Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - Richard Freggi

Pages: 1 2 3 [4] 5 6 ... 8
General Board / Re: Aris vs EA for Process Modelling
« on: April 26, 2018, 12:30:09 pm »
Hi Uffe - welcome to use my analogy!

Re. the discussion for tool choice, you can follow roughly the steps you would for an application fit-gap analysis in an enterprise architecture engagement.  It really is essentially the same thing. 

There are good methodologies explaining how to do this and one of them - TOGAF - is freely available (you can see all of it on the Open Group website).  Yes it can be done!

General Board / Re: Aris vs EA for Process Modelling
« on: April 25, 2018, 10:39:59 am »
Hello Ian
I think it would boil down to what methodology your client wants to follow and what they want to get from their process models.
We had similar discussions in my company and honestly sometimes it felt like people arguing whether baseball bats or tennis rackets were the best tool... the answer of course depends on what game are we playing?  (In that specific case it turned out that we wanted to play basketball, which made the 'tool war' argument even more ridiculous). 
So I would focus efforts on getting maximum clarity on what your client wants to achieve and how they want to achieve it - then the tool choice should be straightforward.

Bugs and Issues / Re: EA using Wine and ODBC
« on: April 21, 2018, 12:50:48 pm »
I think EA should be cross-platform, what with Win10 being a kludgy mess and Microsoft reorganization moving away from OS-centric to service centric. 

I am concerned that EA is evolving into a more and more complex tool that offers features I don't need instead of a bug-free, highly reliable, highly usable, portable implementation of UML standard.  I can't justify spending any money on EA14 license.   

Sparx folks, if you are listening, how about focusing on bug elimination, x-platform, usability and user interface enhancements instead of adding more and more features?  Plenty of suggestions about these in the user forum!  Thanks!

Yes Paolo I got the names mixed up.  Gotta stop posting before my 2nd cup of coffee in the morning.  Original post edited with correct name (Mats).

Sorry Mats I don't think you can be successful in this one (I'd be interested to know how it works out for you...)
Some considerations:
1. Data does not 'flow', it is queried.  Data flows are a hang-on from 1970s mainframe philosophy that stuck around because people liked the idea although it only has very limited usefulness (a bit like flowcharts).  I don's know if an architecture based on information flows can be efficient/effective/flexible (I'd be interested to learn more about this)
2. Since data is queried / provided by each table/class, each query typically contains a mix of different attributes from different classes / tables / whathaveyou + the query logic.  These are called messages (hello sequence diagram!!!)
3. Can we reconstruct a data model from a sequence diagram?  Yes with some effort and some modeler judgement / experience / assumptions, as long as the messages are between participants (I think a Data Flow Diagram maps poorly on to an interaction/collaboration diagram)

TL;DR:  There be dragons where you are going, methinks.  DAMA website has some good resources about data architecture.

I don't know if it CAN be automated but I would STRONGLY recommend against it.  Creating a physical data model from a logical data model or class diagram is not straightforward, and there are so many factors such as level of denormalization, target RDBMS implementation specifics, data modeler style etc.  A perfectly correct physical data model can be totally different from its perfectly correct logical data model.  Anyone else has a better idea?

The behaviour you described is correct.  The DDL transform works at physical level, meaning it assumes all elements to transform are <<Table>> stereotype classes containing physical attributes.  I think the solution is to generate a physical level class diagram (therefore you must decide what to with your parent class) then run the DDL transform.


I am curious what would total participation look like in Crow's Feet (I.E.) notation?  Or what would a SQL DDL code for total participation look like?

If it's 1:1 as suggested in previous post, it's certainly possible to do with EA but 1:1 is usually a sign of a problem in the data model... probably semantic mistakes or treating an attribute as if it was an entity, or using 2 entities where 1 would be enough.  I'm curious to know... thanks!

General Board / Re: Document a whole Solution
« on: March 02, 2018, 11:39:25 am »
Agreed!  TOGAF content metamodel really good, Zachman framework is similar and probably a little easier to use.

Geert, if you use UML for everything you will have a fully integrated model, with consistent elements and artifacts from contextual down to physical and out of context... easier, more efficient and effective than mixing multiple notations.  IMHO UML works even better at contextual-conceptual-logical than at physical level (UML was originally developed to talk to people, and a failed attempt to shoehorn it into a programming language has damaged it quite a lot...)

Willeygi, Sparx has plug-ins for TOGAF and Zachman; you can find full TOGAF documentation online at the Open Group website and there's plenty of Zachman info and examples in the web. 

As for the model organisation, I prefer abstraction layer before domain

The TOGAF content meta-model is by far the best part of TOGAF and is very useful for structuring "content".

Uml Process / Re: Modeling complex component associations and interfaces
« on: February 13, 2018, 02:26:05 am »
Hello Miktaylor,

a good UML diagram shown only the minimum required to convey clearly and unambiguously the concept it wants to express.  For example, what is a functional architecture, who needs it and what for?

You need to ask yourself exactly what is the purpose of your diagram, who is the audience and what they will get out of your diagram.  This will help you choose the right diagram for your idea.

Experienced modelers break down large complex diagrams into several simple ones, each with a specific and clear message.  A very short book "The elements of UML 2.0 stlye" is a great reference.

Suggestions and Requests / Re: Relationship between/to Attributes
« on: January 30, 2018, 05:43:14 am »
Thank you Paolo and qwerty for the info!

Automation Interface, Add-Ins and Tools / Re: Project Template in MDG
« on: January 28, 2018, 07:25:29 pm »
If the template refers to the structure of the model, with maybe some strawman diagrams, maybe a UML profile would be a good way to do it?  Then the profile could be included in the MDG.

Suggestions and Requests / Re: Relationship between/to Attributes
« on: January 26, 2018, 07:44:21 am »
Thank you for your insightful post Simon.
For my own education: about this
"UML doesn't define a notation for rendering relationships between the attributes of a class. (Although, it's not incorrect to render it as a Port or Part to show the relationships."

Understood re. diagramming.  In practical terms, my assumption would be that if some element has a real association with something else, then both elements are classes (i.e. the attributes having the relations are in fact classes on their own right).   Would there be a realistic practical situation where such a relationship exists between non-class elements?  In some programming language or some practical application. 

This isn't an EA issue, but the way UML works: Attributes Do Not Have Relationships.
Not quite true. Dependency, some of its specializations and Information Flow all go between Named Elements, which includes Attributes.

On the other hand, UML doesn't define a notation for rendering relationships between the attributes of a class. (Although, it's not incorrect to render it as a Port or Part to show the relationships.)

General Board / Re: Importing ERwin models
« on: January 20, 2018, 06:40:37 pm »
I am maybe not the most qualified person to post because my imports from ERWin (circa 2015... I forgot which ERWin version but it was the current one) did not go well.  I followed the instructions in ERWin to export as Sparx EA compatible XMI; all models imported but there were missing relationships so I had to manually correct the imported model, very time consuming.  Tried several times with different formats, still not good.

For me the safest way is to forward engineer the source model, create a database with it (in mySQL, SQLite, Oracle express etc), then connect Sparx to that database and reverse engineer it.

Good luck!

General Board / Re: XMI Round Trip - Data Loss?  Your help needed!
« on: January 18, 2018, 09:39:39 pm »
qwerty and vkn, many thanks and this helps.

"Create placeholders for missing External References during XMI 1.1/2.1 Import" locks the placeholders (I have 13.10 professional desktop) and I can't unlock them to change colors or names (I can change fonts though).

Anyways that's already a big step forward, many thanks for your info!

Pages: 1 2 3 [4] 5 6 ... 8