USE CASE Cookbook

[Steve_Straley]

Tim,

I have it and it is _NOT_ thorough enough.  I'm thinking far more exhaustive.   The situations I found myself in at GE with the last project were only .10 covered in the above mentioned book.   I can't cook a meal if the only recipe I've been given is for the soup course.

Steve

[jaimeglz]

Hi Tim,

Some pious cyber-spirit transformed Alistair C-o-c-k-b-u-r-n's (pronounced Co-burn) name in your posting. The same funny change happened when in this same frorum I recommended a link to one of Alistair's writings.

Hi AltatemTC,

There's a Sept. 19 thread with a quick answer to your question, "What do I do when?". See:

http://www.sparxsystems.com.au/cgi-bin/yabb/YaBB.cgi?board=general&action=display&num=1032446412

Be aware that the answer I gave then was very general, and the purpose of this forum is precisely to collectively work out questions on details of the software development process.

For a "light-weight" and very comprehensible view on the software development process, see Martin Fowler's "UML Distilled" (Addison Wesely). And probably the most recognized work is Jacoboson, Buch and Rumbaugh's 450 page "The Unified Software Development Process" (also published by Addison Wesley).

I'm sending a Powerpoint slide to your EMail on the general path I use. (For lack of time, I have not finished setting up a proper web page for this material.)

Everybody,

Have a merry, merry Xmas!

Jaime

[AltatemTC]

Hey jamieglz,

Thank you very much.  I appreciate the link and the slide.  I will study both.  I have read "UML Distilled" and found it very good.  I have written myself a cheat sheet on what each element is for.  I guess my biggest problem is my ignorance concerning a "formal" analysis process.

I'm one of those guys who has been doing everything by the seat of his pants for so long that it is difficult to convert over.  I keep opening up my program editor so I can code instead of determining exactly what I need to code.  Most of my projects have been simple and straight forward, such as AP, GL, AR, etc.  I am starting to work on some pretty complex systems and determined I really need to create a "formal" analysis process.

I have been studying quite a bit over the last year in order to make sure I formulate a reasonably good process.  I have read Steve McConnell's "Project Survival" which I found very informative.  It helped me crystalize some of the thoughts I had concerning my haphazard development process.  I have also read many other books and found several good sights on the web.  What I haven't seen is, someone layout their process.  Sometimes I think they fear the backlash because someone doesn't agree or it won't work for them and so on.  I'd really like to compare notes with someone.  I realize that there may be things that they do that I don't need to and things that I do that they don't need to.  As I said before, my ignorance is showing, and I am just trying to get my arms/mind wrapped around this.

The thing that I have been trying to do since I found Enterprise Architect is to determine how to best utilize EA, a formal process and UML in order to create an efficient project management process.

Thank you again for your help.

[AltatemTC]

Hey Steve,

I agree with you concerning the fact that most books aren't nearly exhaustive enough.  To me the "Cookbook" is kind of like component technology.  Write it once and use it again and again.  Same way with the "Cookbook" if folks are willing to share their analysis, it would save some others from having to completely reinvent the wheel when they attempt a similar type project.

So what do we do to get started?  Where do we put the repository?

[Steve_Straley]

TC,

Well, I kindof thought the same thing... gathering a TON of "examples" but again, the time to do this and organize it and everything else that goes into a book is extensive.  Not sure if people are willing to give up their UC's to their software AND to write a BRIEF outline to the scope and intent of the UC as it fits in with the overall process.     Also, not sure if people will agree with everyone else's UC's... the art form of our industry lends itself to various interpretations.

Steve

[AltatemTC]

Steve,

The price of admission to the sharing could be the submission of a UC with the outline.

As far as folks disagreements, I think that is understood.  Everyone who has been in this industry as long as you and I (almost 20 years for me) knows that there are _AT LEAST_ 10 ways of doing things.  I am concerned that folks will start a war over someone else is doing it "wrong" but I think it is worth a shot.

Organization will have to be done by a group.  If we could get maybe 6 folks to donate 1 hour per week this could work.  They would have to verify that a UC is viable and not someone submitting an empty EAP, categorize it and then issue admittance to the group.

Is there some place to put this or does someone need to pony up some server space?

[Steve_Straley]

TC,

I am very concerned about "I am concerned that folks will start a war over someone else is doing it "wrong" but I think it is worth a shot."  I totally agree.   And I've experienced this myself and didn't like it one bit.   But, like you, the potential or the "marginal benefit" versus the "marginal cost" make it tempting!

I have some space on my server and will have more once we move our web site over to our .COM site.   Once this is finished, the .NET and .ORG sites could be used to help facilitate this.   I'll see if I can hook up some Cold Fusion stuff and to use a database to store these things and then dynamically display them upon request.   Just an idea if people are into this.

Steve

[AltatemTC]

I'm ready.  If there is something I can do to help, let me know.  If you need some additional space, I will chip in a GB.

