I see OpenRefine mentioned a lot here, but I don't see it doing much that R and others can't. What capabilities does it offer that I'm not seeing in the promo page that R or other data packages cannot?
I'm a programmer and I use OpenRefine all the time. Some of the advantages it has over breaking out Python or some other language include:
- results of transformation expressions are previewed interactively with live data
- quick, interactive, filter facets which allow for easy browsing of instances/rows which match a variety of filters
- exploratory analysis of data to do quick visualization via facets and explore interactions among columns
- reconciliation of text data against reference data services containing strong identifiers (Freebase, OpenCorporates, any SPARQL or RDF, etc)
- simple linking of reconciled entities to other info sources like Wikipedia, MusicBrainz, IMDB, etc
- complete provenance/undo history of all modifications
- combination of machine smarts and human review for tasks like clustering of names.
- wide variety of input & output formats including both file formats and online repositories like Google Spreadsheets & Fusion Tables
- one click selection of record boundaries to produce a grid of data from a JSON or XML API is great for exploring new API endpoints
And I'm finding more cool stuff all the time. Just the other day I discovered that the scatterplot facet rotates 45 degrees allowing me to select any area on the diagonal of an OCR character accuracy vs OCR word accuracy scatterplot to investigate in more detail.
Disclaimer: I'm the project lead for OpenRefine, but most the good stuff was done by the original author David Huynh.
If you're comfortable with R, probably not a whole lot. OpenRefine's sweet spot is for facilitating data management for non-programmers, and packaging together a bunch of common data-munging tasks behind nice point-and-click interfaces. Sunlight (my employer) has used it a fair bit for this purpose, if researchers need to be able to clean up some data without writing any code. It also has some nice extensibility facilities that make it relatively straightforward for programmers to add functionality or integrate with existing data services and expose these to less-technical researchers as well.
I think the best benefit from OpenRefine is its GUI. You can always do everything OpenRefine offers you with Python, Java, etc. but for non-programmers it helps them to perform basic (and sometimes not so basic) operations with data without having to learn how to code.