http://blog.wordnik.com/carbonated-frequencies where we talk about Wordnik Fri, 20 Jan 2012 20:19:41 +0000 hourly 1 http://wordpress.org/?v=3.2.1 http://blog.wordnik.com/carbonated-frequencies/comment-page-1#comment-202 Erin Thu, 24 Sep 2009 16:16:20 +0000 http://blog.wordnik.com/?p=80#comment-202 Yep -- we're working with our partners to fill that copyright gap, but it's taking a little while. We hope to have good data in place soon! Yep — we’re working with our partners to fill that copyright gap, but it’s taking a little while. We hope to have good data in place soon!

]]> http://blog.wordnik.com/carbonated-frequencies/comment-page-1#comment-185 Sheelah Sun, 20 Sep 2009 22:06:31 +0000 http://blog.wordnik.com/?p=80#comment-185 Why do most words (except very common ones) have no bubbles at all between circa 1930 and 2005? Is that the gap between out-of-copyright books and the web as a source? Why do most words (except very common ones) have no bubbles at all between circa 1930 and 2005? Is that the gap between out-of-copyright books and the web as a source?

]]> http://blog.wordnik.com/carbonated-frequencies/comment-page-1#comment-106 Dr. Robert Amsler Sun, 02 Aug 2009 20:10:46 +0000 http://blog.wordnik.com/?p=80#comment-106 Couple Comments. (1) Project Gutenberg texts are notoriously lacking in provenance descriptions, esp. years. How are you using them to gauge the years? (2) Perhaps the Library of Congress catalog could supply some info on vocabulary by year by using their book titles and whatever else is available from their entries. It would have the advantage of being comparable per year (albeit with a growing volume (assuming) more and more books were published each year). (3) There is the American Heritage Word Frequency List, which ought to be usable in the 20th century for at least a snapshot of frequencies (as well as the other published word frequency counts, Brown-Kucera, Thorndike-Lorge, West's). Admittedly the problem of compounds not being reported as whole expressions hurts (i.e., 'New' 'York' 'City' as three words hurts). Since these were all dated snapshots it at least provides some input. Couple Comments. (1) Project Gutenberg texts are notoriously lacking in provenance descriptions, esp. years. How are you using them to gauge the years? (2) Perhaps the Library of Congress catalog could supply some info on vocabulary by year by using their book titles and whatever else is available from their entries. It would have the advantage of being comparable per year (albeit with a growing volume (assuming) more and more books were published each year). (3) There is the American Heritage Word Frequency List, which ought to be usable in the 20th century for at least a snapshot of frequencies (as well as the other published word frequency counts, Brown-Kucera, Thorndike-Lorge, West’s). Admittedly the problem of compounds not being reported as whole expressions hurts (i.e., ‘New’ ‘York’ ‘City’ as three words hurts). Since these were all dated snapshots it at least provides some input.

]]>