Sunday, January 30, 2022

TYFK: Nobody Understands What a Data Model Is

 “A data model is a collection of concepts ... used to describe the structure of a database...data types, relationships and constraints...is basically a conceptualization between attributes and entities ...
The building blocks in the data model are as follows:
  • Entity − An entity represents a particular type of object in the real world.
  • Entity set − Sets of entities of the same type which share the same properties are called entity Sets.
  • Attribute − An attribute is a characteristic of an entity.
  • Constraints − A constraint is a restriction placed on the data. It is helpful to ensure data integrity.
  • Relationship − A relationship describes an association among entities.
--TutorialsPoint.com

Fallacies, Misconceptions and Confusion

  • A data model:

- does not describe (just) the structure of a database.
- is not "a conceptualization between attributes and entities" (whatever that means).

  • Entities, entity sets and relationships are not building blocks of a data model.

Friday, January 21, 2022

Read My Lips: If There's NULLs, It's Not Relational

“Let's say I want to store a list of movies that are stored on iTunes. For simplicity, we'll just store a few fields so that the film Avatar has these values:
ID: 354112018
Name: Avatar
Year: 2009
Synopsis: "From Academy Award®-winning director James Cameron comes Avatar, the story..."
However, sometimes the Synopsis is missing...and sometimes the Year is missing. Without giving it a second thought, I would probably create one table to store those four fields, something like this:
ID (INT)
Name (VARCHAR)
Year (INT NULL)
Synopsis (VARCHAR NULL)
Is there any advantage in 'further normalizing' the database so that, for example, I don't store any null values, such as:
Title
 TitleID
 Name

TitleSynopsis
 TitleID
 Synopsis

TitleYear
 TitleID
 Year
To me it seems like doing this would potentially create hundreds of extra tables (on a large database) and make inserts a nightmare -- I suppose a View could be created to flatten out the results so it's queryable, but even though I feel like it would require so much overhead. So is there any reason in the above case to normalize to remove nulls, or in general, what would be the case to do so, if there ever is one?”  --StackOverflow.com

Fallacies

That we see this in 2022 is testament to abysmal ignorance of fundamentals in the industry. Let's enumerate the fallacies:

Sunday, January 16, 2022

OBG: No Understanding without Foundation Knowledge Part 3 -- Debunking an Online Exchange 2

Note: To demonstrate the correctness and stability offered by a sound theoretical foundation (relative to the industry's fad-driven "cookbook" practices), I am re-publishing as "Oldies But Goodies" material from the old (2000-06) DBDebunk.com, so that you can judge for yourself how well my arguments hold up and whether the industry has progressed beyond the misconceptions those arguments were intended to dispel. I may revise, break into parts, and/or add comments and/or references, which I enclose in square brackets).

A 2001 review of my third book triggered an exchange on SlashDot. This six-part series comprises my debunking at the time of both the review and the exchange in the chronological (slightly out of the)  order of the original publication.
Part 1: Clarifications on a Review of My Book Part 1 @DBDebunk.com
Part 2: Slashing a SlashDot Exchange Part 1 @DBAzine.com
Part 3: Slashing a SlashDot Exchange Part 2 @DBAzine.com
Part 4: Slashing a SlashDot Exchange Part 3 @DBAzine.com
Part 5: Slashing a SlashDot Exchange Part 4 @DBAzine.com
Part 6: Clarifications on a Review of My Book Part 2 @DBDebunk.com

Slashing a SlashDot Exchange Part 3

(first published in 2001 @DBazine.com)

The following comments being debunked are by the W3C XML Query Working Group's Activity Lead and by an academic. [The exchange took place when XML DBMS was one of the hottest fads as late as 2013.  Consider them in this context: where are XML DBMSs today?]

“The article seems to say ‘I don’t like SQL and I don’t like XML and I think XML Query is about merging them although I don’t understand it very well, so the people working on XML Query must be stupid, and in any case it’s easier to attack people than understand a specification.’ Perhaps that’s unfair, but it’s clear to me that the writer is a little fuzzy on the design goals of XML and also on the focus of SQL development over the past 10 or 15 years. In both cases the story is about interoperability.”

Saturday, January 8, 2022

OBG: No Understanding Without Foundation Knowledge Part 2 -- Debunking an Online Exchange 1

Note: To demonstrate the soundness and stability conferred by a sound theoretical foundation (relative to the industry's fad-driven "cookbook" practices), I am re-publishing as "Oldies But Goodies" material from the old (2000-06) DBDebunk.com, so that you can judge for yourself how well my arguments hold up and whether the industry has progressed beyond the misconceptions those arguments were intended to dispel. In re-publishing I may revise, break into or merge parts and/or add comments and/or references that I enclose in square brackets). 

A 2001 review of my third book triggered an exchange on SlashDot. This six-part series comprises my debunking at the time of both the review and the exchange in the chronological (slightly out of the)  order of the original publication.
Part 1: Clarifications on a Review of My Book Part 1 @DBDebunk.com
Part 2: Slashing a SlashDot Exchange Part 1 @DBAzine.com
Part 3: Slashing a SlashDot Exchange Part 2 @DBAzine.com
Part 4: Slashing a SlashDot Exchange Part 3 @DBAzine.com
Part 5: Slashing a SlashDot Exchange Part 4 @DBAzine.com
Part 6: Clarifications on a Review of My Book Part 2 @DBDebunk.com

Saturday, January 1, 2022

Schema and Performance: Never the Twain Shall Meet

One of the core objectives of this site (and my work) has been to demonstrate that there will not be progress in data management as long as the industry and trade media require and promote exclusively (mainly tool) experience in the absence of foundation knowledge. I have published and analyzed ample evidence that relational language and terminology are used without grasping what it actually means -- a good way to gauge lack of foundation knowledge.

Recently I posted a four part series titled "Nobody Understands the Relational Model" showing that even a practitioner steeped in the RDM does not really understand it. Consider now a practitioner's mistake at the beginning of career -- "a bad database schema and what it did to system performance" -- which, he claims, belatedly taught him a lesson. Hhhhmmm, did it, really?

View My Stats