00:01:31 jim rothstein: hello to ALL !
00:12:11 Jon Harmon (jonthegeek): readr got a speed-up under the hood a year or three ago: https://vroom.r-lib.org/
00:12:20 Jon Harmon (jonthegeek): https://vroom.r-lib.org/articles/benchmarks.html
00:12:25 jim rothstein: What is the `magic` in data.table ??
00:12:39 Shah Nawaz: data.table and data table package have same speed ?
00:12:55 Jon Harmon (jonthegeek): data.table is the package
00:13:08 Shah Nawaz: Reacted to "data.table is the pa..." with 👍🏽
00:13:13 Jon Harmon (jonthegeek): (not to be confused with data.frames)
00:14:57 jim rothstein: JSON - any comments?
00:15:25 jim rothstein: I loathe JSON
00:35:09 jim rothstein: (re: data.table performance ... ...) Very dated post, but data.table has "radix" algorithm https://www.brodieg.com/2019/06/10/base-vs-data-table/
00:35:25 jim rothstein: Radix: https://en.wikipedia.org/wiki/Radix_sort
00:35:42 Shah Nawaz: thank you
00:39:23 guslipkin: group_by(year) should be able to be re-written as write_dataset(partition_by = year)
^that’s not really allowed
00:42:05 Priyanka Gagneja: Reacted to "group_by(year) shoul..." with 👍
00:49:38 Jon Harmon (jonthegeek): I have to get the door, sorry!