publicly visible section of the repository.
  • A comment about a composition is also a composition amenable to REP.
  • Should a composition contain spam, pornography, hate speech, slurs or other abusive content, the content may be flagged and subsequently “hidden” via a public consensus of at least 3 members who flag the content. Such content may be eliminated from the repository upon a consensus of any 128 members. The concept of hidden content is that the content can be viewed upon a manual action by a person, who thus consents to getting exposed to the content. It is in the best interest of the community and for highlighting valuable compositions, to avoid reposting exasperating content.
  • A composition may also be protected from further editing and remain in public domain via a consensus of any 32 members. Naturally, a copy of it may be edited.

A method for implementing REP1

Consider the public, digital repository – GitHub which, provides a free infrastructure that is already poised for implementing REP1. A composition pushed to a public repository within GitHub is done in this manner:

  • A user must obtain an account with GitHub and use git technology to commit a snapshot of their composition within their local repository and transfer i.e. push that snapshot to the public repository of their GitHub account. This process necessarily records the time and the “user name” while storing the composition in the public repository. This form of user authentication while using git is secured via SSH encryption. So, author attribution to the composition is guaranteed via this process, just that, the user name can be an alias and no other sociological credentials of the author are recorded.
  • The snapshot of a composition when pushed to a public repository in GitHub is stored at a desired remote location which, is identified by a name, such as “origin” and within a desired branch. The first branch is named “master” by default. The composition can also be sent to a different remote location and branch using appropriate commands while using git.
  • Any user who intends to make a copy of a composition within a public repository can clone the repository. This allows the user to download a version of the repository along with its contents to their local workstation. A user can also fork a repository which creates a copy of the composition as a public repository within the user’s GitHub account.
  • If a user intends to push a modified version of a composition to its original branch, he/she is allowed to make a pull request to the owner of the branch. This is not needed if the branch is owned by the user pushing the content. The owner of the branch can acknowledge the pull request and integrate the modified version into his/her branch to create a new snapshot. Thus a chain or stream of snapshots that are monotonic with respect to time are recorded and each author’s contributions to a repository can be identified distinctly by the public.
  • The owner of a branch can reject a pull request. At this stage the user who believes their version of the composition deserves to be published any ways, can create a new branch in the original repository. Thus the public can transparently jude which branch stores a more efficacious composition. Any person from the public can continue to work with a snapshot derived from any of the branches within a public repository.
  • Any two branches can be merged in a way such that conflicts are managed and reconciled among the different versions of a composition.
  • Users can even create an all together new repository and branch via forking. Upon doing so, the original source of the composition can be easily cited because a permanent link to each and every snapshot of every composition is always available. Thus, the streams of snapshots transacted between users is very much like the “block chain of crypto currencies“. This is a fundamental and a very important point to note because a composition pushed to a public repository via git is essentially a token representing the value generated within society by the authors’ efforts to create artifacts and engage in transactions of such artifacts.
  • Any publicly available snapshot can be cloned or forked while maintaining an indubitable public record of such transactions. Thus, if it is ever found that a new composition was less worthy than a base composition it was cloned or forked from, the users have an opportunity to retrace their steps and utilize an older snapshot in their future endeavors. As such, users are not forced to waste their efforts in restarting from scratch and the chain of transactions continues.
  • A caveat in using this method: using GitHub and git has a learning curve and there are various nuances of this technology that can frustrate a novice user. One needs to consider this technology’s usability in the context of learning a new language or a new skill. One may feel inadequate or inept at the beginning but proficiency comes with practice.

Advantages of adopting GitHub based REP

For emphasis, I reiterate, GitHub provides a free infrastructure already poised for implementing REP1. It uses one of the oldest and highly successful version control technologies known as git which, practically creates cryptographically secure block chains of composition transactions. Thus, from the block chains, the value generated within society through transactions of digital compositions is sharply evident from an economic and anthropological perspective. The more poignant advantages of this technology are explained by first analyzing the critical issues within the current industry of paper journal publishing:

