Attendees

• Alan Froggatt

• Andrew Carr

• Anthony Welsh

• Clement Lee

• Colin Eberhardt

• David Withers

• Ryan Stuart

• Simon Laing

• Susan Thomson

Absent

• Matt Richards

• Will Salt

Minutes

Discussion of current position

1. Generator is at version 1

   1. Available as a download from GitHub only

2. Profiler is closed-source, but functions well at present

   1. Clement to organise a demo for 15^th Jan

Discussion of potential features for the suite of products (DataHelix)

1. Produce relational data

   1. For relational databases, e.g. SQL server

   2. As documents for No-SQL databases, e.g. ElasticSearch

2. Usability & distribution improvements

   1. Distribute the generator via more means, e.g. Chocolatey, apt-get, etc.

      1. Research the most appropriate ones

      2. Consider the introduction of a script to save java -jar …

   2. Promote via a (series of) blog posts, that showcase a problem being solved

   3. Apply achievable suggestions from Paul Dykes

   4. Polish the product more to make it more usable out-of-the-box

3. Ability to integrate with API’s

   1. Closely (if not entirely) related to No-SQL relational data

4. Integration with other tools/components/languages (e.g. Python)

5. Meaningful trends and distributions of data

   1. May be required to some degree for relational data

6. Open source the profiler

7. Produce data to train AI models

   1. Suggested that this may not be possible (or appropriate) for the generator

8. Integration with testing frameworks

   1. E.g. Provide test data for Junit/NUnit integration tests

9. Profile UI (and editor)

Vision

1. Can help with promotion to have one big ticket item per release

Votes

Top 3 voted ‘features’ were:

1. Relational data

   1. But in a phased and achievable approach

2. Usability & distribution improvements

3. Open sourcing the profiler

AOB

1. Simon & Andrew to work through blog posts and things to showcase