|
PIEBALDconsult wrote: There are still people who argue that there should be no referential integrity on production databases
That tends to be my preference, I prefer to code the referential integrity through the application code rather than embed it in the database.
I won't give the reasons why this can be a very good idea as there will be plenty of counter reasons as to why it is a very bad idea. 
Continuous effort - not strength or intelligence - is the key to unlocking our potential.(Winston Churchill)
|
|
|
|
|
No referential integrity sometime acceptable.
But if you don't have key or index on every table, it will slow down query significantly. (For some complicated join query, it will be 10 to 100 times slower than the one with indexes).
 Can you imagine you have to wait 20 minutes only to print a invoice? (And during this time, all the other people almost can't do anything)
 If you design database like this, I would really hope I'll never cooperate with you.
|
|
|
|
|
Indexes - pwah!
Real men use table scans in their queries - indexes are for cheats  
Continuous effort - not strength or intelligence - is the key to unlocking our potential.(Winston Churchill)
|
|
|
|
|
I have actually created databases using this model and while FKs are great they should not be used to test the integrity of the data. Wait till you see an app where the developer attempts to save a record and traps the FK exception. Talk about programming by error!
Never underestimate the power of human stupidity
RAH
|
|
|
|
|
A good excuse I heard for that nonsense is that " That is good for preformance" !!
RTFM = Read the F***ing manual.
|
|
|
|
|
cartergu wrote: No primary key or index on every table.
"The clue train passed his station without stopping." - John Simmons / outlaw programmer
"Real programmers just throw a bunch of 1s and 0s at the computer to see what sticks" - Pete O'Hanlon
"Not only do you continue to babble nonsense, you can't even correctly remember the nonsense you babbled just minutes ago." - Rob Graham
|
|
|
|
|
I was talking with small db admin turned to MBA describing data-warehouse on those terms
Yusuf
|
|
|
|
|
I just started at a new place and came on to to help with their database and application stuff. I knew it was bad as soon as I saw their databases. For example they have an article table and a column called ArticleTopics which has "23,5,45,78,23,.." they where pulling it into the app normalizing it and getting all topics which of course had TopicCategories "34,56,..." and such. Not a join could be done in the whole database.
I normalized the database using SSIS script component and rewrote the app from coldfusion to ASP.
Finally I submit a request to shut down the old system do one last pull of the data in SSIS and register the new app with the domain and this admin comes to my office screaming that I should never use foreign keys and all normalization should be done in the app because they cause performance problems and that asp is crap, I should use dhtml. I couldn't believe it when my boss told me I need to rewrite it. I lost it.
Anyway anyone need a good DBA/OLAP Designer?
|
|
|
|
|
I've worked at places where they believed that the database should do as little of the work as possible too. 
|
|
|
|
|
PIEBALDconsult wrote: I've worked at places where they believed that the database should do as little of the work as possible too.
Yip, me too, funny the DBA agreed with the issues, but said there was not enough time to do all the 'optimizations'. He had been working a good few years on that DB...
|
|
|
|
|
"If you don't have time to do it right, when will you have time to do it over?"
|
|
|
|
|
After the demo? 
|
|
|
|
|
No, after the demo the customer has a bunch of new features they need.
|
|
|
|
|
You laugh, but this was the policy (and still is) of my last place of employment... 
|
|
|
|
|
Only there? 
|
|
|
|
|
Much sympathy - obviously the admin is useless and feels threatened. Hope you find a new position soon!
'Howard
|
|
|
|
|
What a start! Too bad, that sounds like some crap. When I apply for a job next time I will ask to see samples of their code. Or at least ask them about their coding standards etc.
_____________________________________
Action without thought is not action
Action without emotion is not life
|
|
|
|
|
I would work for free for a week just so I could find out whats the real deal is. It would be worth the vacation time just to find out all the politics.
|
|
|
|
|
I woulda quit on the spot.
|
|
|
|
|
While quitting on the spot does sound appropriate many have obligations to family/creditors that don't allow that luxury. Still, I would move on as soon as reasonably possible.
|
|
|
|
|
I suppose. I have kids and creditors and the possibility would arise very quickly.
|
|
|
|
|
The admin sounds like he has issue, coming in and screaming like that. Geeze.
"The clue train passed his station without stopping." - John Simmons / outlaw programmer
"Real programmers just throw a bunch of 1s and 0s at the computer to see what sticks" - Pete O'Hanlon
"Not only do you continue to babble nonsense, you can't even correctly remember the nonsense you babbled just minutes ago." - Rob Graham
|
|
|
|
|
a friend of mine - many years ago - was asked to port a physics simulation program from one platform to another. The software performed beautifully, was efficient, and had won awards for educational excellence etc.
the source code was several thousand lines of assembly, which was large but not too large for that platform and era.
The program had two comments in it, one word each. They were not helpful.
The variable naming convention was even more interesting. It seems the original author [a physicist/mathematician, in all fairness] treated variable names like a Pez dispenser. The first variable in the program was named "A", the second was named "B", and so on. When he got to Z he started over with A1, then B1, etc.
The program had a lot of variables, I think the last variable was named BB21.
Porting the application was not too difficult, as the two platforms had very similar instruction sets. Debugging it, on the other hand, was an exercise in insanity!
|
|
|
|
|
|
Steven A. Lowe wrote: The first variable in the program was named "A", the second was named "B", and so on. When he got to Z he started over with A1, then B1, etc.
Steven A. Lowe wrote: I think the last variable was named BB21.
Impossible: no physicist/mathematician will conclude such a series with BB21.
If the Lord God Almighty had consulted me before embarking upon the Creation, I would have recommended something simpler.
-- Alfonso the Wise, 13th Century King of Castile.
This is going on my arrogant assumptions. You may have a superb reason why I'm completely wrong.
-- Iain Clarke
[My articles]
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
