he tightens the straps of his pack
takes a final sip of frivolity
checks that his bottle is filled with solitude
enough to slake any yearning for peace
dim the lights to hear the silence
shut the eyes to taste the stillness
soft sky drips
through the window
pasting light on the wall
sink into my ears
time sinks into a river
causes it to flood
(we paddle with the current)
then, at last
we stoop down
pick up the moon
and piece together the sky
Google Books Ngram Viewer released an update that incorporates changes that I made with two other interns while interning at Google this past summer.
It was a fantastic project that tied my linguistic interests with my computer science interests. My hosts were also beyond amazing - Slav Petrov and Dipanjan Das (in fact, my experience working with them was virtually perfect).
Summary of changes:
-support for wildcard searches, so you can search “the best *” and see what the top ten matches are, graphed accordingly
-support for morphological searches, so you can search “run_INF” to plot all inflections of the verb “run”
-support for case-insensitive searches
-revamped user interface
The project involved both backend and frontend changes. As a team we first worked to understand the backend - how to compute the data, how to store all these frequencies across so many years ranges, etc. Then we gradually moved on to the frontend and considered novel ways to display the data. There were also subtle features to continue support for, such as reliable URL sharing.
Overall, for me, Ngram Viewer 3.0 represents more than just an intriguing project at the intersection of computer science and language; it represents a grand experience this past summer where I grew tremendously.
A site I made years ago for a research lab at Syracuse University. Finally deployed (so the framework and code is a bit old). It’s developed in Ruby on Rails (2.3.x) and features a content-management system so a user can easily manipulate content on the page. In that sense, it’s similar to J-D RamPage.