UtterAccess.com
X   Site Message
(Message will auto close in 2 seconds)

Welcome to UtterAccess! Please ( Login   or   Register )

Custom Search
3 Pages V  1 2 3 >  (Go to first unread post)
   Reply to this topicStart new topic
> Composite Indexes Vs Composite Primary Keys, Wiki Talk    
 
   
ButtonMoon
post Oct 11 2013, 05:38 AM
Post#1


Utterly Banned
Posts: 136
Joined: 11-January 12



I can't reply in this thread, it appears to be locked, so I'm posting here.
e: http://www.UtterAccess.com/wiki/index.php?...te_Primary_Keys
The weasel words of the first sentence aren't "debatable" content, they are just wrong content. Composite keys are never a violation of 1NF and except in this one case I haven't heard anyone claim that they are.
Go to the top of the page
 
genoma111
post Oct 11 2013, 05:42 AM
Post#2



Posts: 2,018
Joined: 2-June 09
From: Bogotá - Colombia


OT: I think posts in that thread are moderated meaning you'll have to wait until a moderator approves your comment before it becomes visible to the rest of us.
Go to the top of the page
 
ButtonMoon
post Oct 11 2013, 05:52 AM
Post#3


Utterly Banned
Posts: 136
Joined: 11-January 12



I can't even comment. When I click reply I get "Sorry, this topic has been locked".
Go to the top of the page
 
argeedblu
post Oct 11 2013, 06:48 AM
Post#4


UA Forum + Wiki Administrator
Posts: 14,057
Joined: 26-September 02
From: Sudbury, Ontario, Canada


Someone will be taking a look at what's going wrong. Thanks,
Glenn
Go to the top of the page
 
CyberCow
post Oct 11 2013, 08:53 AM
Post#5


UdderAccess Admin + UA Ruler
Posts: 19,555
Joined: 27-April 02
From: Upper MI


We're looking into it and have verified the problem. Thank you for pointing it out. We hope to have it fixed by tomorrow. (many of us admins have regular jobs to attend and those of us who have the engine room responsibilities are now aware of the issue and have it at the top of our "fix list".)
or one of the other Engine Room attendees will post back when the issue is dealt with. hat_tip.gif
Go to the top of the page
 
datAdrenaline
post Oct 12 2013, 11:30 AM
Post#6


UtterAccess Editor
Posts: 17,941
Joined: 4-December 03
From: Northern Virginia, USA


"The weasel words of the first sentence aren't "debatable" content, they are just wrong content."
Agreed. I reworded briefly. But the whole article needs help --- I can't give it at the moment
Ideas:
- Explain what a "key" is. Also the claim of a "Composite Index" is misleading ... its an index or not, the key of the index is the composite part. With that its a Composite Key Index, or an Index with a Composite Key.
- Then go into the differences between how a Composite Key Primary Index and a Single Field Key Primary Index can effect your development processes.
Go to the top of the page
 
dmhzx
post Oct 12 2013, 12:30 PM
Post#7



Posts: 7,033
Joined: 22-December 10
From: England


When it is addressed, I wonder if the following could be included please.
Is there a practical difference between an 'index' and a 'key'.
There may only be one Primary Key, but you can have a load of unique indexes. So why is a unique index called Primary Key at all important: - Apart from the convention of the terms PK and FK, and entity modelling, and schema design.
When building 'indexes' in Access, you can designate one of the to be the 'Primary Key': So does it then stop being an 'index', or is it just a special case of 'index'
When working with VBA, you can order by any 'index', including the one called "Primary Key".
So is it that you may have a maximum of one 'key' per table. That one key IS the primary key, but you can call it whatever you want?
And when explaining this, could you point out exactly how Access helps us to understand.
I'm asking because I can't see much other than semantics separating a primary key from any other unique index, and I feel I MUST be missing something..
Go to the top of the page
 
