----------------------------------------------------------------------------------
@MSGID: 3:633/10 fdad8768
@REPLY: 3:633/10 974c6867
@PID: PyGate 1.5.18
@TID: PyGate/Linux 1.5.18
@NOTE: Thoth/1.9.1 (Mac OS X)
@CHRS: ASCII 1
@TZUTC: 1100
@REPLYADDR: blenko@martingalesystems.com
@REPLYTO: 3:633/10 UUCP
@RFC-Message-ID:
<130720261630311554%blenko@martingalesystems.com>
In article <
slrn115a3i3.ki8.spamtrap42@one.localnet>, Robert Riches
<
spamtrap42@jacob21819.net> wrote:
> An analogous thing happens in SQL database design. Academics
> preach normalization at an absolute goal. However, with a
> sufficiently complex system, full normalization can require each
> query to use an excessive number of joins, and that can adversely
> affect performance. Those with practical experience in DB design
> will concede that sometimes you need to denormalize slightly in
> order to get reasonable performance.
This is nonsense and a good hint is that you are putting words in the
mouths of others and then attacking them for those words.
I`ve never taken a database course but I had nearly 20 years of
database design in industry for mid-level systems. I have read some
database books written by academics along the way. The claim that they
"preach normalization at (sic) an absolute goal" is plain wrong in my
experience.
The academic books I`ve read describe multiple normal forms. I don`t
recall ever seeing one singled as superior to all the others, different
normal forms reflect different (generally theoretical) perspectives on
data and the relational calculus. Not surprising, I think, and I am
confident that being informed about different perspectives has helped
me design systems.
Database design requires a designer to balance various factors,
including where the data comes from, how it is to be maintained, and
how it is to be used (including performance of queries). I have no idea
which of all the tables I`ve designed and run in production may or may
not be normalized by any of the multiple normalizations that have been
described. But I read those books in order to improve my skills and it
was time well spent.
Tom
--- PyGate Linux v1.5.18
* Origin: Dragon`s Lair, PyGate NNTP<>Fido Gate (3:633/10)
SEEN-BY: 19/10 50/109 153/757 218/840 840 220/70
221/1 6 360 226/17 100
SEEN-BY: 229/426 240/1120 267/800 301/1 113 812
310/31 335/364 341/66 463/68
SEEN-BY: 633/10 20 280 281 414 416 418 420 422
509 2744 712/848 770/1 3 100
SEEN-BY: 770/340 350 772/210 220 230 5019/40
5020/715 848 1042 4441 12000
SEEN-BY: 5030/49 722 1081 1474 5053/55 5061/133
5075/128
@PATH: 633/10 280 770/1 218/840 221/6 301/1
5020/1042 4441