Hello,

I'm Kevin Garwood, a software developer at the University of
Manchester. We've just finished making Pedro, a free data modelling
tool that renders XML data entry forms for a model expressed in a
particular style of XMLSchema.

The tool is featured in this month's Nature Biotech Issue: "A
systematic approach to modeling, capturing, and disseminating
proteomics experimental data".
(http://www.nature.com/cgi-taf/dynapa.../n3/index.html)

The application was originally developed for the proteomics community,
but it is actually domain independent. In the download, you'll see a
proteomics data model and a crude model on botany. [For the botanists
who do download the application, please know that I built the data
model very quickly and used some field guides on flowers in Britain.
Your models will doubtlessly make way more sense!)

Imagine for a moment that you want a tool that lets you do rapid data
modelling, so that your group can come to a consensus about what kind
of data to record in your experiment files. With Pedro, once you've
developed your own data model, you immediately have a data entry tool
that lets you produce XML files that conform to that model.

Now suppose you want to put in field-based validation that picks up
simple errors such as leaving required fields blank, typing a number
that lies outside of an acceptable domain, or having a date field that
doesn't conform to a particular format. In addition, you can cause
Pedro to validate a field against a regular expression.

Pedro has some neat features:
*it's free, open-source, and we're going to support it on reserved
bug-fix/software development time blocks
*allows your users to tag text fields with terms taken from controlled
vocabulary services (botany has a lot of classification in it so this
might be especially useful).
*context sensitive help that links to HTML document pages made and
maintained by the model developer
*field-based validation, including the ability to have fields
validated against regular expressions
*genomics and proteomics communities are already expressing interests
in the tool

When you first use Pedro, you'll probably incrementally tune a data
model with the fields and validation you want. Then you'll have a
data entry tool others can use to make XML data files.

We're curious about whether some of you might want to use Pedro to
design data models for a variety of scientific fields such as
immunology, chemistry, physics...

Please let us know what you think!

Kevin Garwood

Work:
http://www.cs.man.ac.uk/~garwood
Personal:
http://uk.geocities.com/zaredbaron