My immediate reaction: - <F2> <Shift-F2> etc are really great and will make the form creation go much more quickly than using the mouse - I like the tagging in the editor and visual indicator showing where a field starts and ends - editor does not mind that a file is open in Manager (good!), but each should recognize when a file has been changed by the other
- do Labels need to be treated the same as Fields in recxml? All they really need is location (colour, font eventually?) - paste from Qes identifies labels well, but does not place a label following a field on the same screen line as the field e.g. v10 temperature ##.# C will place the 'C' on the line following v10 - it is hard to place labels on the same line as a field using the mouse - need 'snap' for labels as well (and have them obey font changes) - vertical align might support the snap settings also
- placement of screen elements will be a challenge and this is a great start! I wonder if recxml format should allow for the following: * within <name>, the ability to show the name or not (seems to be 'show' now) * 5 elements for each field: name label that precedes the entry field entry field label that follows the entry field extended explanation of the field (accessed by the help key) * with each of the first 4 elements to be placed horizontally, using <align>, relative to the screen border (for name) or entry field (for name and labels)
No crashes or unexpected behaviour on my PC. My MacBook is coming next week and I look forward to trying it out.
Jamie Hockin
The EpiData Data Manager will gradually replace the existing EpiData Entry software as development goes on.
Why change to EpiData Manager ? Since release of first EpiData software in 1999 many aspects have changed. The main reasons for development of the new strategy are: • New users are increasingly ”graphicly” oriented and focused on mouse usage. ◦ not understanding the well known qes-chk-rec principle • There is a need for multiplatform development (Linux, Mac, Windows, PDA …..) • Font management is changing to UTF-8 to avoid nationalisation problems • We wish to implement GCP (Good Clinical Practice) required for many medical data projects. This demands encryption and logging of editing at a very detailed level. • There should be a common ”engine” for all EpiData software, such that handling of data and metadata (labels etc.) are based on the same internal routines. • When doing data entry it should not be immediately easy to change rules or structure for data entry personnel. Therefore a dedicated EpiData EntryClient will be developed – serving only data entry.
How will a project manager see the new strategy ? The work process is more clearly divided into what the manager does and what the data entry persons do. The EpiData Manager is a tool for the project person, which: • Defines data structures, adds metadata (labels and definitions), documents data and exports for analysis. • Updates data with new fields, changed formats for fields (e.g. More decimals) and is used for control of data. • There is a choice of using computers with either Linux, Mac or Windows
A short introduction document is found next to where the test application can be downloaded. See http://www.epidata.dk/testing.php
Apart from fixing many bugs, the new version now includes an editor, a revised interface and allows for pasting from a qes file or the editor.
Next phase is to add checks and develop the Entry Client.
For the planning please comment on the new release, which includes an editor. Also we should have a discussion on which are the "core" check commands that almost everyone uses and which are enhanced.
regards Jens Lauritsen EpiData Association