Concurrent Baselines vs Versions

[Lee SoCal]

I am a new user of EA, so I don't know if what I am looking for is basic, impossible, or somewhere in between.

I am wondering if there is some way (I am guessing by using versioning and/or baselines) to establish for example three versions of a project, and then be able to go back into version 1, make changes, and have those changes propagate to versions 2 and 3, so I can in effect go back and forth between all 3 versions and add information in any arbitrary order, and then somehow "sync" them up so  the new information added to an older version appears in all the newer versions based upon it.  

I realize this probably violates the basic premises of version control, but I would like to do this so I can back-document earlier versions of a project which were implemented before we started using EA.  This way, I don't have to spend time first completely defining version 1, versioning it, completely defining version 2, versioning it, etc just to establish a coherent model that allows one to see the deltas between 1 and 2, 2 and 3, etc.

For example, say I have all 3 versions partially defined.  I am currently working on the model for version 3 when I realize that I need to add feature X that was present in version 1.  So, somehow, I go back into version 1, add the feature, say the magic words, and then the next time I re-open version 3, there is feature X, having been inheritied from version 2 which inherited it from version 1.

... and yea I know, I'm crazy  but I really would appreciate it if someone could explain to me what the easiest way would be to accomplish this in EA.  We are currently using Corporate edition but without integration with a version control tool.  I'm willing to integrate version control if this would be the only way to do it, but I was wondering if there was a way to somehow make this possible between different baselines.

Thanks for reading.  I look forward to any helpful input.

[Oliver F.]

You are talking about "branching"- start on a HEAD branch, then release it, continue working, notice about mistakes in an old branch, fix it, then merge it back into HEAD- this is called branching.

Despite that several experts say that branching kills productivity (and I have to agree) there is no way around this unless you stop tagging and releasing things to deliver only the latest product version even to legacy customers (in fact I tend to believe this is cheaper than coping with all branching issues but this is a different story)  8-)

Though I have not tried this yet th ebasic principles using baselines would be:

Create a baseline A, then after some time B, then C, etc. and continue working on the HEAD branch after each creation.
Some times later you require a change on B.
Export the baseline B from the model and import it as a baseline into a fresh project (or a copy of the HEAD project). Apply your changes, store it as a baseline B1 and export this baseline.
Open the HEAD model again and import the baseline B1, compare the changes to the model and apply the changes to the HEAD model using the corresponding EA functionality.
Importing it into C requires exporting C first, importing it into a fresh model, importing B1, applying the changes to C, save as baseline C1,  and reimport C1 into HEAD. Something like this.

This is the way it should work- this is not the way I have tried it out until the end but it should give you an idea.

HTH

Oliver