[Quozzbat]

For TC:

I'm currently using Developing Applications with Java and UML by Paul R. Reed, Jr. I believe this might answer some of your questions. I have been following it pretty closely and, so far, my client has been impressed with the documentation I have produced.

Andrew Knight

[doug rosenberg]

Hello

I've written two books on the subject:

Use Case Driven Object Modeling with UML -- A Practical Approach

and

Applying Use Case Driven Object Modeling with UML - an annotated e-Commerce example

The second book (which we call the "workbook") takes an interenet bookstore example, and has examples with mistakes left in highlighted in red, flip the page and see the corrected exercises.

The books have some "cookbook" aspects to them, in that they have "Top 10 Errors" lists for things like writing style, sequence diagrams, etc.  

Info at http://www.iconixsw.com/UMLbook.html and
http://www.iconixsw.com/UMLworkbook.html

I just taught a class with EA for the first time last week and was quite pleasantly surprised at how well it behaved.

regards,

Doug Rosenberg
ICONIX

[Steve_Straley]

Doug,

Hi and welcome to the fray.  I am familiar with those books.  As a published author myself (and the guy who started this thread) I know the problems between the publisher's needs standing between the author's desires and the public's wishes.

A single application's Use Case set is the problem IMO.  I mean, we can get language reference guides and books and even pocket reference guides (C# and O'Riley for example).  It just would be nice to have a TON of various types of application (from legacy to new and a mix between) UC's in one guide.   That way whenever someone had a scenario they had to UC document, they could get some idea of how the diagram might be established.

Just an idea...

Steve

[MichaelRuschena]

Steve,

I just wanted to throw in another idea... I've recently been thinking about a similar issue. I write Use Cases for apps, and so do half a dozen people sitting around me. We all work in pretty much the same domain, and deliver products not hugely dissimilar.

Given this, you'd expect that we could share Use Cases. (I take it this is what is meant by Use Case cookbook.) Well, it never worked out that way. Sharing tended to lead to cut-and-paste, with minor modificiations, which tended to get lost. (Of course, it was also a rather amature effort never strongly embraced. Maybe a serious author could put together something worthwhile.  )

I've been thinking that the problem is that a Use Case represents a way of resolving a bunch of issues. To reuse a use case, all of these issues have to be the same. In practice, this doesn't really happen.

This lead me instead to the idea of patterns (mentioned briefly above). Instead of trying to present completed solutions which may or may not fit, I'm working at the moment to sort out the particular patterns. The idea is that these patterns can be used to build use cases.

I've been thinking about this for only a few weeks when I have time, so work hasn't progressed very far. I'll see how it comes out.



(Doug, loved Applying Use Case Driven Object Modeling with UML... I'm evangalising Robustness diagrams to anyone who'll sit still long enough    )

Michael

[Andrew_Webster]

I'm just getting used to working with the UML (got the idea ages ago, but it's purchasing EA that really got me going).
I can see that a Cookbook would be an excellent idea.  

I'm thinking of starting something similar for data structures, since most of my work is in database design. Design Patterns a la Gang of Four are great, but almost too re-usable and abstract. For my project, I'm thinking of 'Organisation has several Addresses of various Types' for instance, and publishing a table structure that contains just the basics for that situation. There are so many similar structures in databases is seems mad to keep re-building the same damn thing over and over.

I'd fall over myself to grab a UML cookbook that did the same thing. Having come into this game via experience and an MCSD, rather than through an IT degree, and now working as a lone contractor, I find lot of the analysis and design work to be a tremendous challenge, although I'm keen to avoid the mistake of trying to be too agile and just jumping in.  I make most of my living trying to de-kludge databases that were built too fast (why to they let users have MS Access? To make me a living!) Anything that would let me browse for a UML 'situation' that's similar to one I've come across would save an awful lot of wheel re-invention.

I'm more of a database guy than a website guy, otherwise I'd volunteer to try and start a site that could collate people's working UML/data model 'recipes'.  I'd gladly work with someone who wanted to get such a thing going.

Steve, this is a great idea, I'm right behind you!

Andrew Webster

[Andrew_Webster]

Hah! Just found the second page of postings for this thread...duh!

I'm up for the hour a week idea, I can always spare an hour when I've got a feeling that somewhere down the track it'll save me two hours.

I don't think it would be a huge problem if people were supplying different ways to solve the same problem. I applaud the effort to try and distill patterns from similar Use Cases, but I could see that a Cookbook could perhaps benefit from people suypplying their own flavours!  If the goal is to provide a resource that analyst programmers could browse through to find something to simplify their lives, then seeing several takes on the same issue could be helpful.  If the Cookbook was a website, then users could vote for recipes, and this would make them self-sorting.

All the best

Andrew Webster

[PhilR]

I'm thinking of starting something similar for data structures, since most of my work is in database design.

Andrew,

Take a look at Data Model Patterns: Conventions of Thought
by David C. Hay.
http://www.amazon.com/exec/obidos/tg/detail/-/0932633293/qid=1044935170/sr=8-1/r
ef=sr_8_1/104-2446082-3500760?v=glance&s=books&n=507846


PhilR