Re: Natural keys vs Aritficial Keys
Date: Thu, 21 May 2009 18:51:37 -0300
Michael Schuerig wrote:
>>Michael Schuerig wrote: >> >>>Would you say that these principal functions are essential for, say, >>>blogs, wikis, or social networking sites such as Flickr? >> >>To repeat, no one here says synthetic keys are always wrong. Those >>things you list may well benefit from them.
> Bob Badour just wrote in response to me:
>>If you need to manage data, it's stupid to cripple the primary functions >>of the data management system which is exactly what ORMs and rails do. >>The concept is fundamentally flawed. I am not going to counter basic >>ignorance piecemeal. Go learn the fundamentals and then come back.
> I took him as claiming exactly what you say no one does claim.
> EOD for me, it's going nowhere.
Don't put words in my mouth, idiot. What I said says exactly what I said: no more and no less. ORMs and rails cripple the primary functions of a data management system: the integrity function and the manipulation function for a start. If one has a need for data management, it is incredibly stupid to cripple those functions.
My statement says nothing about choosing keys.
I repeat: If one doesn't have a need for data management, one ought to discuss that somewhere it is relevant and on-topic--not here. If anyone is talking past anyone, it is you. Is there an application that has no need for data management? I neither know nor care. The topic is out of scope in a data management forum.
If you want to get started on a basic education, the design criteria for choosing keys are: uniqueness, irreducibility, simplicity, stability and familiarity. The answer to: "Simplicity is a design criterion" is "Well, duh!" That doesn't address the other 4 criteria nor does it address the frequent conflicts among the criteria. Go away and learn. Come back when you get a clue. Received on Thu May 21 2009 - 23:51:37 CEST