Interview with Joshua N Pritikin

Joshua N Pritikin has recently started a peer-review service based on free software he developed. Being the editor of a magazine about free software, the idea immediately intrigued me. So, I asked Joshua a few questions. Here are his answers.

TM: Hello Joshua. You’ve been involved in the free software community for quite some time... please introduce yourself to our readers!

I authored the Event module for Perl in 2001 and I have implemented Linux Terminal Server Project at C&M Poultry’s head office in India. However, I have not been as prolific as some others in the community. Perhaps I have spent a lot of time pondering how to improve society and proportionally less time actually coding. But I do believe in Mr. Stallman’s vision and when I decide to get behind a software project then I put my heart into it.

TM: You recently contacted me about your latest project, Peer Proofreading Exchange. What is it about?

Over the last few years, I’ve been writing a lot in connection with my interest in experimental psychology. Psychology is a challenging topic for me and I have always been helped by feedback on my writing from other people. I have noticed that my writing improves radically when I get other people to read and comment on it.

Then with the One Laptop Per Child project on the horizon, it dawned on me that we are soon going to have a lot of more writers in cyberspace and not enough teachers. It will be a practical necessity for writers to solicit peer reviews. I knew I could quickly build a web site targeting this need so I did. Perhaps the proprietary program trading software I built Deutsche Bank in the late 1990s influenced the theme and design.

TM: It’s very similar to the concept of free software, but working on papers...

I think it is similar in the sense that a key software development tool, diff (and wdiff), is prominently featured on the site for comparing the original and revised draft. I believe diff is far superior to the “Track Changes” feature of word processors.

On the other hand, the concept is dissimilar to free software in that the exchange tries to respect the anonymity of authors and reviewers.

TM: How does the reward system work exactly?

When you complete a proofreading assignment then the number of words in the manuscript is deducted from the author’s account and deposited into your account. The idea is that you are obliged to proofread as many words as you ask other people to proofread. There is a bit more to it though because skill in writing is not equally distributed. Manuscripts are rated on a coarse grade-level scale and authors rate the quality of revisions. The idea is to match people of roughly equal skill and commitment in order to minimize abuse and disappointment.

TM: Now, the interesting part: the software you are using is fully available under a free license. Is that right? Has releasing the code been beneficial?

Yes, the software is licensed under the Affero General Public License which is nearly identical to the GNU GPL except for the addition of section 2 (d). Section 2 (d) adds the requirement that if you (as a user) interact with the program over a computer network then you have a right to the program’s complete source code. All the ancillary library code is free software as well.

For my purposes, the benefit of the Affero GPL is not that you can run your own copy of the exchange (although you could). The efficiency of the exchange is proportional to the number of members so there is no advantage to having more than one. The more members, the faster the turnaround on editing.

At the time of this interview, the exchange has not reached “critical mass” yet in terms of membership. I imagine that critical mass is about a thousand contributors. With that many contributors, I anticipate a trading volume of 3-5 manuscripts per day. That should be enough to keep everybody thoroughly entertained. As it is now, you can submit a manuscript and wait for days without getting a single review.

Oops, I am going off on a tangent. To get back to your question, the benefit of the Affero GPL is that you can study how the reward system works and insure that it is as fair as possible. I doubt most people are going to invest a lot time in the demanding process of meticulous proofreading unless we all have confidence that the reward system is actually fair and that there is no practical way to cheat or game the system. The best way to nurture that confidence is complete disclosure.

TM: Thank you for your time!

It is my pleasure. Thank you for your interest.