this is for holding javascript data
Vijay Krishna Palepu renamed sectionsc_Implementa.tex to Implementation.tex
almost 11 years ago
Commit id: 6cfbbb6d4d79704e0042fc04b2cd1e25dc70a0eb
deletions | additions
diff --git a/Implementation.tex b/Implementation.tex
new file mode 100644
index 0000000..93fd044
--- /dev/null
+++ b/Implementation.tex
...
\section{\sc Implementation}
Our application is a web browser based application that does not require a server. This allows users to interact with the application, as long as they have a browser and at least two data sets from Google Trends. By doing a web based application we have a larger potential user base and effective distribution to all operating system platforms.
Users are able to interact with the application via placing the csv files into predefined locations that are labeled “Drop Files Here”. This cause the data set to be read by the application and display the time and volume of the data that is in the CSV file.
\subsection{Google Code and SVN}
The application is designed on top of HTML5, JavaScript, and CSS. The code can be checked out on Google code repo \url{http://code.google.com/p/3d-statistics-cube/}.
\subsection{\texttt{D3.js}, \texttt{CrossFilter.js}}
The application is web based also because of the incorporation of the open source JavaScript interaction frameworks Data Driven Documents (D3) [Citation Here]. Using D3 we are able to generate the graphical representation of the graphs. D3 also provides us the basic interactive framework that allows us visually select portions of the data set to be used in the t-test. Our initial goal was to incorporate D3 and CrossFilter to allow the user to explore different dimensions of the data set. In particular we can allow the user to discover what happens if we pick certain volume range with a user defined time range.
\subsection{Twitter \texttt{Bootstrap}}
The application is styled and in part made functional using the Bootstrap CSS and JS open source libraries provided by Twitter. These libraries enable uniformity and consistency for both the look-and-feel and the functionality of the site. The most notable use of the library was made in the implementation of the entire notification system that serves the ``Feedback and Discussion'' component of the application. This notification system is built on top of the \texttt{ScrollSpy} JS plugin from \texttt{Bootstrap} with final touches from the \texttt{D3.js} library.
It is worth noting that we have not performed any cross-browser-compatability tests for this application. The reason for the same is two fold. Firstly, this is largely meant to serve as a prototype that can be extended into a full fledged tool. To that end we have tested its functionality extensively on the Google Chrome and Mozilla Firefox browsers primarily within the Linux and Windows ecosystems. Secondly, the Bootstrap libraries are a popular framework and being used extensively by many organizations and companies for their own website solutions. Such large scale usage and adoption provokes a thorough community driven testing process, of the libraries across various browsers and operating systems. Due to the open source nature of the libraries any faults or issues detected as a process of such usage are likely to be fixed. We rely on this central notion that Twitter Bootstrap is extensively used and such usage guides the maintainance and enhancement of the library; and do not perform any cross browser testing for the application ourselves.
Not having to test the application for such issues and a readily available framework allowed us to develop what we percieve to be a fairly robust website to learn about hypothesis testing.