ace
post Oct 12 2013, 01:08 PM
Post#8


UtterAccess VIP
Posts: 5,446
Joined: 26-November 05
From: Upstate NY, USA


While we're at it, a wiki article really should be an actual article as opposed to a few sentences introducing
a couple of files to download.
Go to the top of the page
 
genoma111
post Oct 12 2013, 02:56 PM
Post#9



Posts: 2,018
Joined: 2-June 09
From: Bogotá - Colombia


That's not even an abstract evilgrin.gif
Go to the top of the page
 
CyberCow
post Oct 12 2013, 04:16 PM
Post#10


UdderAccess Admin + UA Ruler
Posts: 19,555
Joined: 27-April 02
From: Upper MI


The beauty of our Wiki is that ANY member can add/edit the content. These "Talk" threads are merely for further/deeper discussion about the actual topic before/after such content is added/edited. hat_tip.gif
Go to the top of the page
 
ButtonMoon
post Oct 12 2013, 05:53 PM
Post#11


Utterly Banned
Posts: 136
Joined: 11-January 12



In relational database terms a table must have at least one key and it may have more than one. A key is a minimal superkey, which means a minimal set of attributes (zero, one or more attributes) which collectively are required to be non-null and unique at all times. The word key is also used for key constraints, i.e. the implementation of a logical rule in a database that a certain set of attributes must be unique. For example the NOT NULL/UNIQUE and PRIMARY KEY keywords in SQL are syntaxes that are used to define key constraints. By convention when a table has more than one key then one key of that table is deemed to be a "primary" one, meaning it is the "preferred" identifier in that table or has some other special significance. The other non-primary keys of the table are then called secondary keys or alternate keys. The distinction of a primary key is only a convention however and is only as important as you want it to be. A primary key doesn't have to be different in form or function from any other key unless you choose to make it so. The relational model itself makes no theoretical or practical distinction between primary and alternate keys. All keys are equal under the relational model.
While keys are logical features of a database, indexes are physical. An index is a performance and storage optimization feature. Key attributes may or may not be indexed. Usually they are indexed because keys are usually very good candidates for performance optimization. Some DBMSs actually require keys to be indexed and even provide a "unique index" feature as an alternative way of defining keys. Allowing an index feature to define a key is an unfortunate compromise of the principle of Physical Database Independence but it can also be seen as mere syntactical sugar: the fact that a key is defined as a "side effect" of index creation doesn't alter the fact that it is a key.
Access won't help you understand. There is no accounting for the sloppy and confusing way in which keys, constraints, indexes and other database features are presented in the Access UI. Any student wanting to learn about database principles would be well advised to stay away from Access.
Go to the top of the page
 
genoma111
post Oct 12 2013, 06:43 PM
Post#12



Posts: 2,018
Joined: 2-June 09
From: Bogotá - Colombia


QUOTE (ButtonMoon)
Any student wanting to learn about database principles would be well advised to stay away from Access.

That's like saying that anyone who wants to learn music should stay away from jazz...
Go to the top of the page
 
CyberCow
post Oct 13 2013, 08:58 AM
Post#13


UdderAccess Admin + UA Ruler
Posts: 19,555
Joined: 27-April 02
From: Upper MI


ButtonMoon - I agree with Diego. Making such a statement on a forum that primarily supports Access is counter-productive.
I knew nothing of database principles until I encountered Access. Before Access, I worked with DB4 and Clarion - both were horrible ways to learn database concepts. (IMO)
Some sort developing environment is required to learn db principles and IMO Access is the fastest path to that end.
Access helps us to understand db principles by providing enough controls to shoot one-self in the foot or render a beautifully designed data model and everything in between. LEARNING database principles is a process - no matter what environment is selected/used to engage in that learning. While Access is not the only environment to learn db principles, I challenge anyone to find a faster tool.
Any member wishing to participate here at UtterAccess would be well advised to support the very thing for which this forum was designed, is maintained and exists. :ha
Go to the top of the page
 
