Sign in or Join FriendFeed
FriendFeed is the easiest way to share online. Learn more »


Ph.D. in English literature, singer-songwriter, technology enthusiast.
New Analysis Shows Problematic Boom In Higher Ed Administrators | New England Center for Investigative Reporting -
Over the past 25 years, the number of university administrators has more than doubled, adding 517,636 positions: - Yoni Appelbaum (YAppelbaum) - Amanda
How to Disappear Completely: On Modern Anorexia -
Work (painting) - Wikipedia, the free encyclopedia -
@ariddell What a great passage. But I doubt the goddess Sloth looked kindly on Carlyle. He is emblematic of Work. - Amanda
RT @librarythingtim: @amandafrench Regex doesn't need special justification. It's a powerful tool that, once you know it, you use all the time. Like algebra…
I once taught a regex workshop and a literary scholar asked me “What would you use this for?” This is an answer. Search. #digitalcrucible
So, for instance, searching EEBO for “What an X am I” can find potential ancestors to “What a rogue and peasant slave am I” #digitalcrucible
The talk by @danashore is an argument for sophisticated search as a simple but powerful form of text mining. #digitalcrucible
RT @peasandpoetry: .@DanAShore closes out #digitalcrucible with "Cyberformalism: Search and the History of Linguistic Forms."
A single backend (& corpus) should be able to support multiple presentations, says @benmschmidt #digitalcrucible
BookWorm is in many ways an attempt to improve upon Google nGrams: you can create your own corpus, for instance. #digitalcrucible
Many large scale analytical projects in history analyze only metadata, not full text. #digitalcrucible
Tool talk from @benmschmidt on the BookWorm tool. See #digitalcrucible - Amanda
Tool talk from @benmschmidt on the BookWorm tool. See #digitalcrucible
Says that to reproduce @mljockers work, for instance, he would need his corpus. Strong plea: “Make your corpora available” #digitalcrucible
Specific problem they faced: some epigraphs marked as <epigraph> and some as <quote>. #digitalcrucible
“Cleaning up the data is 80% of the work; analyzing the data is 20% of the work” - @lincolnmullen #digitalcrucible
Trying to remember that great @lincolnmullen quotation about how most of the work is cleaning the data. :) #digitalcrucible
Hearing from NYU English grad students about attempts to study epigraphs in MONK XML texts. Hard to get good code, results. #digitalcrucible
RT @peasandpoetry: Next up, @AaronPlasek + Rob Koehler on Mediating Genres of Prestige, Credit and Authority #digitalcrucible
Text mining often indicates very sharp spelling change for some words ca. 1630. Not gradual. Others peak earlier. #digitalcrucible
Nice. The Early Modern Print database allows text mining of early English texts: #digitalcrucible
Big spike in number of books published circa 1640 was the real beginning & cause of standardized spelling. #digitalcrucible
A 1569 “Orthographie” by Mulcaster was already recommending some spelling standardization, but took time to catch on. #digitalcrucible
Basu makes the point that standardization and modernization of spelling are not the same thing. #digitalcrucible
Now hearing from Anupam Basu on his large-scale data analysis of the Early English Books Online database. #digitalcrucible
@librarythingtim Explanation of sampling method or at least source here: #digitalcrucible - Amanda
Wikipedia:The Wikipedia Library - Wikipedia, the free encyclopedia -
@ariddell Links from dinner convo. Reading Experience database: and @WikiLibrary - Amanda
The Reading Experience Database 1450-1945 (RED) -
@ariddell Links from dinner convo. Reading Experience database: and @WikiLibrary - Amanda
Kirstyn Leuner Day of DH 2014: Stainforth | Just another Day of DH 2014 site -
Leuner blogged about her project for Day of #DH 2014: - Amanda
Other ways to read this feed:Feed readerFacebook