  • Currently, peer reviewed journals recruit a few referees (often three individuals) to judge the veracity and publishability of a submitted manuscript. During this process a mechanism may be employed wherein the authors of the manuscript are not known to the referees nor the referees known to the authors. This mechanism supposedly, allows the referees to consider the manuscript on the basis of its contents rather than the social attributes of its authors. And when comments about the manuscript are sent to the authors, the referees are protected via anonymity, from personal conflicts. This style of disseminating research has gone past its obsolesce because of these reasons:
    • Contents of a manuscript give away the research groups and potentially the affiliations of the authors as the topic, references, paradigms, methods and equipment reported in the manuscript can be very typical to certain groups of authors. For example, a very few number of people in the whole world use machine learning techniques for analyzing their data generated from a 7 Tesla fMRI machine for researching the topic of visuomotor coordination. If such a paper is submitted to a journal, it is nearly certain who the authors are. Also, various topics and experimental methodologies evade scrutiny from the larger public by being published in extremely niche journals. All these issues open up a distinct possibility for kinship or grudge based decision making on the part of the referees and journal editors.
    • The use of a mere two or three referees is absurd when a vast crowd can be recruited for free to consider a composition. It is not guaranteed that the few recruited referees are competent in judging a given composition even if they happen to be the most renowned investigators of the field. It is also not guaranteed that these referees will not deliberately dismiss a valuable line of research merely to protect their personal research and career motives. Recruiting a larger pool of investigators from across the world and from different fields can provide a more rich, deliberative and constructive form of peer review.
    • When the research topic cuts across fields or is a novel area of research, it is too difficult to find a competent referee to review the manuscript. Thus, it is necessary that the greater pool of researchers across the world be recruited as adjudicators for such topics. This pool might as well be recruited for analyzing any composition.
    • The comments about a paper from a referee are not available to the public for scrutiny. The comments are not guaranteed to be constructive and this fact may go undetected within the closed circuit correspondence between authors and referees. More importantly, making the process of critiquing a composition through discussions that are transparent to the public can help other authors learn ways for improving their compositions by preempting the type of critique they may come to see.
    • The time taken in waiting for referees to read and provide comments on a submission and the subsequent correspondences can take months or even years before the research is published. Such large scale delays in promoting valuable findings is unacceptable in our era where freely accessible technologies can support real time communications. Delays in publishing a research may squash the research.
  • It is very difficult to search different databases of journal papers simultaneously. One needs a generic search engine like Google. So the search tools within academic databases are redundant to Google’s search engine. Why then must anyone pay for a journal publisher’s search tools?
  • Institutions can save funds by not subscribing to academic databases such as EBSCOHost, JSTOR, Elsevier, Thomson Reuters, PubMed, etc. and by promoting their research through free public repositories. The most absurd aspect of a paid database is that researchers hand their composition to the journal for free along with its copyrights, referee other such compositions for free and even form the editorial board of the journal for free while the journal publisher makes money off of such efforts in perpetuity without ever transferring any portion of that monetary benefit to the authors. This, by definition, is slavery.
  • Many institutions cannot afford a subscription to sufficiently many databases and opt for only a few of them. This forces the institution’s members to get boxed into referencing only those few journals that could be afforded and are thus bereft of other important literature. Such exorbitantly expensive databases whose prices go up each year do not serve any valuable purpose to society anymore. Every purpose of paid databases can be exacted via crowd sourced research stored in free, public repositories. The money thus saved (millions of dollars per year) by adopting free and open source research can be transferred to students and faculty or utilized for infrastructural improvements. (See Harvard University Cannot Afford Journal Publishers’ Prices)
  • Persons not affiliated to institutions which have subscribed to academic databases face immense difficulties in obtaining the articles within the paid databases. The contribution of such persons to academics, who may be top researchers within private companies, is thus diminished.
  • Journal papers do not match up to the bandwidth of audio and video compositions in conveying knowledge. Such audio and video compositions are vital for research in music, journalisms, business marketing, communication studies, etc. A simple video about an experiment or procedure is likely to cut across language barriers and make the research more usable to persons across the world. Such digital artifacts obviously need to be stored in a digital repository which, the current journal publishing industry is not capable of providing and certainly not for free.
  • The language barrier is a special issue, it is notable that research done by English speakers seldom references knowledge from articles published by other speech communities. For instance, are the Chinese researchers along with their work in Chinese not peers of English researchers merely due to differences in language? The effort to bridge this barrier and generate translated versions of a piece of knowledge into various languages is not taken up by the journal publishing industry. This issue can indeed be handled by crowd sourcing the required effort if the research publications were free and open source and easily accessible from a public repository.
  • Often a piece of research work dies with the demise of the researcher. With free and open sourced research that is publicly available, the orphaned work can be taken up by any interested party and hence promote the sustainability of valuable research. When the work is more important than the worker, it needs to be preserved and promoted by others!
  • When there are multiple authors of a publication, it is impossible to delineate the contributions of each individual in the current approach to paper publishing. It may also be the case that while one individual’s contributions were numerous, another’s, though few, were more crucial. It may also be the case that a popular author’s name was included merely for gaining social traction and feigning legitimacy. The most fine grained analysis of contributions to a composition to the level of analyzing authorship of each symbol in a composition, is possible via GitHub’s tracking mechanisms. Of course, one may hack this mechanism by usurping someone else’s identity at the level of the local workstation but doing so is a federal crime in many countries.
  • It is notable that special interest groups that influence publishing houses can block the efforts of minority groups due to the special interest groups’ biases pertaining to gender, region, religion, economic class, etc. This is perhaps the most significant and critical hurdle to societal improvement that can be overcome by free and open source development practices.

Conclusion

The reasons explained in this article illustrate that the very concept of “peer review” has been compromised and undermined by the current scenario of hegemony, cronyism and incompetence among many of those who control research publication and distribution channels. The economic utility of the existing processes of disseminating research via paper journals is negative!

In sharing our research and development endeavors, we can save lots of trees and many more resources to improve our ecology by adopting GitHub or an infrastructure like it. GitHub is already in a position to overcome the above mentioned flaws within the current journal publishing industry. It also has indubitable tracing mechanisms to track the development of a piece of work submitted to GitHub. We can also expect continuous, real time, enhancement of compositions pertaining to any area of interest using such technologies which is simply impossible within the current journal publishing industry.

References

  1. http://git-scm.com/
  2. https://help.github.com/articles/fork-a-repo/

Source code for REP

https://github.com/samkhan13/REP

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s