Candidate experiments

Update 10th April 2015

I have now spent quite a bit of time exploring scripting capabilities and found it really interesting.  I only wish I had been fully aware of their capabilities some time ago as scripting really can help with the small daily customising tasks.  However, although much easier and quicker than writing AddIns they can be frustrating to debug!

Having done that, it was time to review where I should focus my energies next. I started by reviewing the hot topics (and posted my observations) and look at a set of my mindmaps in which I dump my thoughts.  From this I have extracted a subset of topics, following on from my last candidate list – this is the diagram below which you can click to get a larger version.

I have colour coded as follows:

  • Dark blue – groups with uncompleted items
  • Green – topics covered and in most cases a post created
  • Light blue – potential topics
  • Purple – group of related potential topics
  • Red – Hot topics – questions

The challenge now is to:

  • Check if there are any additional topics that should be included – hence my post and interest in getting feedback; I probably need to do a pass through the various forums (Sparx and Stackoverflow) to get some inspiration
  • Assess the aim of the experiment
  • Prioritise
  • Action

And with that done I’ll carry on and do some more posts to report on progress.

Candidates Updated Apr 2015

Candidates Updated Apr 2015


Update – 4th November 2014

Just having a look back at my somewhat random series of experiments over the last few months and thinking about the areas of EA that I don’t understand fully. In fact there more I find out and think about what EA offers the more I see there is so much more to check out.  Perhaps that’s inevitable with any broad ranging product and, of course, Sparx are adding more all the time.

The mindmap below gives some idea of the topics of interest to me, in particular the areas of application development and customising EA, in particular those aspects that enable me to use EA as my central tools and interacting with others as and when needed.  This mindmap is not complete – otherwise it would be unreadable!!

Some of the areas to explore

Some of the areas to explore


Update – 11th December 2013

Since my last visit here I’ve been playing with:

  • EA workflow/scripting
  • EA – Java Automation

So starting to go into a few new areas.

More to do on the above in the next few weeks and will post as information arises.

Brief update – 23rd November 2013

I occasionally look at this page and realise that what I wrote months ago hasn’t changed much, but I’ve being doing bits and pieces when other work permits.

I’ve worked with Full End 2 End development of simple applications – and so far not convinced it’s scaleable.  I feel that this shouldn’t be the case and feel I should persevere.  I think the problem I have faced is working in 2 tools VS and EA, and concerns with sync between the two. I can see that if I write, compile and build from within EA then it should be fine, however EA doesn’t have the richness of development aids provided by VS.  I suppose if I were working with systems which did not provide a windows GUI (e.g. some embedded systems) it would be fine.  However, … I probably need to explore this further.

In the last few moths I done some of the interesting stuff I’ve been doing recently has included using:

  • MDGs – both in terms of providing customised elements and toolbox
  • I’ve used the tagged values associated with the MDG elements coupled to AddIn’s to provide some nice UI for managing element attributes
  • I’ve done some AddIn work both within EA and MS Project to provide developers and project managers, respectively, with RW access into the EA repositories/Project files.

However, what I haven’t done is write too much about them!

  • Full End 2 End development of code – I have used EA to do part of my developments but want to check out how to work from a concept to producing and testing code and produce some real examples; guess a lot to learn here – so could be a long time doing this, possibly the never ending task! As above – found some limitations which for my daily work make it a show stopper – but can see that there are probably areas where it really would be useful.
  • Integration with tools – I suspect that as part of doing the End 2 End development I will need to look at the use of other tools.  These may be through Add-Ins  or in exploring any other means that helps in bringing stuff together; accessing EA from other tools; synchronising data between tools, etc e.g. with Project management tools, Mind mapping software, image handling I’ve written several test AddIns to see what can be done working with some tools – the sync problem remains the constant issue.  With defined rules there can be some usable workarounds (seems to work fine with MS-Project) but for other Office tools it is very much a one-way activity either import or output (unfortunately)
  • Testing some features of EA Extensions (I write EA extensions in my day job but there are some things I want to try out that I just don’t get time to do) – some recent work has brought into using MDGs.  So far just used to create elements and toolboxes – but in creating the elements with the MDG I know tagged values are available to rely on.  Next is to explore MDG’s further – and really find out what all the MDG  AddIn events enable me to do
  • Managing EA based projects which will look at: Have documented some of this but feel that I could write masses on the model management stuff – not least as its a constant battle
    • Model administration
      • Databases e.g. Using mySQL
    • User Management
      • Users
    • Model control
    • Modelling standards
    • Model management
      • Shared
      • Master Replica
      • Backups
    • Version control
    • Data transfer
      • Import
        • Harvesting content
      • Export
    • Requirements management – the reason I got into using EA in the first place; the idea that I could put some structure and order into product development.  In my mind requirements management cannot stay as a list in a spreadsheet it needs to be managed within the context of the product that is being developed and to which it can be linked – it is part of the design process. 
    • Code testing
  • Plus many more I can think of …