Wednesday, May 18, 2005
There are several problem with the table-based user interface show in my last entry:
- If scanned documents do not have any coding data associated with them, the table view is worthless
- The original context of a document (where it falls in the stack, how it is bound, etc) creates meaning for its reader and is lost in the table interface
- In a table-based design, the image of the document is so small as to be unusable.
After thinking about the table based interface for a few days, I decided that I should either scrap the whole project or come up with something much better.
I started over from scratch, this time with a pen and paper instead of sitting in front of a computer. I drew a design that restored the document image to center stage and displayed its original context using a row of graphical icons at the top of the page. The document data, which used to be the primary way to find a document, is now sent over to the left side of the screen. I also wanted to create use thumbnails at the bottom of the screen to give the user an idea of what pages come immediately before and after the current page.
I was much happier with this design as it would more closely match the physical way of handling documents and has a more more document influenced design rather than a database influenced design. However, implementing this radical new design brought new design challenges and technical issues to the project. I'll show you the next phase tomorrow.