ace
post Oct 13 2013, 10:18 AM
Post#14


UtterAccess VIP
Posts: 5,446
Joined: 26-November 05
From: Upstate NY, USA


One could sit down in front of Access with a good book on relational database design and learn about relations, tuples, keys, constraints and
what have you by using nothing but the SQL view of the query builder.
laming a tool for someone not taking the time to learn the basics of relational database design is really kind of silly.
Go to the top of the page
 
argeedblu
post Oct 13 2013, 10:23 AM
Post#15


UA Forum + Wiki Administrator
Posts: 14,057
Joined: 26-September 02
From: Sudbury, Ontario, Canada


Well said, Mark and Ace.
That's the old saying, "A good workman doesn't blame his tools."
Glenn
Go to the top of the page
 
ButtonMoon
post Oct 13 2013, 10:34 AM
Post#16


Utterly Banned
Posts: 136
Joined: 11-January 12



That's true but then why bother with Access at all if you aren't using its UI.
I'm not blaming the tool for people not trying to learn, I'm simply saying that Access's irrelevant and misleading terminology and UI features tend to hinder rather than contribute to the learning process. I think there's plenty of evidence in these forums and elsewhere to confirm that impression. That isn't a criticism of Access for what it was intended for: as a development tool it isn't a problem. But there are far better ways to learn about databases.
Go to the top of the page
 
ace
post Oct 13 2013, 10:36 AM
Post#17


UtterAccess VIP
Posts: 5,446
Joined: 26-November 05
From: Upstate NY, USA


Can you make any suggestions?
Go to the top of the page
 
CyberCow
post Oct 13 2013, 11:06 AM
Post#18


UdderAccess Admin + UA Ruler
Posts: 19,555
Joined: 27-April 02
From: Upper MI


I believe the venue and end-target of what a student is intending to learn is key to the starting point. Access was never intended for totally secure, mission critical db development tool (IMO); I've always seen it as the best and fastest pilot-application development to ever hit the market.
Oalso believe Microsoft is now targeting Access to provide better web-based tools for those who have little to zero programming experience/education.
And as Ace suggests, please point out other suggestions.
Go to the top of the page
 
GroverParkGeorge
post Oct 13 2013, 12:10 PM
Post#19


UA Admin
Posts: 31,240
Joined: 20-June 02
From: Newcastle, WA


When I was studying linguistics back in the 70's, we identified a specific type of writing primarily associated with academics. Hence, we referred to it as the academic style, although that term can include a lot of other things in addition to the one that is relevant here. What we meant by "academic writing" is that many journal articles are written NOT to illuminate or elucidate a particular topic, but rather to establish the writers' superior intellect and understanding. The way that is done is to employ highly technical language, complex theory and an obscure style to convey the notion that, "I know more than you; therefore it's your fault if you can't understand what I'm saying." The genius of some well known science writers like Carl Sagan and Richard Feynman is, IMO, that they realized "academic pontification" was not going to help anyone understand anything. I think Stephen Hawking's books show that he shares that same understanding. There are others. Unfortunately, there are also far too many academics who try to establish their superiority through journal articles whose hallmark is obscurity rather than elucidation.
The point I'm trying to make is that we offer the most help to our members by avoiding the urge to show off our superior knowledge and just help them try to understand the nature of the problem they are facing, along with some practical suggestions about solving it. Again, IMO, the typical member here at UA is not at all interested in mastering relational design, no more than Sagan's legion of readers intended to master biology, nuclear physics or astronomy. When our members pick up some sound principles along the way, so much the better.
Go to the top of the page
 
CyberCow
post Oct 13 2013, 01:27 PM
Post#20


UdderAccess Admin + UA Ruler
Posts: 19,555
Joined: 27-April 02
From: Upper MI


George - very well said! order="0" alt="thumbup.gif" />
Go to the top of the page
 
3 Pages V  1 2 3 >


Custom Search
RSSSearch   Top   Lo-Fi    17th December 2017 - 10:28 AM