2006 Fiscal Year Final Research Report Summary
Development of a spreadsheet environment supporting trial-and-error exploration by flexible combination of processing
| Project/Area Number |
17500053
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Media informatics/Database
|
|---|
| Research Institution | Hokkaido University |
|---|
Principal Investigator |
LUNZER Aran edward Hokkaido Univ., Grad. School of Information Science and Technology, Specially Appointed Associate Professor, 大学院情報科学研究科, 特任助教授 (50374608)
|
|---|
| Project Period (FY) |
2005 – 2006
|
| Keywords | computer systems / information systems / user interfaces / subjunctive interfaces / spreadsheets / Web application linkage |
|---|
| Research Abstract |
Spreadsheet environments support users in creating ad-hoc applications, and exploring the outcomes of alternative choices within those applications. However, spreadsheets typically limit the kinds of exploration that can be carried out easily on a sheet: once a sheet's cells and formulas have been defined, the cells' values are easy to change but the formulas are relatively fixed. This asymmetry is unhelpful for, say, a researcher who wishes to explore variation not just in the parameters to a simulation, but also in the calculations used to implement the simulation.
In this project I developed an extended spreadsheet model, and associated interface, such that users can explore a variety of calculations as easily as they can explore a variety of inputs to those calculations. I then implemented an environment that incorporates the model. The environment also includes locally developed mechanisms that allow processing offered by form-style Web applications to be used within the calculations on a sheet.
As a concrete example of what the.implemented environment supports, a bioinformatics researcher could place onto a sheet a homology-search facility captured from one Web site, and connect it on the same sheet to a phylogeny-tree generator offered elsewhere. Along the way, the results from the homology search (a list of protein sequences) can be viewed in a cell and selectively highlighted by the researcher, so that only chosen sequences are passed to the tree generator. Similarly, the researcher can set up a cell supporting selection among a list of tools for generating the tree. Thus it is equally straightforward to explore a range of sequence sets or a range of analysis tools. We believe that such flexibility will provide benefits in a wide range of computing domains.
|
