Re: An Analysis Pattern for Inventories

[Dirk Riehle]

Open source and the patterns community are pretty much on the same line. 
The patterns community is still by far the biggest publisher of patterns. 
To qualify as a pattern description, a write-up has to follow the "rule of 
three", i.e. it must show three known occurrences of what it claims is a 
pattern. Typically you do this by referencing articles, published system 
documentation, etc. Unless there is something wrong with this prior art, a 
pattern basically claims it is only documenting something that has been a 
long-standing well-known problem solution. Which probably means it is in 
the public domain.

There have been attempts to patent patterns, but non of them was really 
successful, and we are still kind of expecting a shootout between a 
corporation attempting this and the patterns community (as represented by 
the Hillside group).

When it comes to system design, in the area I know best (object-oriented 
design), things have pretty much been said and patterns are there to use 
without problems. Algorithms are a different beast.

But the question is what you mean by using patterns. For business models? I 
guess a good textbook on double-entry bookkeeping gives you all the 
patterns (in textbook disguise) you need to design a decent bookkeeping 
solution. What's been called "patterns" most of the time just means the 
description form and not the actual contents, so the fight is on about the 
contents, of course, and not the presentation form.

I just see your other email. The copyright on the first page of Eduardo's 
paper prohibits that you republish the paper or make use of the text in any 
form. (Copyright, not patent!) It does not prohibit you to use the 
described design, the actual contents, which is what most readers are 
interested in. Take a look at the end of the paper. The Known Uses section 
serves to address the Rule of Three, stating that the concept has been 
known and used before by at least three independent parties. Ideally, if an 
author wouldn't state this, the paper would not be accepted for the 
patterns conferences.

On business domain patterns, again, Martin Fowler's book on Finance and 
Healthcare patterns (the book is called Analysis Patterns) is a good one. 
Also has lots of organizational modeling, even though much of it may seem 
obvious to the seasoned designer. David Hay's Data Model Patterns are very 
good too. I'd bet they would be willing to confirm by email that you are 
free to use the patterns in whatever form  (again, the concepts, not the 
writeup).

Dirk


----
Furniture for sale: www.riehle.org/for-sale/furniture/index.html
Electronics for sale: www.riehle.org/for-sale/electronics.html
Kitchen stuff for sale: www.riehle.org/for-sale/kitchen.html
Computer (science) books for sale: www.riehle.org/for-sale/books.html
----
For work etc. see: www.riehle.org