[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: Discover data patterns or Create data patterns?

• From: Chin Chee-Kai <cheekai@s...>
• To: "Costello, Roger L." <costello@m...>
• Date: Sat, 20 Sep 2008 21:59:21 +0800

It might boil down  to semantics again, but I think the presence of a 
"pattern" necessarily requires more than one item, and possibly many of 
them.
Pure mathematicians would create a "formula" when they find a pattern, 
while statisticians would term a pattern hypothesis "one of high degree 
of confidence" when sufficiently many instances match the pattern 
hypothesis.  One "creates" a pattern through formula, while the other 
"describes" a pattern through measurements.  So I don't think it is 
necessary for pattern discovery and creation to be mutually exclusive - 
they both could co-exist peacefully.  Likewise for descriptive-ness and 
prescriptive-ness.

My take is:
A "master"  data designer  finds a construct that "best" match all the 
requirements, including technical, financial, transitional,  
operational, user and various other imposed constraints.
A "novice" data designer creates a construct, then attempts to find a 
problem that best match the construct which s/he has created.
Others probably lie somewhere in between.

For your last question, on the basis that nobody is born a master data 
designer, it is safe to say that there must exist steps which novice can 
take to gain mastery (ie, assuming your question is an existential 
query).  Now as to what steps, let's say that a novice designer who 
can't perceive the patterns of success of master designers would have a 
smaller chance of success?

I just found a "lack of pattern" - my reply has no relationship with XML....


Regards,
Chin Chee-Kai


Costello, Roger L. wrote:
> Hi Folks,
>
> Recently I read this:
>
> "One of the most important axioms in the discipline of Information
> Architecture states that designers are the ones who uncover patterns
> inherent in data, and expose them in an interface." [1]
>
> I was particularly struck by these words: "uncover patterns inherent in
> data".
>
> I got to thinking,
>
>     - How does one uncover patterns in data?
>
>     - Does data have inherent patterns?
>
>     - What is a data pattern?
>
> I looked at regular expressions, e.g.
>
>     [a-zA-Z]*[0-9]*
>
> This regular expression specifies a "pattern"; namely zero or more
> letters of the alphabet followed by zero or more digits.
>
> This is a low-level pattern, at the character level.
>
> What are patterns at the document level?
>
> When I create an XML instance document I am (implicitly) stating,
>
>    - Here is a data pattern I think is useful.
>
> Do you agree?
>
> For example, I discovered a pattern in BookStores: they are comprised
> of multiple Books. And I discovered a pattern in Books: they are
> comprised of a Title, an Author, a Date of publication, an ISBN, and a
> Publisher. I can expose these patterns like this:
>
> <BookStore>
>      <Book>
>          <Title>The Art & Science of Web Design</Title>
>          <Author>Jeffrey Veen</Author>
>          <Date>2001</Date>
>          <ISBN>0-7897-2379-0</ISBN>
>          <Publisher>New Riders</Publisher>
>      </Book>
>      ...
> </BookStore>
>
> QUESTIONS
>
> 1. Are data patterns discovered, or, are data patterns created?
>
> 2. Are data patterns descriptive, or, are data patterns prescriptive?
>
> 3. What's the difference between a "master" data designer versus a
> "novice" data designer?
>
> 4. Does this characterize the difference: a master data designer
> discovers patterns intrinsic in data whereas a novice data designer
> engineers a pattern?
>
> 5. Are there steps that a person can take to transition from a novice
> data designer to a master data designer?
>
> /Roger
>
> [1] The Art & Science of Web Design by Jeffrey Veen
>
> _______________________________________________________________________
>
> XML-DEV is a publicly archived, unmoderated list hosted by OASIS
> to support XML implementation and development. To minimize
> spam in the archives, you must subscribe before posting.
>
> [Un]Subscribe/change address: http://www.oasis-open.org/mlmanage/
> Or unsubscribe: xml-dev-unsubscribe@l...
> subscribe: xml-dev-subscribe@l...
> List archive: http://lists.xml.org/archives/xml-dev/
> List Guidelines: http://www.oasis-open.org/maillists/guidelines.php
>
>   

• References:
  • Discover data patterns or Create data patterns?
    • From: "Costello, Roger L." <costello@m...>