release candidate of Manager
Dear All
A "feature complete" version of EpiData Manager has been uploaded today - the expectation (read as "hope") is, that this version with bug removal can be released as version 1.0
* *NEW: *Create reports in HTML format using "Tools" * Default settings for a project * Change font and colour in program settings
The reports on project structure (question lists, value label lists, combined lists) are produced when you click on tools menu. All epx/epz files in that folder are included. (Notice they must be of most recent format).
For Entry Client a similar version will be released within the next days .
The following aspecs will be fixed/clarified before release: a. A Proper About box will be added. b. For reports a grid of included files will be shown to the user - just like "add structure" to allow for in- and exclusion of single files. c. known and reported bugs. d. any bugs reported from now on by users. e. updating of documentation on the wiki.
Another interesting development has taken place in the last week. We have localised an open-source based cross-platform testing system for interactive simulation of user behavior. We will spend some energy and time to develop validation tests to secure that the visual interface behaves properly after each new compilation. More news on this later.
Therefore please: Now is the time to challenge the software and report problems and viewpoints on the products - before releasing as v1. Documentation and instruction files could be more elaborate obviously - this will come later. In practice it has turned out, that unless features work almost without instruction, users will not use it.
Kind regards Torsten Christiansen Jens Lauritsen
EpiData Association Denmark
Explanation on reports.
The implementation of reports is now rather crude. Will be extended, e.g.:
a. The report window after creating the report is "fixed". Meaning you are locked to that report.
Will be changed, such that each report window is a separate entity.
b. Only the simple reports are included. Further documentation at project level (study information) and at field level (extended calculations etc) will be developed.
What you can test currently is the reporting format. When you create the first report a style sheet is created called "reports.css" which defines the various classes used in the report.
Examples:
h1 {color: blue; font-size: 1.25em; font-family: proportional,monospace; font-weight: bold}
defines the header level h1
table.simple .caption {font-size: 1.1em; font-weight: bold; border-bottom: solid 2.0px; text-align: left;}
defines the caption of each table, e.g. for a given value label set. If you preferred centering of the caption you would just in a standard editor change that part to:
table.simple .caption {font-size: 1.1em; font-weight: bold; border-bottom: solid 2.0px; text-align: center;}
For an introduction to formatting with stylesheets read: http://www.w3.org/Style/Examples/011/firstcss
regards
Jens Lauritsen EpiData Association
Thanks, Torsten and Jens:
But I was wondering If would be better to release the first stable version with almost all the capabilities of the actual stable one. I'm talking about the implementation of other important case of use like the relational database function and the validation of double data entry.
Regards,
2011/6/6 epidata-list@lists.umanitoba.ca:
Dear All
A "feature complete" version of EpiData Manager has been uploaded today - the expectation (read as "hope") is, that this version with bug removal can be released as version 1.0
* *NEW: *Create reports in HTML format using "Tools" * Default settings for a project * Change font and colour in program settings
The reports on project structure (question lists, value label lists, combined lists) are produced when you click on tools menu. All epx/epz files in that folder are included. (Notice they must be of most recent format).
For Entry Client a similar version will be released within the next days .
The following aspecs will be fixed/clarified before release: a. A Proper About box will be added. b. For reports a grid of included files will be shown to the user - just like "add structure" to allow for in- and exclusion of single files. c. known and reported bugs. d. any bugs reported from now on by users. e. updating of documentation on the wiki.
Another interesting development has taken place in the last week. We have localised an open-source based cross-platform testing system for interactive simulation of user behavior. We will spend some energy and time to develop validation tests to secure that the visual interface behaves properly after each new compilation. More news on this later.
Therefore please: Now is the time to challenge the software and report problems and viewpoints on the products - before releasing as v1. Documentation and instruction files could be more elaborate obviously - this will come later. In practice it has turned out, that unless features work almost without instruction, users will not use it.
Kind regards Torsten Christiansen Jens Lauritsen
EpiData Association Denmark _______________________________________________ EpiData-list mailing list EpiData-list@lists.umanitoba.ca http://lists.umanitoba.ca/mailman/listinfo/epidata-list
Good point by Omar to discuss "when to release as v1" of the new software.
The following aspects are not developed yet for Manager and Entry Client: - several databases in a given project - key handling within a given database - relationships between databases (relate) - duplicate data entry procedures for entry and validation
But neither are: - multilanguage databases (translated questions and headings, structure fixed) - user access control and encryption - good clinical practice principles of logging
- Adaptation of Analysis to new principles (core module and UTF8) and cross platform is also a challenge.
Many projects can be done with just one database and the level of field control now possible in Manager/Client. With a few more additions to documentation of structure (all definitions and project level information) a large number of projects can be handled within the current functionality.
I see this strategy: a. All current software available from the Http://www.epidata.dk/download.php will remain there b. When considered stable Manager and Client will be available also from that same page. From that point onwards all created data files will be readable by future versions. c. New points will be developed, while fixing bugs in released versions.
A discussion of "when to release as v1" is welcomed.
regards
Jens Lauritsen EpiData Association
I think the release of a simple core version before all planned features are implemented is a good idea, because:
1) It has a fruitful and inspiring precedent in the current EpiData Analysis which allowed with an early release a large user base to test, provide input, and revision for subsequent enhancements 2) Not to release until all possible features are implemented would mean to restrict testing to a very small user base and thus perhaps missing key problems in the core functionality 3) Relatively early release will demonstrate to a wider public the progress being made and allow establishing through user-feedback the core functionality 4) Early release will, in addition to already planned enhancements, enable input coordination for future development
Hans
On 20:59, epidata-list@lists.umanitoba.ca wrote:
Good point by Omar to discuss "when to release as v1" of the new software.
Thanks for the responds, Jens and Hans!
- I do agree that a stable release can be benefit from the bug reports of a wider public. - I do agree, also, that most projects can handle the implementations done in the current test version. - I do agree with the strategy that propose Jens.
But maybe a release version with fewer implementations than the current stable can lead to some misunderstandings with some users. For example, maybe a user that is not informed with this wonderful list can think that "the new product of EpiData is a lower level of the old one".
I Anyway, this can be informed on the download web page, to avoid this probably confusion. And also maintaining the old version to download. Maybe with a table indicating the implementations that the new version have and by other hand with the future challenges that are already implemented with the old version can illustrate this easily. Something like this:
Old version New version One database One datatabase Multi-database Not yet And so on...
Regards,
2011/6/8 epidata-list@lists.umanitoba.ca:
I think the release of a simple core version before all planned features are implemented is a good idea, because:
- It has a fruitful and inspiring precedent in the current EpiData Analysis
which allowed with an early release a large user base to test, provide input, and revision for subsequent enhancements 2) Not to release until all possible features are implemented would mean to restrict testing to a very small user base and thus perhaps missing key problems in the core functionality 3) Relatively early release will demonstrate to a wider public the progress being made and allow establishing through user-feedback the core functionality 4) Early release will, in addition to already planned enhancements, enable input coordination for future development
Hans
On 20:59, epidata-list@lists.umanitoba.ca wrote:
Good point by Omar to discuss "when to release as v1" of the new software.
-- Hans L Rieder, MD, MPH Jetzikofenstr 12 3038 Kirchlindach Switzerland
Tel: +41 31 829 4577 Mob: +41 79 321 9122 Web: http://www.tbrieder.org
EpiData-list mailing list EpiData-list@lists.umanitoba.ca http://lists.umanitoba.ca/mailman/listinfo/epidata-list
participants (1)
-
epidata-list＠lists.umanitoba.ca