A list of cool “features” or techniques that I’ve seen in papers.
Direct PDF linking. The DOI isn’t there, which isn’t great though (link could break?)
Looks like this:
From the HamPath user guide: http://www.hampath.org/wp-content/uploads/2016/09/user_guide_hampath_V3.pdf
In legal documents, footcites containing quotes from the work cited
Acknowledgements on first page, funding on last
Lin RR, Assistant R, Kascak AF, Professor A, Alexander RM. Active Control of Transient Rotordynamic Vibration of Optimal Control Methods,". 1988. doi.org/10.1.1.1052.3537
Detailed and exhaustive nomenclature. Might be improved by adding a definition for each part.
“They’re communicating to the public” sgu 832 title
Determination of bang-bang controls for large nonlinear systems, David Donald Niemann
Has an entire chapter devoted to “conclusions and recommendations”, which also includes key drawbacks. I like this layout.
Google scholar only indexes papers (outside accepted) with explicit biblographies. My comon
What a pleasant chart theme.
Technical abstract / plain language summary
In fact, could the summary be written by other people - the reviewers, say?
One thing to keep in mind is when scanning in old documents to institutional repositories, the OCR of some scanners can completely mangle math
Darwin usually speaks in the first person plural when analyzing empirical evidence he has collected and only uses the first person singular when he is specifically speaking about his own actions, such as, “…many special facts which I have collected,” or when he is speaking about his own qualms, such as “I am well aware that there are on, on this view, many cases of difficulty, some of which I am trying to investigate.” However, when analyzing his evidence, he always uses “we”, such as “we notice”, or “we understand”(2). Darwin’s change in footing when he is explaining his theory places himself and the reader on the same level and makes him a more “humble” presenter, allowing us to suspend disbelief for at least the time being and trust him.
GPG sign papers (maybe only the abstract) that I publish in a key-signing party - ask security.stackexchange.com how best to sign a paper - maybe tie with ORCID?
arxiv - latex rewrite, can’t do signed pdf
X.509 - researchers in low-income countries may not be able to afford a new cert every year? pro for gpg
ORCiD could have the public key, except then there’s hardly any trust. Workflow would be
- get public key from orcid
- Decrypt abstract
authentication [full paper abstract][researcher name][researcher orcid][something distinctive about the researcher - last paper, thesis title, perhaps?]
actually, having a way for things to be validated after they’re published doesn’t seem very useful - the researcher will usually put the publication on a page they own anyway (university institution site, etc), and you can just ask them if they wrote something.
The main way this would be useful is pre-publication for the journal to verify. An alternative to signing would be to put a challenge-response code on a page that is known (a university webpage accessed from a past paper, say).
A fraudster could set up a fake researcher page to dupe the journal. I think the whole thing relies on the key / orcid pair being key-party signed by other scientists not associated. ( that imposes some headache on non-institution members… could make the code of conduct such that you must verify anyone that asks regardless of affiliation )
The sad truth is the vast majority of people signing GPG keys are not experts in detecting document forgery. I am quite convinced that with a half-decent false passport it would not be difficult to get many well-connected and widely trusted people to sign your key.
You’re right: take one of those barely secure pre-2017 Italian Identity cards (typewritten and a stapled passport image on folded cardboard) to a key signing party, and probably nobody will even care.
actually, this is a downside of anonymous peer review! The burden of authenticating the author is on the editor, rather than the reviewers; the reviewers cannot spot a forged author.
How is getting a CA to verify government ID superior to the journal itself verifying? The cost is to the journal; because the chaff is such a big cost, they can’t afford to spend as much effort. On the other hand, a digital certificate signed by a CA is valid for any journal, so even if they reject and you go to another journal, they don’t need to validate again. I trust CAs slightly more to be scrupulous than journals.
I guess one alternative would be to first get reviewers to review the content, then unblind and get them to verify identity.
Add PROCESS.md with instructions on how to repeat, guidelines on validation requirements, things to watch out for, etc
add automated unit test with virtualenv to check that installation process goes without a hitch - check python version
Adding line numbers to preprints makes text difficult to select and copy. The same is true of two-panel layouts, which usually copy improperly.
Nomenclature with units
Everyone says that their research is “non trivial” and “a novel contribution”. screw that.
An abstract with background/objective/methods/conclusion - neat!