The Birth of JOIN: Chapter 3 (Final)

War & Victory

Various Academic Institutions…

“Wow!”
“What Chris?”
“There is a guy in California who has invented a multipurpose n-dimensional data model.”
“Really!  Hashtables are hard enough to work with and they only have 2 dimensions.
“He has a way to manipulate them using set theory and enforce rules based on Boolean and predicate logic. 
“A building full of mainframes couldn’t handle that!”
“Don’t worry about it. Moore’s law will hold.”

Somewhere in Corporate America…

“We already have a data bank system!”
“But there appears to be a better and more correct way to do it”
“All the engineers are saying that it will be magnitudes slower.”
“What price correctness?”
“Pardon?”
“The current system does not have a formal model and is limited to specific problems.  This new system will make it possible to implement any model, perform an almost unlimited number of operations and make it possible to ask any kind of query without requiring programmers to rebuild the application.”
“I seriously doubt the board is going to fund another system or that the engineers are going to be pleased to know their system is outdated already.”
“If we don’t do it, someone else will.”

ACM workshop - Michigan

“Let me begin by saying that in order to compare the 2 approaches, I had to make sure there was a level playing field.  This required me to attempt to create an abstract model for the opposition.  Frankly, I can’t believe none of you even tried!  If I was a man of short temper, I would say the debate is over, but in the interests of data management I will elaborate…..”

Post ACM Workshop

“He kicked their arses!”
“So many in fact I think he lost a shoe!”

CIA - Langley, Virginia

“We can’t keep up with the data requests Phil.”
“I know! It takes too long to write the data extraction routines.  So many pointers….”
“We need a new system to handle all this quickly; otherwise the “reds under the bed” will take over America.”
“You are such a drama queen Steve!  You are right about a new system though.  We also need a way to be able ask any question we like.  Who knows what the brass will request next.”

CIA - Langley, Virginia – 2 months later

“Gentlemen, the time has come to implement an analysis engine to track the enemies of this great nation.  I understand that a theoretical solution to this problem exists and we must move quickly to build it.  I am giving you an unlimited budget for this.   The code name for this project will be known as ORACLE....

Chapter 2

Chapter 1

Print | posted on Thursday, January 05, 2006 8:09 AM

Feedback

# re: The Birth of JOIN: Chapter 3 (Final)

left by spirit1 at 1/6/2006 3:15 AM Gravatar
Now that is cool :)

# re: The Birth of JOIN: Chapter 3 (Final)

left by rockmoose at 1/6/2006 8:34 AM Gravatar
That is great!
Was Moore's law thought of at the time ;)

# re: The Birth of JOIN: Chapter 3 (Final)

left by DavidM at 1/6/2006 9:02 AM Gravatar
>>Was Moore's law thought of at the time.

Sure was! 1965 for Moore's Law and 1970 for the public release of Codd's Paper.

Everything in the article is accurate (with a bit of poetic licence). And there really was a CIA project named Oracle. Ellison was given the contract and took the name with him....

# re: The Birth of JOIN: Chapter 3 (Final)

left by Jon Hermiz at 1/6/2006 11:49 PM Gravatar
Is this based on true events :-p

# re: The Birth of JOIN: Chapter 3 (Final)

left by rockmoose at 1/7/2006 5:07 AM Gravatar
>> Sure was! 1965 for Moore's Law and 1970 for the public release of Codd's Paper.
I did not doubt you, but wasn't sure. Well a belated google confirmed everything.
Comments have been closed on this topic.