These are chat archives for weaverbel/LibraryCarpentry

2nd
Apr 2017
Belinda Weaver
@weaverbel
Apr 02 2017 01:37
Can I say we should loop @brendam in to the python discussion as she told me she thought the existing LC lesson should be cut by half as there was just too much material to get through.
carmi cronje
@ccronje
Apr 02 2017 02:13
OpenRefine (advanced GREL) could help build a pathway to programming. GREL introduces learners to variables, functions, arrays, regex...it seems a solid stepping stone (and OpenRefine supports Jython). In terms of automating the boring stuff, I don't think you can go wrong combining APIs and GREL. I found that after the pilot OpenRefine workshop last year, participants wanted to learn much more about extracting data via APIs (e.g. OADOI, Worldcat, Ulrichs, Elsevier/Scopus). The tricky/powerful part is using GREL to slice and dice extracted data. I ended up including more API-GREL examples in the advanced OpenRefine episode but had to remove reconciliation and extensions due to time constraints.
Richard Vankoningsveld
@richyvk
Apr 02 2017 07:14
@remerjohnson true. That was kind of my point though. Python is a lot more than pandas. I think the python lesson should introduction e programming concepts more than it should teach how to use pandas. When Open refine lesson already exists. I know pandas and open refine aren't the same thing. But hopefully you get my point.
API usage would be a great thing to base some of the python lesson on too I reckon.
Owen Stephens
@ostephens
Apr 02 2017 11:42
+1 re "The tricky/powerful part is using GREL to slice and dice extracted data". GREL offers a way into CSS selectors and JSON parsing. If you move to Jython then xpath can be introduced
Can't believe Gitter turns x + p into an emoji
Jez Cope
@jezcope
Apr 02 2017 17:02
XP