I’m writing a paper. More about its content in future posts.
So, ideally, I want to either link to myself or “plagiarise” myself. I already wrote the stuff I want to say, and the text is mine. The intro could go one of two ways. In its entirety:
As previously discussed, it is useful to examine the association between data archiving and citation rate [Piwowar 2007]. This study re-examines this association based on a much larger dataset.
or, since the version above isn’t very self-contained, I think it would be better to do this:
“Sharing information facilitates science. Publicly sharing… include verbatim text from previous paper inside quotation marks” [Piwowar 2007]
This study re-examines this association based on a much larger dataset.
Such clear text verbatim quotation is rarely if ever done, afaik. It should be: it would save everyone a lot of time and hassle. Frankly since I kept copyright the self-attribution is probably optional, but it helps for clarity.
With gold oa, I think it is no problem. “Self-plagiarism” is totally fine in this case. It isn’t a copyright violation and it isn’t without attribution. Is there any reason it isn’t fine?
Admittedly it would take people by surprise. For that reason, on the first paper or two where I do this, I think I’d want to add this footnote:
Note: Rewriting text for the sake of variation is a poor use of resources. The cited text comes from an article licenced under CC-BY, eliminating all concerns about fair use.
Blogging this now because there is a chance that in rewrites of the manuscript we’ll decide to change the paragraph and not use it verbatim… but I wanted to say that I THINK THIS IS A GOOD IDEA IN GENERAL.
There has to be substantial new content, of course — I’m not talking about duplicate publications here. In normal ethical papers there is a lot that is redundant. Intros, methods, limitation sections… so much room for reuse . Let’s spend time doing more science and less rewriting.
I’m far from the first person with this idea (there was a Friendfeed conversation at some point musing about how a paper could be a link to a prev paper for the motivation and lit review, another link to somewhere else describing the methods for this novel application, a link to a dataset, and then the body of the paper itself could be a few figures and a discussion)… but I gotta say — I really want to DO it.
Some say that for self-plagiarism “the best thing you can do is avoid it completely.” No way. Let’s push the envelope to where it can and should be.