Development Research in Practice: The DIME Analytics Data Handbook. By Kristoffer Bjärkefur, Luíza Cardoso de Andrade, Benjamin Daniels, and Maria Jones
[x] I softened the recommendation for Overleaf, since we don't necessarily strongly recommend it for DIME teams. Would be good to briefly discuss
[x] i think the note that we used LaTeX/Git to prepare this book would be better in Chapter 7 than where it currently is in Chapter 6 (since that dynamic docs section focuses on intermediate outputs) @bbdaniels
[x] i found the usage of output to mean variously tables/figures/research papers to be confusing. i made revisions to try to clarify. - Luiza used code output vs. research output
Preparing and publishing data
[x] de-identifying data for publication subsection: there isn't a very clear flow between paragraphs in this subsection. i made substantial revisions. Also, the first paragraph of the next subsection (publishing raw and constructive datasets) was repetitive so I cut from there and combined with the last paragraph of de-idnetification. it would be good for someone else to read through and make sure that the logical flow of this sub-section is now clear.
Preparing a complete reproducibility package
[x] add link to our reproducibility checklist
[x] add link to DIME standards
General
I made a lot more revisions to this chapter than others, so it would definitely be good to have someone take another pass
Need to add a final paragraph tying it together and transitioning to the conclusion
[x] Very focused on academic publications, but this should include dashboards, infographics etc and other types of research products useful in the context of development.
Add paragraph to alternative types of output, not just academic publications. Not go in to details as we do not have much resources, just point to good DIME examples. Say that this is a good way to disseminate development result to non academic audince @luizaandrade #500
[x] page 108 -pandoc example - where does this example come from? who has quality controlled that this is the best way to do this? @bbdaniels - remove and move to wiki
[x] page 108 - "This makes it possible to set up a dedicated GitHub repository for each output and manage contributions and version histories and manage contributions and version histories using the same system we recommend for analytical work. using the same system we recommend for analytical work." - sounds like we are recommending setting up a repo per output. Do we want to do that? 14993fff79ed8f8b14f4d11246036c8547b796b2
[x] page 109 - Internet or internet? I know Internet is technically most correct but I am always a bit annoyed on people that capitalize Internet. - non capitalized @kbjarkefur bd66879246b6eaa4b5b864634adbea741525666f
[x] page 112 - Options for public release. Mention and link to WB options only, and then link to wiki page with more options?
@kbjarkefur will rewrite so that it shows that datahub is part of dataverse c2048692d794a586d32cd5cbfcf83fb7ea9ef00f
Preparing written documents for publication
Preparing and publishing data
Preparing a complete reproducibility package
General