[EpiData-list] EpiData Analysis build 134 ready for testing

epidata-list at lists.umanitoba.ca epidata-list at lists.umanitoba.ca
Sun Apr 20 10:47:20 CDT 2008

Many minor problems have now (hopefully) been solved in build 134 of 
EpiData Analysis. Which is available from the usual testing page: 

Some issues remain:
- small data / zero cell inaccuracies (Most likely not solved, but 
better information should be developed)
- update had to be disabled in this build
   - a major rewrite is needed before stable function
        use EpiData Entry for changes to data instead.
- some wishes reported to Mantis database

- The user base seems divided in where to put some aspects in the menu 
  - should logfiles be in file menu or results
  - should edit setup ini be in file menu or edit

Issues solved: Version 2.0.0 build 134
  * Extended or adapted function
    * CIPLOT: adapted to changed order of variables and defaults
           new plot with confidence intervals for proportions
           (e.g. cases in subgroups of other variables)
    * BAR and Histogram: Y axis can be scaled as percent: Option /PCT
    * Weight variables can be non-integer in tables
    * merge accepts comparing upper string with string key
    * Update disabled this build (inconsistent behaviour must be 
  * Documentation
    * Command reference updated (F1) 
    * added quick guide (see F1)
  * Bugs
    * Upper string variables written as standard string
    * Various bugs removed  

For those wishing to test or comment on the adapted menu system, please see
www.epidata.dk/testing.php to get the updated build.

Following testing we are approaching release as v2.1 with first 
translations in Spanish and French.

I do acknowledge in particular Gilles Desve for the adaptation of the 
multinational translation module and
Jamie Hockin, Pedro Arias, Peron (Juan Carlos Merino), Bjørnar Nyen, 
Hans L. Rieder and Neville Verlander for discussions and inputs. But 
also the EpiData-list discussants in general.

In the previous year the Association of Public Health Epidemiologists in 
Ontario have established an important expert panel, which have been very 
active in the specification of the CIplot command and other recent 
aspects. More on this latter collaboration will be announced later.

Update is not permanently disabled, but has shown many problems. 
Therefore I am considering NOT to wait a full rewrite of update, but 
instead disabling update, to be able to release v2.1 with translation. 
Comments welcomed on this aspect also.

Regards Jens Lauritsen
EpiData Association

More information about the EpiData-list mailing list