The House of David

"dawnbreak in the west"

Thursday, September 06, 2012

Refactoring to publons

I am a bookish sort and have a side interest in the physical sciences, so I am always up for learning about new elementary particles: the top quark, the Higgs, the administratium nucleus, whatever. So I stumbled onto a scholarly article just today: Publons. Peter Kovesi (and James Trevelyan) posted this, its last version anyway, in 10 November 1998 (7:06:42 PM). h/t Wikipedia. (The two footnotes in that article seem not to lead anywhere. They just show that the essay was written between 1991 and 1998.)

Spoiler alert. Read that link first.

Back now? (and still here?) Okay. Here is my first thought: simply amazing. Second thought: WHERE HAS THIS BEEN ALL MY LIFE. Because I've been doing "publons", actually published or not, since I first got the posting-original-research bug in... 1997.

As for publons themselves, also called LPUs, Condensed Concepts praises them here. And I can see why. I write code for a living. I don't get graded on the lines of code I generate; "it is like judging a battleship's seaworthiness by weight", as I've been emailed. I get graded on the number of class definitions I write...

Oh wait. I don't. I get graded on quality. Academics get graded on class definitions. Hence, Kovesi-Trevelyan. But this still does not obviate the virtues of object-oriented architecture (done right); likewise, that people snark about "publons" does not mean that publons are evil. In fact publons are exactly what prospective scholars should write. Condensed Concepts is correct.

So I'd like, here, to offer how I, personally, refactor my projects to publons.

It is a truism of original research that it generates footnotes. A footnote is a digression in the project's argument. Sometimes the footnote becomes very large. This is a code smell. Specifically, this is a hint that here is another avenue of research that you want the reader to accept, before you finish writing up the rest of your stuff.

Maybe I could reduce footnotes by adding a new section / lemma to the project, but this isn't common. (At least, not for me. If it's a footnote, it's not really part of the project - it's an annoyance that I have to swat off, so I can get to the project.) Instead the typical avenue of research that most footnotes describe is "go read this article by someone else first". Sometimes they describe "go read lots of articles first"; these boring lists also have to stay as footnotes.

Less often (than I'd like) someone else - or many someones else - has/ve already organised the stuff for me; at which point, I link to the most comprehensive summary/ies I can find and then just pick out the highlights. As an example, we have the (numerous) summaries by Hoyland, Jeffery, Périer et al. that the Qur'an - rather, a predecessor, "the Furqan" - was put together under al-Hajjaj's direction at the earliest 700s AD and then secondarily ascribed to 'Uthman in subsequent decades. As we can see, all serious Orientalists from Jeffery to Hoyland have stated agreement with this hypothesis. So there is no point, for me, in doing anything more than just sticking their stuff into a footnote too. (Blonde joke!)

But sometimes I have to extract the footnote and make it a separate project.

I think the first publon I ever did was that "John and Enoch" HTML page, in 2000; but I am no longer posting such juvenilia around here. So another example, which I have at the Madrassa, is "A Test for the Jami' of Hakam from Shu'ayb". Come to think of it, its mother-essay was another publon, "The ‘Parable of the Workers’". All I'd wanted to do was to find out when sura 57 was written. (That's "Monks, Muslims...".)

So, yes; we have to refactor our "code" - our projects - to the least publishable unit, down to the most nitpicking publon. Otherwise what we send out is going out as a massive unreadable pile of undersourced assertions like John Wansbrough's "Qur'anic Studies" (at worst) or else as a mess of footnotes, each of which will have to be picked apart by somebody. Wansbrough got away with it; but he shouldn't have, and you won't.

posted by Zimri on 19:46 | link | 0 comments

On this site



Random crap

Powered By Blogger TM

Property of author; All Rights Reserved