IDEA: Stereotyped Template Diagrams

[Paolo F Cantoni]

Hi,

We've started to use the Template Package option to create a consistency in diagramming for our users.  (See: Settings|Template Package...)

It wasn't obvious until we looked at this option that there needs to be a new diagram type (say) Data Model to handle database related diagrams.  There's a new UML Data Modelling specification coming, and it's probable that they define new diagram types, however, in the absence of any specific direction from that source, adding Data Model under the Extended list of diagrams would be very useful.

However, what would be even more useful is the addition of the diagram stereotype to the diagram type.  (Actually, that would also solve the Data Model template problem

If, when we created a new diagram, the dialog asked for an optional stereotype¹, the template package could contain multiple diagrams of the same type, differentiated by stereotype.  The dropdown lists could indicate whether the type/stereotype had an associated template.

The stereotyping of diagrams is an increasingly useful mechanism.  This type of support would really make it useful.

Thoughts? Votes?
Paolo

[size=0]©2007 Paolo Cantoni & Darren Sampson, Ripple Systems[/size]

¹ Which the creation of any object should anyway...

[«Midnight»]

Yes, I agree.

While EA does seem to recognize diagram stereotypes in some cases, actually manipulating these through the UI, or even with an add-in, can be a real pain. At best, results are not great as things now stand.

David

[Paolo F Cantoni]

We've received a reply from Sparx indicating they are considering various solutions, but no commitment on implementation.  Although the solution is not likely to use templated diagrams.

Neil has used the terms Re-use of extended diagram definitions to cover this concept.  So perhaps on-going discussions could run along those (more general) lines...

Paolo

[darren.sampson]

While extended diagram definitions may be helpful for some situations (such as the Database diagram example), it seems to me that more flexibility in the existing template mechanism may be sufficient for many uses.

I mentioned one such use in my other post ([size=13]Idea: Show Diagram Stereotype in Project Browser[/size])

As we define classes of diagrams, it would be extremely useful to have a template for each class of (say) component diagrams used in the Architecure.

From an outside point of view, it would seem that adding a stereotype combo to the New Diagram dialog would be a simple matter (given stereotype is a property of all diagrams).  From that point, it should be simple to select the particular template to use when creating a new diagram by matching the stereotype.

If this is as small an effort as it seems from an outsider's point of view, it would be really handy while we wait for a more sophisticated solution.

Comments?

ps. I've noticed that diagram notes and text elements don't seem to be copied from the template. Getting a consistent base diagram including things like Diagram Notes would seem to be a typical use for a template.  The properties seem to come over OK though.

[alicecbrown]

Where can I find a definition of each stereotype given with the EA software?

[Paolo F Cantoni]

Where can I find a definition of each stereotype given with the EA software?

Alice,

See Settings|UML>Stereotypes...  However, I think there are still some hard-wired stereotypes that aren't distributed in the list.

HTH,
Paolo

[vnedelcu]

Hi Paolo,

This would be a good feature that I guess will help a lot the users.
So far I am using different apps to deal with Data Models (and diagramming) - I can use stereotypes, I can swith from ERD to IDEFIX whenever I want (actually I like these better than UML for DBs), I can easily define "abstract" tables with properties that are then inherited by the childern etc.
Having a better Stereotyped Template Diagram would perhaps convince me to use just one tool (EA) instead of using several.

Well done.
Vasile

[alicecbrown]

I looked at the 'definitions', but it's AGAINST THE LAW to define concepts in their own terms.    Not really, but certainly a wee bit disingenuous or ignorant, as the case might be.   :-/
As unwise as the use of examples that use software or business terms in their example while trying to explain a concept that IS software or business development, so the cyclical use of the same terms in definitions.  And I'm sure as few of you understood that last sentence as those of us out here who try to understand concepts without examples or poor ones.

[sargasso]

I think    I recall that some if not all the "old" standard stereotypes were, to some extent, sort-of, described in the old 1.4 or 1.5 specs.

They are, or were the last time I looked, available in the OMG  vault.

As for the new standard stereotypes, I guess you could search the OMG development areas for some commentary, but it would probably be unwise to accept that any such definition found was the truth.  Just accept them as ideas.

(As for cyclic definitions, I think all steretypes should be defined as recursive TLA's - just to keep the buzzards guessing   )

bruce

p.s.  Sorry Paolo, I didn't think or mean to keep the highjack going.

[thomaskilian]

Alice, please define definition first. Please.

[alicecbrown]

In German or English?  Long ago, when AI was the rage, we learned that German was the easiest language for a computer to translate, being so rules-ridden.  And Guess what? our chosen requirements language, English, is the worst.  Why?
The problem with definitions in English is that words are so context driven:  Consider the following mean everything, mean nothing words (that you should put in your Requirements Tutorial):
entity
data
phase
mode
thread
cycle
etc, etc, ad nauseum :-/
Hey that's it: let's just go back to Latin and the Good Old Days when flowcharts were all we needed.  

[mikewhit]

let's just go back to Latin and the Good Old Days when flowcharts were all we needed.  

...or Ancient Geek ...

[«Midnight»]

But "nauseum" isn't an English word, and certainly means something.

Also I was there in the "Good Old Days" and I read classics in school back then. I don't recall needing flowcharts for Latin at all.

David

[Paolo F Cantoni]

But "nauseum" isn't an English word, and certainly means something.

Also I was there in the "Good Old Days" and I read classics in school back then. I don't recall needing flowcharts for Latin at all.

David

But then David neither is "et cetera".  And besides, what's an English word anyway?

See the excellent book: [size=13]The Fight for English: How Language Pundits Ate, Shot, and Left [/size] by David Crystal.

As a person of Italian extraction, born in England, living in Australia for the last 40 years;  having worked for a number of American companies and now working for a wholly owned subsidiary of a Singaporean conglomerate  (see [size=13]Singlish[/size])  ; I've learnt the benefits of trying to be as clear as you can be when you communicate  .

Paolo

[sargasso]

what's an English word anyway

AFAIK "English" isn't.

'smatta anyway, wots ambiguous about engrish?

The next sentence is false.

The previous sentence is true.

b