1
0

Update on Overleaf.

This commit is contained in:
2023-02-22 19:00:30 +00:00
committed by node
commit c5e0a01713
133 changed files with 24196 additions and 0 deletions

35
TODO Normal file
View File

@@ -0,0 +1,35 @@
DONE 1. As IRR is not even a major focus now: on Page 8, just keep the citation to "the Bible" (Krippendorf 2004) and remove all the other single/double-serving references: Gwat; Krippendor
2007; Hayes & Krippendof.
DONE 2. On page 9: I know you are super animated by the Wikipedia example. But I would argue reviewers would be more animated by the Perspective API. But whatever your choice, I would suggest
keeping just one: Wikipedia / Perspective.
DONE 3. I don't mind you cut Dobbrick et al. and keep the big name: Boukes et al., 2020
DONE 4. pilny_using_2019 doesn't sound very important.
5. This "citation dragon": \citep{obermeyer_dissecting_2019, kleinberg_algorithmic_2018, bender_dangers_2021, wallach_big_2019, noble_algorithms_2018, gillespie_custodians_2018}
bender_dangers_2021, wallach_big_2019, noble_algorithms_2018, gillespie_custodians_2018 are single-serving. We must cut Tarleton Gillespie at least to save space, he has enough fame and
doesn't need our citation. Maybe we can keep bender_dangers_2021, it's meaningful. Yes, bender_dangers is my favorite of these.
6. another dragon: \citep[see]{carroll_measurement_2006, yi_handbook_2021, fuller_measurement_1987, buonaccorsi_measurement_2010}
?cut fuller_measurement_1987
7. I don't recommending removing the DAGs. Those lazy reviewers can skip the descriptions of the simulations and read the DAGs and pretend to be understanding the whole paper.
By considering the above, probably we can cut around 1 to 2 pages (~37p).
8. There are also some room to cut in the recommendations
8a: Could we cut the caveat about the API and PL in R I? We don't want to promote the usage of PL and it would save 1/3 page.
8b: I trimmed some words in R II. I think R II is quite compact already and it is super important.
8c: Strategically for this ICA submission: Should we just promote MLE in R III, i.e. don't say anything about adding another adjacent method? Other than that, R III is quite compact also.
8c: Recommendation IV: Cut the last paragraph for this submission. We are facing the same problem and learn from us.